ABSTRACT
PART I
DIRECT OXIDATION OF ALKENES TO KETONES
USING PEROXYTRIFLUOROACETIC ACID—BORON FLUORIDE
PART II
THE ACID-CATALYZED REARRANGEMENT OF
OCTAMETHYL-8-OXATETRACYCLO[4.3.0.02'5.07I9]NON-3-ENE

by Lawrence Robert Lerner

The purpose of the first part of this thesis was to
determine the synthetic utility of peroxytrifluoroacetic
acid-boron fluoride etherate as an oxidant for effecting
a one-step transformation of alkenes to ketones. In a
number of cases this transformation was carried out suc-
cessfully (1). The products can be explained as the result
of attack of positive hydroxyl on the double bond, followed
by Wagner-Meerwein rearrangement, to yield the correspond—
ing ketone. The products are shown to be the result of

hydrogen, methyl, acyl, chlorine or bromine migration, as

R R R
\\\c‘// , ' R — é — R R - é - R
0H*' ~R
9-” KART 4.,
R/// \\\R 5a R

well as ring contraction and expansion. The following trans-

formations exemplify these migrations:

76

Lawrence R. Lerner

H
+ CH3 _ C - CH3
OH > I 53%
rvH C = 0
/
CH3
252.
OH+ $H3
~CH3> CH3 - c]: - eta 7796
C = O
/
CH3
4,33.
Cl
CH3 " C " CH3
OH+ g ‘ 0
~ c1 > / 77%
CH3
£4.
Br
OH+ CH3 - (I: " Br
.__————> 69%
“'Br C = O
/
CH3
95

 

(ring expansion)

+
OH '
Acyl migration? . 43%

Lawrence R. Lerner

 

CH3 + CH3
. OH . > 76%
ring contraction
CH3 l/C' —CH3
36 38 O

rw rw

The second part of this thesis deals with the structural
determination of a product isolated from the reaction of 22'
with peroxytrifluoroacetic acid. The synthesis of this
compound, by the acid-catalyzed rearrangement of 122/ has
been reported by Maier (2), but the previously assigned
structure, 122/ is shown to be incorrect. The actual struc-
ture, 122” is determined by the nmr and ir spectra of 192,
and the spectra of its hydrogenation and hydroxylation

products (106 and 112, respectively).

 

 

 

 

 

 

 

CH3 0 CH3
CH CH3
\
/
CH CH ‘
CH3 3 3 CH3
, 100
CH3
CH3
CH CH3
H2804 V \ O
acetone ' / l
CH3 CH3
CH3
H3
103

 

 

 

 

Lawrence R. Lerner

CH CH3 0 CH3
3
\7/ CH3
H2 7\ H ‘ /
Pd/C H )
CH CH H3
CH3 3 3
C CH
CH3 H30 3
CH3
1' x H t m
_ fl
2. OH .

In addition, it is shown that 102 thermally rearranges

to 104 as was reported by Maier in the same article, but

some of the previously reported details of the nmr spectrum

of 104 are shown to be incorrect.

180

 

102

 

 

 

 

CH3
CH3 CH
CH3 / 3
/
o
C \\‘ \
H3 CH
CH3 3
CH3
103
REFERENCES

1. H. Hart and L. Lerner, J. Org. Chem. 3g, 2669 (1967).

G. Maier, Chem. Ber. 26, 2238 (1963).

PART I
DIRECT OXIDATION OF ALKENES TO KETONES

USING PEROXYTRIFLUOROACETIC ACID-BORON FLUORIDE

PART II
THE ACID-CATALYZED REARRANGEMENT OF

OCTAMETHYL-B-OXATETRACYCLO[4.3.0.02v5.0719]NON-3-ENE

BY

Lawrence Robert Lerner

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1968

To Lenore

ii

ACKNOWLEDGMENT

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

Appreciation is extended to the National Science Founda—
tion for financial support from June, 1965 to August, 1965
and from October, 1966 to June, 1968.

Appreciation is also extended to the Army Research
Office (Durham) for financial support from March, 1966 to
September, 1966.

Finally, the author would like to thank his family for

their support and encouragement.

iii

TABLE OF CONTENTS

PART I

DIRECT OXIDATION OF ALKENES TO KETONES USING
PEROXYTRIFLUOROACETIC ACID-BORON FLUORIDE

INTRODUCTION

RESULTS AND DISCUSSION

of 2,3—Dimethyl-2-butene (16) . . .
of 2-Methyl-2-butene (g1) . . . .
of cis and trans-3-Methyl-2-pentene

(g6 and g1). . . . . . . . . . . . . . . . .

of l-MethyICyclohexene (22) . . .
of 1,2-Dimethylcyclohexene (26) .
of Agtlo-Octalin (21) L . . . . . .
of 2-Methyl—3-phenyl—2-butene (22).

of 3,4-Dimethyl-2,5-dihydrothiophene—

1,1-dioxide(§g)..............

of Isopropylidene Malononitrile (21)
of 3—Ethyl-2-methyl-2-pentene-

nitrile (Q2) . . . . . . . . . . . . . . .

A. Oxidation
B. Oxidation
C. Oxidation
D. Oxidation
E. Oxidation
F. Oxidation
G. Oxidation
H. Oxidation
I. Oxidation
J. Oxidation
K. Oxidation
L. Oxidation
M. Oxidation

(2.9)..
N. Oxidation
O. Oxidation
P. Oxidation
Q.

of Z-Methyl-Z-butenal (£9) . . . .
of 3,4-Dimethyl-3-penten-2-one (fig)

of 2—Cyc10pentylidenecyclopentanone

of 3-Chloro—2-methyl-2-butene (gg).
of 3-Bromo-2—methyl-2—butene (21) .
of 2,3-Dibromo-2-butene (2%) . .

Generalizations about the Oxidation of Alkenes
to Ketones Using Peroxytrifluoroacetic Acid
- Boron Fluoride . . . . . . . . . ... . . .

1.

~Migratory Preference . . . . . . . . . .

2. The Effect of Substituents on the
Double Bond . . . . . . . . . . . .

iv

Page

20
20
23

25
26
28
28

29
29

33

TABLE OF CONTENTS (Cont.)

EXPERIMENTAL . . . . . . . . . . . . . . . . . . . .
A. General Procedures . . . . . . . . . . . . .
1. Melting Points .
2. Microanalyses . . . . . . . . . . .
3. Infrared Spectra . . . . . . . . . . .
4. Mass Spectra . . . . . . . . . . . . .
5. Nuclear Magnetic Resonance Spectra . .
6. Gas Chromatography . . . . . . .
B. General Oxidation Procedures . . . . .
C. Preparation of Alkenes .. . . . . . . . .
1. A9110-Octalin (41) . . . . . . . . . . .
2. 1,2-Dimethylcyclohexene (36) . . . . .
3. 2-Methyl-3-phenyl—2-butene (45) . . . .
4. 3,4-Dimethyl-3-penten-2-one (64) . .
5. 4—Methyl-g3-3-methyl-3-penten-2-one—
1,1,1,5,5,5-g5 (11) . . . . . . . . . .
6. 2-Cyclopentylidenecyclopentanone (22)
7. 3-Chloro-2-methyl-2—butene (g2) . . . .
8. 2—Chloro—3-methyl-2-butene-1,1,1—g3 (g6)
D. Results of Alkene Oxidations . . . . . . . .

1.

coooslcnonmwm

H P‘
HO

2,3-Dimethyl-2-butene (11) . . . . . .
2-Methyl-2-butene (21) . . . . . . . .
gisy3-Methyl-2—pentene (26) . . . . .
trans-B—Methyl-Z-pentene (21) . . . .
l-Methylcyclohexene (32) . . . . . . .
1,2-Dimethylcyclohexene (32) . . . . .
A9'10-Octalin (41) . . . . . . . . . .
2-Methyl-3-phenyl-2-butene (45) . . .

3,4-Dimethyl-2,5-dihydrothiophene-1,1-
dioxide (52) . . . . . . . . . . . .

Isopropylidene Malononitrile (52) .
3-Ethyl—2-methyl-2-pentenenitrile (52)

V

Page
37

37
37
37
37
37
37
37

37

39
39
4O
41
43

44
44
45
46

46
46
47
47
47
48
49
49
50

50
51
51

TABLE OF CONTENTS (Cont.)

SUMMARY

OCTAMETHYL-S-OXATETRACYCLO[4.3.0.02I5.07'9]NON—3-ENE

INTRODUCTION

12.
13.
14.

15.
16.
17.
18.
19.

2-Methyl-2-butenal (59) . . . . . . . .
3,4-Dimethyl-3—penten-2-one (64) . . .

4-Methylfig3~3-methyl-3-penten-2-one—
1,1,1,5,5,5-g__6 (1%) . . . . . . . .
2-Cyclopentylidenecyclopentanone (Zn)
3-Chloro-2-methyl-2-butene (82) . . .
2-Chloro-3-methyl-2-butene-1,1,1-g3 (Q6)
3-Bromo-2—methyl—2-butene (91) .

2,3-Dibromo-2-butene (93) . . . .

PART II

THE ACID CATALYZED REARRANGEMENT OF

RESULTS AND DISCUSSION 0 O O O O O O O O O O O O O O O

A.
1.
2.
3.
4.
5.
6.
B.
EXPERIMENTAL
A.

The Acid-catalyzed Rearrangement of Octamethyl-

Page
52

54
54
55
56
56
56

58

6O
62

8—oxatetracyclo[4.3.0.02'5.07r9]non-3-ene (102)62

Thermal Rearrangement of Octamethyl-B-oxatetra-

Nuclear Magnetic and Infrared Spectra '.
Hydrogenation . . . . . . . . . .

Bromination . . . . . . . . . . . . . .
Hydroxylation . . . . . . . . . . . . .
Attempted Photolysis . . . . . . . . . .

meChanj—sm O O O O O O O O O O O O O O O

cyclo[4.3.0.02'5.O7I9]non-3-ene (102) . . .

Preparation of Octamethyl-s n-tricyclo-
[4.2.0.02'5]octa-3,7-diene 92) . . . . . .

vi

62
65
67
68
7O
71

74

76

TABLE OF CONTENTS (Cont.)

B. Preparation of 1,2, 3
8-oxatetracyclo[4. 3.
(102) . . . . . . . . .

C. Preparation of 1
oxatricyclo[4.3.

1

2.

I

2
O.
D. Preparation of ,3,
oxatricyclo[5. 0.

E. Reactions of 1, 3,4
oxatricyclo[5. 2. 0.

I

001 Ch 00.!
:31!

1. Hydrogenation

2 Hydroxylation . . . .

3. Attempted Photolysis . . . .
4

Bromination . . . . . . . .
SUMMARY . . . . . . . . . . . . . . . . .
SPECTRA . . . . . . . . . . . .

Nuclear Magnetic Resonance Spectra

Infrared Spectra

LITERATURE CITED

vii

7,9- Octamethyl-
.07 9]non-3—ene

,6,7,9-Octamethyl-8-
9nona-2,4—diene (104).

,7,8,9-Octamethyl-2-
nona-4, 8-diene (100).

,8,9- -Octamethyl- -2-
nona-4, 8 -diene (100).

Page

77

78

78

79
79
80
81
82

83
84

85
99

108

LIST OF FIGURES

FIGURE page
1. Nmr spectrum of 1,2-dimethylcyclohexene (88). . 86
2. Nmr spectrum of a 63:37 mixture of 2-methyl-3-
phenyl-2-butene (48) and 3—methyl-2-phenyl-1-
butene (42) . . . . . . . . . ',' . . . . . . 87
3. Nmr spectrum of an 80:20 mixture of 48 and 42'. 88
4. Nmr Spectrum of 3,4-dimethyl-3—penten—2—one (82) 89

5. Nmr spectrum of deuterated 3,4-dimethyl-3-
penten—2-one (1;) . . . . . . . . . . . . . . 89

6. Nmr spectrum of 3-chloro-2-methyl-2-butene (88) 90

7. Nmr spectrum of deuterated 3—chloro-2-methyl-
2-butene (82) . . . . . . . . . . . . . . . . 90

8. Nmr spectrum of 1—methylcyclohexYLtmifluoroe
acetate (8%) . . . . . . . . . . . . . . . . . 91

9. Nmr spectrum of 3,3—dimethyl-2,4-pentanedione
(88) . . . . . . . . . . . . . . . . . . . . . 92

10. Nmr spectrum of deuterated 3,3—dimethyl-2,4-
pentanedione (11). . . . . . . . . . . . . . . 92

11. Nmr spectrum of spiro[4.5]decane-6,10-dione (Z8) 93
12. Nmr spectrum of 3-chloro-3-methyl-2-butanone (82) 94

13. Nmr spectrum of deuterated 3-chloro-3-methyl-
2-butanone (82) . . . . . . . . . . . . . . . 94

14. Nmr spectrum of a 43:57 mixture of octamethyl-S-
oxatetracyclo[4.3.0.02I5.07I9]non—3—ene (£28)
and octamethyl-B—oxatricyclo[4.3.0.07:9jnona-
2,4-diene(123) 95

15. Nmr spectrum of octamethy1-2-oxatricyclo-
[5.2.0.03:6]nona-4,8-diene (100) . . . . . . . 96

16. Nmr spectrum of the hydrogenation product (106)
Of 100 . . . . . . . . . . . . . . . . . ... . 97

viii

LIST OF FIGURES (Cont.)
FIGURE page

17. Nmr s ectrum of the hydroxylation product
(112 of 100 . . . . . . . . . . . . . . . . . 98

18. Infrared spectrum (CC14) of A9v10-octalin (41). 100

19. Infrared spectrum (CC14) of 1,2—dimethylcyclo—
hexene (88) . . . . . . . . . . . . . . . . . 101

20. Infrared spectrum (liquid film of 2-cyclo- .
pentylidenecyclopentanone (28' . . . . . . . . 102

21. Infrared spectrum CC14) of 1-methylcyclohexyl
trifluoroacetate 81) . . . . . . . . . . . . 103

22. Infrared spectrum (CCl4) of spiro[4.5]decane-
6,10-dione (Zé) . . . . . . . . . . . . . . . 104

23. Infrared Spectrum (CCl4) of octamethyl-Z-oxa-
tricyclo[5.2.0.03I5]nona-4,8-diene (100) . . . 105

24. Infrared spectrum (CCl4) of the hydrogenation
product (106) of 100 . . . . . . . . . . . . . 106

25. Infrared spectrum (CCl4) of the hydroxylation
product (112) of 100 . . . . . . . . . . . . . 107

ix

PART I

DIRECT OXIDATION OF ALKENES TO KETONES

USING PEROXYTRIFLUOROACETIC ACID-BORON FLUORIDE

INTRODUCTION

Peracids have long been used to oxidize alkenes to
epoxides (1). The reaction is generally believed to in-

volve electrophilic attack of the peracid on the double

 

 

 

bond.
R /R R /R R /R
>c = c\ > >c\__ c\ > >c\-—-—- c\
R‘ R R \x R R 0/ R
l /O\\
H’ o
H-O\ : :J 0
(VO 0 C\R H - o - c - R
c
‘6—é \R

The electrophilic nature of the reaction is borne out
by the fact that substituents on the double bond which are
electron-donating (e.g. alkyl) increase the reaction rate,
whereas those which are electron-withdrawing (e.g. carboxyl)
decrease the rate (2). In addition, the stronger the per-
acid, the faster the rate. Peroxytrifluoroacetic acid is
particularly reactive (3). It has a much weaker oxygen-
oxygen bond than other peracids and will react readily with
negatively substituted double bonds (e.g. ethyl crotonate),
whereas other peracids will react only very slowly at best.

Acid-catalyzed ring opening of epoxides, which leads to
carbonyl compounds yi§_Wagner-Meerwein rearrangements, is

also well known (4).

 

 

R O R OH O R
\/\/ HB I + ~R u n
C“""""C >R-C-C-R > C-C-R
/ \ I I R/ I
R R R R R

This type of rearrangement is effectively catalyzed
by Lewis acids such as boron fluoride. The direction of
ring opening is normally such that the more stable carbonium
ion is formed. Subsequently or simultaneously, one of the R
groups migrates. Migratory aptitudes are generally in the
order aryl > acyl > H > t-butyl > ethyl > methyl. A few

examples follow:

¢ 0 H -OBF3 H o
\c/ \c/ BF EtO «pg é H H «.2: c//
// \\ ___::I.__._.2...> . . ...) ' \

¢ H ¢ H ¢ H

2, a a

Cope, Trumbull and Trumbull (5) (1958).

 

 

H o H -OBF3
\ / \ / + l
C C BFg-Etzo H'C - C "H 1.9;— 3
/ \ > - - > ~
¢ ¢ ' ¢ ¢
2. i
Reif and House (6) (1963).

H O H + -OMgBr H O
\ / \ / MgBrz-EtZO ' ~ H ' "
/C _ C -—> H-‘C ""'—" C-H > H-C -' C‘C5H11

I I I

OCH3 CH2CH2CH2CH2CH3 OCH3 C5H11 OCH3

203

1 8

~

Stevens and Dykstra (7) (1954).

4
In some cases ring contraction or expansion may occur

as in the following two examples:
CH3 CH3
CHO

MgBI 1—Et20 >

OCOCHa OCOCH3
2. l9.
Braude gt. _1 (8) (1958).

O
EH4) BFR -E th >
acyl migration

11 12

('W rw

House and Wasson (9) (1956).

The ring opening and migration may or may not be con-
certed, though there is evidence for a concerted mechanism
in certain cases (10).

Closely related to the acid—catalyzed epoxide ring

opening, is the acid catalyzed pinacol rearrangement (11).

 

OH OH OH R O

I I HB + I N R I II
R-C — C-R > R-C - C-R :fi-‘F—> R-C - C-R

I I I I I

R R R R R

In the pinacol rearrangement, as in epoxide ring open-
ings, the more stable carbonium ion is usually formed: this

is followed by migration of a group and proton loss. The

5

order of migratory aptitudes, in general, is the same as
that for the epoxide ring opening reaction.

Previous work in this laboratory has demonstrated the
usefulness of peroxytrifluoroacetic acid - boron fluoride
as an electrophilic oxidant of substituted aromatic compounds
(12). With hexaalkylbenzenes, hexaalkylcyclohexadienones
were obtained in good yield by electrophilic attack by the
oxidant (where boron fluoride aids heterolytic cleavage
of the peracid), followed by a Wagner-Meerwein rearrangement
and proton loss. For example, hexamethylbenzene (l8) was

converted into hexamethyl-Z,4-cyclohexadienone (18) in 88%

yield (13).

CH3
CH CH3 $F3
i
x-yo/H OD
VS£"_ . CH CH3
0 C CF3 , 3 CH
CH3 H3 3
CH CH3 CH
CH: CH
~CH 3
OH I 9
or 13 -H O
'”” CH3 CH3
H3 H CH3
3 H3
CH3 H3H 14 15
/
O 9 MV rw
\gP-C-CFs
CH3 H3 BF
3
CH3
13

The type of carbonium ion intermediate (£4) postulated

here could be envisioned in a similar oxidation of an isolated

6
double bond, in which case the intermediate ion would be
the same as that suggested in the pinacol rearrangement and
should lead to formation of a ketone yi§_migration and

proton loss.

\. /’ H F3
C/BI / c3
I 0 q n
/C\ \o - C - CF3 R R
R R . .
R — C+ R - c - R
or > I “'R > I
R — C — OH _H* C = o
' ./
R R R R
\ -/ H
/ o
fiflo\ ..
C Q‘ "' C - CFa
./
R \R BF3

A ketone would also result if the initial reaction was
epoxidation of the double bond by the peracid, followed by

boron fluoride catalyzed rearrangement of the epoxide.

 

 

R R R R R
\ / \ / I

lCl CFICOQH > C\ BF: > R ‘ (f " R

C O ~R c-o

/ \ / _
R R C R/
/’ \. '
R R

An initial experiment (14) carried out on the oxidation
of 2,3-dimethyl-2-butene (18) with peroxytrifluoroacetic
acid - boronfluoride resulted in a 75% yield of the antici—

pated product, pinacolone (11).

 

CH3\\ CH3 CH3
fill CF3CO3H > CH3 - ? - CH3
/C\ BF3 /C=O
CH3 CH3 CH3
kg 17

Without boron fluoride, pinacolone was produced in

only 16% yield, and the major product was the hydroxytri-

fluoroacetate, 12;

 

CH3 CH3 CH3
\. // '
C CH3 - C — OH
16 CF3C03H ‘ '\\O CF3C02H I
,CC 7? // > CH3 - C - OCOCF3
C I
/ \ CH3
CH3 CH3
18 19

m

This hydroxytrifluoroacetate was quantitatively converted
to pinacolone upon the addition of boron fluoride, pos-

sibly via a pinacol rearrangement type intermediate (20)

as shown.

 

CH3 CH3

I I
CH3 ' C - OH CH3 " C - OH CH

n A, > I —-e> 1.2.
CH3 - C — o - C - CF3 CH3 - C + - H

I u I

CH3 (p: CH3

‘BF3
12 2.9.

This would be a third possible mechanistic pathway to the
ketone, when the oxidation is carried out in the presence of

boron fluoride.

8

Part I of this thesis describes the results of a study
on the SCOpe of the oxidation of alkenes to ketones in one
step (15) and a comparison of migratory aptitudes, reflected
in the products obtained, with those found in the acid-
Catalyzed rearrangement of the corresponding epoxide or

glycol.

RESULTS AND DISCUSSION

A. Oxidation of 2,3-Dimethyl-2-butene (lg)

 

It was found to be most convenient to convert 2,3-di-
methyl-Z-butene (lg) to pinacolone (11) by substituting
boron fluoride etherate for gaseous boron fluoride which
had been used previously (14). Under these conditions the»
yield of l1,was 72%, as opposed to 75% with gaseous boron
fluoride. (For simplicity the attacking oxidant will be
represented as OH+, though any of the mechanisms discussed

in the introduction may be operative.)

 

CH3 CH 3 CH3 CH3
\\C// + CH C+ CH C CH
3 - ~ 3 - - 3
H OH > | ICIH 3_>
C CH ‘ C - OH — C = O
/ \ 3 I /
CH3 CH3 CH 3 CH3
16 20 17

It was also found that if the reaction mixture was not
cooled (as was originally done) the yield of lz'increased
slightly to 75%. Therefore, by simply adding the peracid
in methylene chloride concurrent with the addition of the
liquid boron fluoride etherate, at such a rate that the
reaction mixture refluxed gently, the entire experiment could
be carried out within two hours. Since the total synthesis
could be carried out without isolating any intermediate
(e.g. epoxide or glycol) and within this short period of
time (the peracid was completely consumed within 15 minutes

9

10
after mixing the reagents), this procedure promised to be
an excellent method for effecting the transformation with
other alkenes.

All the oxidations to be discussed were therefore car-
ried out using boron fluoride etherate, but some oxidations
were carried out at either 0-80, at reflux (40°) or at both
temperatures, since attempted oxidation without cooling was

not tried until this study was well underway.

B. Oxidation of 2-Methyl-2-butene (gl)

 

When 2-methyl-2-butene (21) was oxidized at 0°, the
exclusive, volatile product was 3-methyl-2-butanone (Ea)
as would be expected by initial attack of the electrOphile
to form the most stable ion, 22, followed by hydrogen migra-

tion and loss of a proton.

 

CH3 CH3 CH3 CH3
\ / I 1
C + CH3 - C+ CH3 - C - H
H OH > | «’H > |
C CH3 - C — OH _H‘F C = o
‘// ‘\ ' /’
CH3 H H CH3
2,1, 2,2. 22,

This product (gg) was also the major one (> 95%) in

the pinacol rearrangement (16) of the corresponding glycol

\

22, in aqueous perchloric acid. It was also demonstrated
by the same workers, through deuterium labeling, that the
hydrogen migration was intramolecular; therefore the reac-

tion did not proceed via the enol (gé) of 22;

11

 

 

 

CH3 CH3 H(D)
I I I
CH3 " C " OH + CH3 " C+ CH3 " C " CH3
H ~IH D
I > I ji¥L> I
CH3 - C - OH CH3 - C - OH /’C = o
I I
H(D) H(D) CH3
2.2, 22. 232.
A
+
—H (D+) CH3 CH3
\ /
w a, > ll
./ ‘\
CH3 OH
25

C. Oxidation of cis and trans—S-Methyl-Z-pentene (gg'and 21)

 

The oxidation of both gig (gg) and trans (gl)-3-methyl-
2-pentene, at 0°, resulted in the formation of 3—methyl-2-
pentanone (22) in 63 and 70% yield reSpectively.

This product, as in the oxidation of g}, is the one
expected on the basis of formation of the most stable ion,

§§x followed by hydrogen migration and loss of a proton.

 

 

Et\ /CH3
TI
CH/C\H 9% g
3 Et - C+ Et — C - CH3
gfi OH+ I ~'H I
> CH3 — C - OH -——;—> C = 0
Et\\ l/CH3 .. -H /
C H CH3
'3 * 2,82, 22.
H/ \CH3

12

D. Oxidation of 1-Methylcyclohexene (g2)

 

A serious side reaction was found in the oxidation of
l-methylcyclohexene (Q2). When the oxidation was carried
out at 0°, two volatile products were detected. They were
identified as 2-methylcyclohexanone (gg) and l-methylcyclo-
hexyl trifluoroacetate (2}) obtained in 8 and 15% yield,
respectively. When the oxidation was carried out at reflux,
the yield of §§ was improved to 41% whereas the yield of
g; dropped to 9%. Compound g} was undoubtedly formed by
the addition of trifluoroacetic acid to Q2; Q; was the only

product when the peracid was omitted.

9

CH3 CH3

 

OCOCF3
CF3C02H
>
BF3 -Et20
CH2C12
2.9, 9.1,

The expected oxidation intermediate 22 might give
rise to two products, §§ by hydrogen migration or aldehyde

gg by ring contraction, but only g2 was observed.

CH3
0
CH3
' OH
gg 0H+ H -H+ 33
0
CH3 C-H
34

13
The two possible carbonium ion intermediates are 32a
and 32b. Either could lead to ring contraction but only

32a (axial hydrogen) leads to hydrogen migration.

CH3 CH3

OH
32a £2.12,

Ion gga with the equatorial hydroxyl, should be the more
stable of the two. From these results 222 is the preferred
intermediate and rearranges exclusively by hydrogen migra-
tion. This migratory preference might be expected since
hydrogen migration is preferred over alkyl migration (ring
contraction) in Wagner-Meerwein rearrangements (4).

Similar results have been reported (17) in the epoxid-
ation of g2 with perbenzoic acid. Here the expected
epoxide gé'was found in 48% yield and the only rearrangement
product, formed in 24% yield, was §§3 The authors suggested
that §§ might have been formed from £2“ which, in turn, was
formed by direct attack of positive hydroxyl on £2; This,
of course, is one of the mechanisms postulated in this
thesis for the oxidation with peroxytrifluoroacetic acid —

boron fluoride.

14

OH ~I:I

\L/

:91

\V

32
30 A. m

CH3

¢CO3H

 

v,

 

35

E. Oxidation of 1,2—Dimethylcyclohexene ( 6)

m

 

In contrast to the results on the oxidation of g2, the
oxidation of 1,2-dimethylcyclohexene (fig) at 00 led ex-
clusively (76% yield) to the ring contraction product, 1-
acetyl-l-methylcyclopentane (Qg), rather than 2,2-dimethyl-
cyclohexanone (g2), which would result from methyl migra-
tion. In the pinacol rearrangement of the corresponding
gi§_(18) and trans (19) glycols (22) in sulfuric acid, the
only product formed was also §§x presumably through the same
intermediate, g1;

When the glycol rearrangement was carried out in per—
chloric acid at various temperatures (20), the amount of
§§’in the rearranged product was found always to be greater
than 90% (the remainder was g9). Bunton and Carr (20)

suggested that of the two possible intermediates, 37b

15

 

 

 

C
+
OH
OH §§.
3,32,
CH3
OH OH 0 CH3
CH3 CH3 + 37 CH3
H m
-H20
40 39

should be more stable than 37a since the carbonium ion cen-
ter can be stabilized by participation of the hydroxyl group.
In addition, 37b would be favored on steric grounds; the

methyl group, rather than the smaller hydroxyl group, would

 

+ CH3 + CH3
OH CH3
CH3 OH
37a 37b
be in the equatorial position (21). Conformation 37b can

only lead to ring contraction (methyl group not axial).

16
Regardless of the intermediate, ring contraction (sec-
ondary alkyl migration) would be favored over methyl migra-
tion according to migratory aptitudes in Wagner—Meerwein

rearrangements (4).

F. Oxidation of Ag'lo-Octalin (11)

 

Oxidation of A9'10-ootalin (21) at 00 afforded an
additional comparison with the pinacol rearrangement of the
corresponding trans (22) glycol (22). As expected A9I10-
octalin (21) gave the same spiro ketone (22) as the glycol,

in 86% yield.

OH OH
+
O OH 5 . E HgsoI
+ :
OH
41 4N1
”W 2.2..
+
-H
o
44

G. Oxidation of 2—Methyl-3—phenyl-2—butene ( 5)

m

The oxidation of 2—methyl-3—phenyl-2-butene (32) was

of interest because of the possibility of oxidation of the

17

aromatic ring (12) as well as oxidation of the double bond.

Oxidation of 42/ at reflux, showed that the only pro-
duct, 3-methyl-3-phenyl-2-butanone (4Q), arose from attack
on the double bond. The sample of 42 used was contaminated
with the isomeric 3-methyl-2-phenyl-1-butene (42), but oxida-
tion of two different mixtures of fig and 49 indicated that
under the reaction conditions the two became equilibrated,
with 42 being preferentially oxidized. This assertion is
based on the fact that an 80:20 mixture of fig and gg're-
sulted in a 96% yield of 4§ based on the original amount of

42, or a 77% yield based on the total amount of fig and 42;

 

CH3 a»
~ I
[I > CH3-C-¢ ——C§¥a—> CH3-C-¢
c _
/’ ‘\ 1
CH3 CH3 CH3 - C - CH3 /C = o
45 OH CH3
’W of 22, 2.12.
H
H OH
I I
C, CH3 - C - ¢ N
H/ \\ -¢ 1 (D
.. _ -H
CH3 —i -CH3 CH3 3. CH3
I
H 22.
49

On the other hand, a 46:54 mixture resulted in a 117% yield
based on the original amount of fig; Obviously some of 42’
must be rearranging to 42 in order to account for the large
amount of 43 found. On the basis of the total amount of

gfi’and 49 used, the yield of 4§ was found to be 73%. The

18
presence of 42, therefore, does not detract from the syn-
thetic utility of this oxidation.

The intermediate in the oxidation is probably 46 (the
more stable ion), which by methyl migration gives rise to
the product. The other intermediate, 41, could be involved,
with phenyl migration (preferred over methyl) leading to
4g” Whichever is the intermediate, 4§ would be the expected
product. Unfortunately, there are no reports in the litera-
ture on the rearrangement of the corresponding epoxide or

glycol for comparison purposes.

H. Oxidation of 3,4-Dimethyl-2,5-dihydrothiophene-1,1-
dioxide (éQ)

 

 

The oxidation of §Q’ was unsuccessful from a synthetic
point of view. The oxidation took place readily at reflux,
but only a small amount of material was isolated from the
organic layer (30% based on the amount of starting material).
These results would indicate that a highly water-soluble
product(s) was formed, rather than the expected product 22.
or E3; Perhaps this was a sulfonic acid (or degradation
product), though there does not appear to be a precedent
for this type of a reaction. It is also possible that Q2,
and 63 are sufficiently water soluble so that any of E2 or
Q3 formed were lost during the work-up procedure. Since
this oxidation was not of synthetic utility, no further in-
vestigation was carried out into the exact nature of the

products.

19

 

 

I. Oxidation of Isopropylidene Malononitrile (24)

 

To test the effect of strong electron—withdrawing
substituents, the oxidation of isopropylidene malononitrile
(Q2) under refluxing conditions, was attempted. No reac-
tion occurred and apparently the two nitrile groups caused
the double bond to be too electron deficient for an electro-

philic attack to take place.

NC CN
‘\C/’ +
H 9§——> No reaction
1’ ‘\
CH3 CH3

20

J. Oxidation of 3-Ethyl-2-methyl-2-pentenenitrile (éé)

The oxidation of Q2 showed that even one nitrile group
was sufficient to deactivate the double bond, so that elec-
trophilic attack did not occur under refluxing conditions.
Though a reaction did take place, the only volatile material
detected was starting material. A considerable amount of
tar was formed indicating polymerization and/or oxidation
of the nitrile group had occurred.

CH3 CN
\/

z/ ‘\
Et Et

0
+

 

O

55

m

K. Oxidation of 2-Methyl-2-butenal [tiglaldehyde (£2)]

Oxidation of tiglaldehyde (£2) was of interest, since
products could be formed by either acyl or methyl migration,
or by Bayer-Villiger oxidation (23) of the aldehyde to the
formate ester.

Both oxidation of the double bond and of the carbonyl
group have been noted in the oxidation of other a. B -

unsaturated carbonyl compounds as in the following example

(24):

21
O ‘o O

 

H C H C H -C—CH
\ / \ \ / \ f) 3
C CH3 C CH3 C
H CchoiHA / /
C 20-2507 0\| + O\l
CH3 CH3 / \ / \
56 CH3 CH3 CH3 CH3
£1 22:

Upon oxidizing tiglaldehyde (59) at reflux, it was
found that any initial oxidation product was further oxidized.
When a slight excess of peracid was used, it was all rapidly
consumed, but out of 25.2 g of 52 used, 14.6 g was recovered
and only 3.2 g of volatile product was isolated by distil-
lation. This product appeared to consist of two compounds.
They were tentatively assigned the structures 62 and 62;

A mixture of 62 and 62 showed strong carbonyl absorption
in its ir spectrum at 1722 cm_1 and strong absorption be-
tween 1150 and 1250 cm-1, which could be associated with
an ester group. In its nmr spectrum, 62 showed a singlet
at T 1.96 (1 H) assigned to the formate proton, a quartet
at T 4.89 (1 H) assigned to the proton a to the carbonyl, a
singlet at T 7.87 (3 H) assigned to the methyl attached to
the carbonyl, and a doublet at T 8.62 (3 H) assigned to
the remaining methyl group. Compound 63 showed a quartet
at T 5.03 (1 H), assigned to the proton attached to the
carbon bearing the methyl group, a singlet at T 7.90 (6 H)
assigned to the methyl protons of the two acetyl groups,
and a doublet at T 8.67 (3 H) assigned to the remaining

methyl group.

22
Product 62 could be rationalized as the result of
electrophilic attack on the double bond followed, or pre-

ceeded, by Bayer-Villiger oxidation of the aldehyde.

 

 

 

O
CH C-H OH O CH
3\ / I II I 3
+ _ .. =
C CH CH3 C C\H ~CHO C 0
II > _H. > |
C H - C - CH - H - C - CH
/ \ + 3 I 3
H CH3 /C = o
22. 22. H 2;
CH3
I
C = O
|
H - C - CH3 < CF3COqH
O
C = O
/
H
:52,

The presence of fig, if the structure is correctly as-
signed, cannot as easily be explained, but it is apparently

an oxidation product from the reaction.

 

0 CH3 0
II I II
CH3 - C - o - C C - CH3
H
:35:

House has observed acyl migrations in the rearrangement
of a number of a. fi—epoxy ketones (25) so that conversion

of Qg’to gl'in the above scheme is not without precedence.

23

L. Oxidation of 3,4-Dimethyl-3-penten-2-one (64)

 

In contrast to 52x 3,4-dimethyl-3-penten-2-one (64)
was oxidized (reflux) at the double bond, leading to an
81% yield of 66; The fact that some starting material was
recovered (conversion of 64 to 66 was 61%) indicates that
some further oxidation of the initial product did take place,
but apparently the addition of two methyl groups (compare
Qg'with 64) activated the double bond sufficiently for prefer-
ential attack at that position.

Product 62 could have arisen yia_methyl or acyl migra—

tion. House and Wasson (26) have observed acyl migration

 

 

 

 

o o cH3
" + II /
CH3 C ' CH3 CH3 " C " C - CH3 0 = C
\\C// + .| I
H OH > CH3 - C — CH3 ~CH3 > CH3 - C - CH3
I + |
C OH -H _ '
// ‘\ O - C \‘CH’
CH3 CH3 65 66 a
w 11"”
QC
, OH O
+ I II
OH CH3- C - C - CH3 «vCOCH
> H 1L
‘ 1"
CH3- C - CH3 *H
+
9,2,

in a related syStem, where they converted gg'to 72 by boron

fluoride etherate catalyzed epoxide ring opening.

24

 

o
H C - CH H o
\ / 3 - I II
C F3BO ‘- C " C
0/ BEL-EtQO > \ CH
\\C ¢H CH C CH 3
CH / \CH 3 3
3 3 ~C0CH
3 69
68 ’“‘

In order to decide between acyl or methyl migration in
our system, deuterium labeled starting material 21' was pre-
pared and oxidized. The resulting deuterated diketone 22'
showed a single peak in its nmr spectrum at T 7.93. Acyl

migration must have taken place.

 

0
CH3 C - CD3 OH O
\ II
C OH+ CH3 - C - C - CD3
II ——-->
C CD3 - C - CD3
/’ ‘\ +
CD3 CD3
4%
7.1,
CH3
\\C = O
I
NCOCD
CD - C - CD 3
3 I 3 T
C = O
./
CD3

25
If methyl migration had occurred, the resulting di-
ketone 12 would have had a single nmr absorption at higher

field (52- 1- 8.7).

 

0 \
C "‘ " C = 0
H3 - C - C - CD3
71 OH+ > I 1%93—> CH — C - CD
m CD3 ' C - CD3 “H 3 ' 3
OH C = o
,/
CD3
74 75

M. Oxidation of 2—Cyclopentylidenecyclopentanone (22)

 

An additional example of acyl migration was found in
the oxidation, at reflux, of 2-cyclopentylidenecyclopentan-
one (12) which gave spiro [4.5]decane-6,10-dione (lg) in

43% yield (conversion, 27%).

OH 0 0
(XE-fl; O —9°R O.
O .
2.8..

Surprisingly, Z§ does not seem to have been previously
prepared. Its structural assignment rests on spectral
data and chemical transformations. It had infrared carbonyl
bands at 1720 and 1695 cm-1. The nmr Spectrum showed a
triplet (4H) at T 7.38 (methylene protons a to the carbonyl)
and multiple absorption between T 7.65 and 8.60 (10H). Com-
pound Zg'was converted with base to 5-cyclopentyl-5-oxo-

pentanoic acid Z§J whose semicarbazone had a melting point

26

in agreement with the reported value.

0

22 5:21: —> MW

2.2.

 

Interestingly House (26) could not affect a similar
rearrangement of the epoxide of 2-cyclohexylidenecyclo-
hexanone (§2) with boron fluoride etherate, though it could

be rearranged thermally to give the expected product g};

0

BF3 -Et20

O

o.
...) 1..
f 260° V

N. Oxidation of 3-Chloro-2-methyl-2-butene (§Z)

 

 

Oxidation of 3—chloro—2—methyl-2-butene (§2) offered
the interesting possibility for methyl or chlorine migra—
tion. The less common chlorine migration has been reported
in other olefinic systems (27). The only oxidation product
of §£ was 3-chloro-3-methyl-2-butanone (fig), obtained in
77% yield. The structure of the product was assigned on
the basis of its nmr and infrared spectra (see Experimental
Section). This product could arise from two possible inter-
mediates, §§x in which a methyl migrates, and g2” in which

chlorine migrates.

 

 

CH3
+ I
CHa-C‘Cl CHs-C-Cl
«CH I
‘_> > c = 0
CH3 - c — CH3 ”H /
OH CH3
CH3 c1
\/ §.§. £4.
C +
H OH
c OH
/\ .
CH3 CH3 CH3 - c - c1 «'c1
_> I
§g’ CH3 — c - CH3 -H
+
85

To decide between these alternatives, deuterium labeled
starting material §§,was prepared and oxidized. The product
isolated had an nmr spectrum consisting of one singlet at
T 8.36. If methyl migration had occurred the product §§’
would have had two singlets. The actual product must have

arisen, therefore, from chlorine migration where the expected

CD3 C1 C1 c1
\ / ' I
C OH+ CD3 - c-+ «'CH CD3 - c - CH3
H X > I +3 > I
c CH3 - c - CH3 ‘H c = o
/’ ‘\ - /
CH3 CH3 OH CH3
5‘3 él 9.2

ketone 22 would have only a single peak in its nmr spectrum.

c1 CD3
' '\
+CD3-C’OH C=O
OH «'c1
§§. ___‘> | ‘—-*r-> l
CH3 - c - CH3 -H CH3 - c - CH3
+ I
Cl

89 90

m m

28

O. Oxidation of 3—Bromo-2-methyl-2-butene (21)

 

To determine whether halogens, other than chlorine might
migrate, 3-bromo-2-methyl-2-butene (2}) was oxidized. The
major product, obtained in 79% yield, was the expected 3-
bromo-3-methyl-2-butanone (22) identified by its infrared
and nmr spectra. Since suitably labeled bromobutene could

not be synthesized easily to test whether methyl or bromine

CH3\ /Br CH3,
C C = 0
II _Qli_+___> I
C CH - C - CH
/ \ 3 . 3
CH3 CH3 Er
2.1. l 22,

migration had occurred, attention was turned to the oxida—

tion of the following compound.

P. Oxidation of 2,3-Dibromo-2-butene (22)

 

In the oxidation of 2,3-dibromo-2-butene (22) two dif-
ferent products fig and 22/ would be obtained from bromine
and methyl migration, respectively. Vpc analysis showed
only one product from the oxidation of fig” identified as
3,3—dibromo-2-butanone (22), obtained in 69% yield (con-
version, 61%). The structural assignment rests on infra-
red and nmr spectra, elemental analysis, and independent
synthesis (see Experimental Section). Since the product
was a ketone rather than an acid, bromine migration must
have taken place. Although bromine migration in other car-

bonium ion rearrangements have been reported (28) it appears

29

 

CH3 Br Br
I
\\c’/ + . CH3 - c - OH
OH
H >
c Br - c - CH3
/’ ‘\ +
Br CH3
2.12. 2.4.
~ CH3
«Br
CH3 Br HO
\. \x \
c = o c = o c = o
I I 522—> I
CH3 - C - Br CH3 - c - CH3 CH3 - C - CH3
I I I
Br Br Br
2,52. 2E. 2.2.

that these are the first examples of a pinacol—type rear-

rangement with bromine as the migrating group.

Q. Generalizations about the Oxidation of Alkenes to Ketones

Using Peroxytrifluoroacetic Acid - Boron Fluoride

1. Migratory Preferences. As stated earlier, the
migratory aptitudes we have observed parallel those found
in other Wagner-Meerwein rearrangements. The actual prefer-
ences that we have found are H > 2° alkyl, or methyl;
acyl > 2° alkyl >~ methyl; and chlorine or bromine > methyl.
The reasons for various migratory aptitudes are not
at all clear since steric hindrance, neighboring group par-
ticipation (during formation of the carbonium ion), and

solvent or temperature effects may play important roles in

3O
determining the ability of a group to migrate (4,11,29).
In general, the ability to stabilize the positive charge in
the transition state is always an important factor. The
order of the migratory aptitudes of alkyl groups (3° > 2°
> 1°) can be explained in this way. Stabilization of the
positive charge is increased with an increase in the number
of electron-donating groups attached to the migrating carbon
atom. The addition of alkyl groups (which are electron-
donating) would then lower the energy of the transition

state and thus increase the probability of migration.

 

R R R R
HO \\;g< O{\c_T—c/,R ;R\Q+ C/IR
(I: “ W H O /
x /)|{\x x /3 \x X :\ x

The high migratory aptitude of hydrogen may be due to
its small size. It could easily assume the proper posi-
tion (parallel to the vacant p orbital) for migration to
the electron-deficient center. The fact that an s orbital
is involved in the migration might be an important factor
in the ability of hydrogen to migrate. The spherical s
orbital may be capable of better overlap with the other
orbitals in the transition state than would the sp3 orbital

of carbon.

31

 

R R
R
HO R
H04 +_/R. \ _ 4. ;R\ + /R
l R HO h

The rather high migratory aptitude of acyl groups
might be explained by the possibility of additional de-
localization of the positive charge to oxygen. This addi-
tional delocalization should lower the transition state

energy and thereby favor the migration of this group.

 

R
R HO R
H. O) \ / R
\\C + 5‘R —9 \C—TC’ ~_) R\ QLC QR
L , 30'73
/ l/
R 0

Migration of bromine or chlorine might be favored
over methyl migration because of the possibility of overlap
of the available p orbitals of these halogens in the

transition state. In fact, Olah (30) observed the nmr

 

x = Cl, Br

32
Spectra of halonium ions, when 2,3-dihalo-2,3-dimethyl-
butanes were ionized at -60° in antimony pentafluoride-

sulfur dioxide.

CH3 CH3 CH3\ /CH3
' ' SbF -SO

CH3 - ‘3 " ‘5 ‘ CH3 720—04" /C\T/C\
x x CH3 x CH3

X = I, Br, Cl.

Another possibility is the formation of an ion-pair
from the initially formed epoxide (if the peroxytrifluoro-
acetic acid - boron fluoride oxidation of alkenes involves

an epoxide intermediate). Such an intermediate has been

0 o R 0 RR
/f-/\ II + II I
R-c—c-R >R-C-C,‘/ >R-C-C’

x? \z x" R x

suggested by McDonald and Tabor (27) to explain the pro-

 

 

ducts they found in the thermal rearrangement of a-chloro-
epoxides. Even if such an ion-pair were formed in the
rearrangement of the chloroalkenes under our oxidation con-
ditions, it would not necessarily be true for the bromine
migrations we have observed. Since bromine is less electro—
negative than chlorine, ion-pair formation may not be as
favored and the formation of a bromonium ion intermediate

or transition state may occur instead. In fact, any of the
migrations that we have observed could be explained on the
basis of a bridged intermediate rather than a transition

state. Whatever the factors are which are involved in

33
determining the migratory aptitudes that we have found,
they seem in no way different from those observed in re-

lated systems.

2. The Effect of Substituents on the Double Bond.
Since the attack of "positive hydroxyl" on the double bond
of alkenes is electrophilic, it would be reasonable to be-
lieve that such an attack would be more highly favored when
the electron density of the double bond is increased and
less favored when the electron density is decreased. The
fact that we found that tetra-alkyl substituted alkenes
gave better yields of ketones than tri-alkyl substituted
alkenes supports this belief. Since alkyl groups are
electron-donating, then the more alkyl substituents present,
the higher would be the electron density at the double bond.
Since the peracid can attack the initially formed ketone
to give further oxidation products, there is competition
between oxidation of the double bond and the oxidation of
the ketone. The more susceptible the alkene is to oxidation,
the less likely is the probability of further oxidation of

the product.

 

 

R /R R 91}
\ Ro-c-c-R
0H+ C = 0 CF co H '
II > I 1*" > R
C R - C - R or
/\ '
R R R o R

34
The over-oxidation of tiglaldehyde (Q2) is also a re-
flection of reduced electron density about the double bond.
Compound Q2 is only tri-substituted and one of the substitu-
ents is an electron-withdrawing aldehyde group. Upon re-

placing the hydrogen and aldehyde substituents with methyl

H\\ z/CH3
C:::C 3 ester formation
/ \ ,H
CH3 C
22. O
I...
+ I 3
CH3 ._
O

and acetyl groups, the increased electron donation to the
double bond is apparently sufficient to cause the oxidation
of the double bond of £2 to be preferable to the oxidation
of the diketone product §§'(though, some oxidation of the
product fig must also take place since some starting material
is recovered; conversion of §g to fig was 61%). In the case
of both §2 and Q2“ oxidation of the carbonyl group of the
starting material may take place prior to oxidation of the

double bond as well as after.

 

CH3 CH3 CH3
\xca/ + O _ C/’
H OH > I 81%
c CH3 - c - CH3
/ \c - CH3 |
CH3 5 o = c
\.
CH3
64 66

W w

35

The presence of halogens about the double bond might
also be eXpected to lower the yield of the anticipated
product due to the electronegativity of halogens. But even
in the oxidation of 2,3-dibromo-2-butene (2g), the yield
of ketone 22 was 69%. The relatively high yields of ketones
obtained in the oxidation of the haloalkenes that we have
studied may be due to the ability of halogens to supply
electrons to the double bond through participation of their

non-bonding electrons.

:éf CH3 Br
\\c’/ OH+ CH3 - C - Br
ll '———————9 I
c\\ c = o
/’ ‘. I
CH3 éf- CH3
2Q. 2Q.

l

.0 -
3?? - c - CH3
l
c

/’

CH3 Br.
0 O

The decrease in electron density of the double bond

when nitrile substituents are present is apparently great
enough to deactivate the double bond toward effective elec-

trophilic attack. In the case of the alkylidenemalono—
nitrile g2, no reaction with peroxytrifluoroacetic acid -

boron fluoride took place at all.

36

[I -———————+ No reaction

From our results it is apparent that the conversion
of alkenes to ketones is most effective when the alkene
being oxidized is highly substituted with electron-donating

groups, and least effective when substituted with electron-

withdrawing groups.

EXPERIMENTAL

A. General Procedures

1. Melting Points. Melting points were measured on

a Gallenkamp melting point apparatus and are uncorrected.

2. Microanalyses. Microanalyses were carried out by

 

Spang Microanalytical Laboratory, Ann Arbor, Michigan.

3. Infrared Spectra. Infrared spectra were obtained

 

on either a Unicam SP 200 or Perkin-Elmer 237B spectro-

photometer in CCl4 solution unless otherwise stated.

4. Mass Spectra. Mass spectra were obtained with a

 

Consolidated Electrodynamics Corporation 21-103C instrument.

5. Nuclear Magnetic Resonance Spectra. These spectra
were obtained with either a Varian A-60 or JEOL JNM-C-6OH
spectrometer, using CC14 solutions with tetramethylsilane'

as an internal standard.

6. Gas Chromatography. All gas chromatographic work

was carried out with Varian-Aerograph instruments.

B. General Oxidation Procedures.

1. Using Fisher certified methylene chloride, a
methylene chloride solution of peroxytrifluoroacetic acid

(31) was prepared by adding in sequence 15.3 g (0.073 mole)

37

38
of trifluoroacetic anhydride (City Chemical Corporation)
and 1.78 ml (0.066 mole) of 90% hydrogen peroxide (FMC
Corporation) to 20 ml of methylene chloride at 00.. The
solution was gently swirled at 0° until homogeneous, then
allowed to warm to room temperature for several minutes
before once again cooling to 0°.

A solution of 0.060 mole of alkene in 100 ml of
methylene chloride was prepared in a 500-ml three-necked
flask bearing an ice-jacketed addition funnel, thermometer
and pressure—equalizing addition funnel. All openings were
equipped with drying tubes containing drierite. The solu-
tion was cooled to -5° in an ice-ethanol bath and agitated
by means of a magnetic stirrer. The cold peroxytrifluoro-
acetic acid solution was then added dropwise at a constant
rate to the alkene solution by means of the ice-jacketed
addition funnel over a period of 20 minutes, during which
time 8.3 ml (0.066 mole) of 47% boron fluoride etherate
(Eastman Organic Chemicals) was added, also dropwise at a
constant rate, to the alkene solution by means of the pres—
sure-equalizing funnel. The temperature was kept below 8°
during the addition of the reagents.

The mixture was stirred at 0° for 15 minutes after
addition was completed, then hydrolyzed with 35 ml of water.
The organic layer was washed consecutively with two 35-ml
portions of water, three 35-ml portions of saturated aqueous
sodium bicarbonate, and two 35—ml portions of water. The

solution was then dried over anhydrous magnesium sulfate,

39
filtered, and the solvent partially evaporated by means of
a rotary evaporator. The residue was diluted with methylene
chloride to exactly 100 ml. A small portion of this solu-
tion was set aside for yield determination by Vpc comparison
with a standard methylene chloride solution of the product.
The solvent was removed from the remainder of the solution
and Vpc analysis carried out on the residue. It should be
noted that many alkene solutions turned deep blue or violet
during the addition of the reagents and that these solutions

faded to yellow or orange during the workup procedure.

2. The above procedure was followed except that the
methylene chloride solution of alkene was not cooled, and
the rate of adding reagents was adjusted so that the solu-
tion refluxed gently. Addition time of the reagents was

10 minutes.
C. Preparation of Alkenes.

1. A9r1°-Octalin (21). The procedure of Campbell and

 

Harris (32) was followed. To a mechanically stirred solu-
tion of 200 g of phosphorus pentoxide in 2000 g of 85%
phosphoric acid was added 190 g of mixed B-decalols. The
temperature of the mixture was raised to 150° for 10 min-
utes. While this temperature was maintained, a slight
vacuum was applied and the octalin was steam distilled by
the dropwise addition of water. The distillation was con-

tinued until no more octalin was obtained. The product was

40
taken up in ether, dried over anhydrous magnesium sulfate
and filtered. After the solvent was removed, the product
was distilled from sodium (bp 190—1940). The yield was
99.3 g (56%).

This product was heated, with stirring, with 44 g of
phosphorus pentoxide at 140° for 1.5 hrs. The flask was
cooled and ice added to react with the phosphorus pentoxide.
The octalin was taken up in ether, dried over anhydrous mag-
nesium sulfate, filtered, and the solvent removed. Distil-
lation on a stainless steel spinning band column gave 35 ml
of product, bp 193-194° (lit. 32, 190-192°) which was 94%
pure according to Vpa analysis on a 5-ft Carbowax column
at 150° (ir spectrum, Figure 18). The major impurity, with
a lower vpc retention time and bp, was apparently the iso-

meric A1v7-octalin (lit. 32, bp 189.5-193.5°).

2. 1,2-Dimethylcyclohexene (32). Methylmagnesium

 

iodide was prepared from 24.3 g (1.0 mole) of magnesium

and 93 ml (1.0 mole) of methyl iodide in a final volume of
300 ml of anhydrous ether using the standard procedure.

The Grignard solution was cooled in an ice-ethanol bath.
2-Methylcyclohexanone (100 g, 0.89 mole) was added to the
vigorosuly stirred solution by means of a dropping funnel,
at such a rate that the temperature of the reaction mixture
was kept below 5°. After addition was completed the mixture
was allowed to warm to room temperature and hydrolyzed with
40 ml of concentrated sulfuric acid in ice and 200 ml of

water. The two layers were separated and the aqueous layer

41

was extracted with two 100-ml portions of ether. The
combined ether layers were then washed consecutively with
100 ml of water, to which had been added 5 ml of saturated
aqueous sodium bicarbonate, and twice with 100-ml portions
of water. The ether solution was then dried over anhydrous
sodium sulfate, filtered and the solvent removed with a
rotary evaporator. The resulting oil weighed 88.6 g (78%
if all the oil was alcohol).

The alcohol was dehydrated by adding a small crystal
of iodine and distilling through a 6-in Vigreaux column.
The product which distilled from 88 to 133° was saturated
with salt and the organic layer separated. After being
dried over anhydrous calcium chloride, the filtered product
was redistilled through a stainless steel spinning band
column. The fraction boiling at 135° (17 ml, lit. 33,
136°) was 93% pure according to Vpc analysis on a 5-ft
SE-30 column at 100° (nmr and ir spectra, Figures 1 and 19

respectively).

3. 2-Methyl-3-phenyl-2-butene (45). In a one-liter

 

three—necked flask equipped with a reflux condenser protected
with a drying tube, mechanical stirrer and pressure-equaliz-
ing funnel was added 260 ml (0.78 mole) of three molar
phenylmagnesium bromide (Arapahoe Chemicals). The stirred
solution was cooled with an ice bath and by means of the
pressure-equalizing funnel, 70.5 ml (0.67 mole) of 3-methyl-

2-butanone in 75 ml of anhydrous ether was added, dropwise.

42
After addition of the ketone was completed, the reaction
mixture was stirred at room temperature for one hr, then
poured into a mixture of ice, 200 ml of water and 40 ml of
concentrated sulfuric acid. The layers were separated and
the aqueous layer extracted with ether. The combined ether
layers were washed consecutively with 100 ml of saturated
aqueous sodium bisulfite, two 100-ml portions of water and
100 ml of saturated aqueous salt solution. After the pro-
duct was dried over anhydrous magnesium sulfate, filtered,
and the solvent evaporated with a rotary evaporator, 60 g
of potassium bisulfate was added to the residue and the
material was vacuum distilled. The distillate (65 g, 66%)
was a clear colorless liquid, bp5 68-69° (lit. 34, bpzo 95-
970). Vpc analysis on a 5-ft SF-96 column at 170° showed
one peak but nmr analysis showed that the product was a mix-
ture containing 63% of 2-methyl-3-phenyl-2—butene and 37%
3-methyl-2-phenyl-1-butene. The nmr spectrum (Figure 2)
showed a complex absorption, assigned to the phenyl protons
of both compounds, centered at T 2.90. The slightly split
doublet (J = 7.0 cps) at T 4.95 (2 H), multiplet (J = 7.0
cps) centered at T 7.23 (1 H) and the doublet (J - 7.0 cps)
at T 8.93 (6 H) were assigned to the latter compound while
the three broad and slightly split singlets of equal in-
tensity at T 8.08, T 8.25 and T 8.40 were assigned to the
former. This material was distilled through a stainless

spinning band column to give three fractions; these consisted

43
of 15 ml (bp5 65-690) of material which contained 46% of
the desired isomer, 20 ml (bp5 69—69.5°) of material which
contained 57% of the desired isomer and 10 ml (bp5 69.5-69.50)
of material which contained 80% of the desired isomer (nmr

spectrum, Figure 3).

4. 3,4-Dimethyl-3-penten—2-one (64). This compound

 

was prepared following the method of Colonge and Mostafari
(35). Stannic chloride (4.48 ml) was added, dropwise, to
a stirred, ice—cold solution of 150 ml (1.43 moles) of 2-
methyl-Z-butene and 50 ml (0.66 mole) of acetyl chloride
in a 500-ml three-necked flask equipped with a mechanical
stirrer, thermometer and pressure-equalizing funnel. The
temperature of the reaction mixture was kept between 10 and
22° during the addition. The reaction was very exothermic
during the addition of the first two ml of stannic chloride.
The reaction mixture was stirred at 0° for 15 min after the
addition was completed. After being warmed to room tempera- '
ture, the reaction mixture was poured onto 400 ml of 15%
hydrochloric acid, washed consecutively with water and
saturated aqueous sodium bicarbonate, dried over anhydrous
magnesium sulfate, filtered, and the solvent removed with
a rotary evaporator. The residue was vacuum distilled and
the distillation stopped at 31° (45 mm).

The residue was refluxed with 85 g of dimethylaniline
for one hr. The upper layer was then washed with water,

dried, filtered, and distilled through a 30-cm column packed

44
with glass helices. The product collected (22.2 g, 30%,
bp 144—148° (lit. 35, 146°)) was 94% pure according to Vpc
analysis on a 5-ft FFAP column at 140°. The nmr spectrum

of this compound is shown in Figure 4.

5. 4-Methyl-g3-3-methyl-3-penten-2—one-1,1,1,5,5,57§5(71).

 

3,4-Dimethyl-3-penten-2—one (18 g, 0.16 mole) was stirred
for 77 hrs with 238 g (7.2 moles) of MeOD, to which a small
piece of sodium had been added. The alcoholic solution was
then poured into 250 ml of methylene chloride and washed
with 50—ml portions of water until the aqueous layer was
clear and neutral to litmus. The methylene chloride layer
was then dried, filtered and the solvent removed with a
rotary evaporator. The colorless, distilled product, bp
144-1480, was 90% pure (vpc analysis) and the nmr spectrum
(Figure 5) showed that the two olefinic methyl groups were
66% deuterium labeled and the methyl group in the one posi-

tion at least 98% labeled.

6. 2—Cyclopentylidenecyclopentanone (76). The procedure

 

of Mayer (36) was followed. To a two liter flask cooled
with an ice bath and equipped with a dropping funnel, con-
denser and drying tubes containing drierite, was added 450
ml of absolute ethanol followed by 30 g of sodium. After
all the sodium had reacted, 400 g (4.8 mole) of cyclopentan-
one was added, dropwise, to the chilled solution. The re-

sulting red-brown reaction mixture was kept at 0° for one

45

day and the solution then decanted from the solid which
had formed. -Most of the ethanol was removed with a rotary
evaporator. -Water was then added to the residue and the
two resulting layers were separated. The organic layer,
after being dried over anhydrous magnesium sulfate and
filtered, was vacuum distilled. -The resulting pale green
liquid (85 g, 24%) lope,7 83-850 (lit. 36, bp2 88.5-90.00),
was greater than 99% pure according to Vpc analysis on a

10-ft FFAP column at 235° (ir spectrum, Figure 20).

7. 3-Chloro-2-methyl-2-butene (82)” This alkene was

 

prepared following the procedure of Behal (37). 3-Methyl-
2—butanone (68 g, 0.8 mole) was added dropwise, with stir-
ring, to 166 g of phosphorus pentachloride cooled with an
ice bath. After addition was completed, the reaction mix—
ture was allowed to warm slowly and refluxed 3.5 hrs (HCl
was evolved). The resulting solution was poured onto ice
and the layers were separated. The organic layer was washed
with saturated aqueous sodium bisulfite followed by water.
After being dried over anhydrous magnesium sulfate and
filtered, the crude product was distilled through a Teflon
spinning band column. The colorless liquid, bp 93-98O
(lit. 37, 97-980) was collected (7.9 g, 9.3%) and by vpc
analysis on a 5-ft XF-96 column at 70° was 90% pure (nmr
spectrum Figure 6). The major impurities appear to be the
starting ketone and the other dehydrochlorination isomer,
2-chloro—3-methyl-l-butene (ir bands at 1710 and 1635 cm-1

respectively).

46

8. 2-Chloro-3—methyl-2-butene-1,1,1-g3 (g3). This

 

compound was prepared as above except that 3-methyl-2-
butanone-1,1,1,45QA was used. This ketone was prepared by
refluxing, for two days, 90 ml (0.85 mole) of unlabeled
ketone with 125 ml (6.9 moles) of D20, to which had been
added a Spatula of anhydrous potassium carbonate. The two
layers were then separated and the aqueous layer was ex-
tracted with methylene chloride. The organic layers were
combined, washed with water, dried over anhydrous magnesium
sulfate, filtered and the solvent distilled (the distilla-
tion was stopped at a head temperature of 48°). The nmr
spectrum showed that the one position was about 80% deu-
terium labeled. The alkene prepared from this ketone also
showed, in its nmr spectrum (Figure 7), about 80% deuterium

label in the one position.

D. Results of Alkene Oxidations

 

1. 2,3-Dimethyl-2-butene (17). This alkene (7.09 ml,

 

0.060 mole, Aldrich ChemicalCo.) was oxidized by procedures
B1 and B2. Vpc analysis on a 20-ft SE-30 column at 80°
showed one major product peak. This product was identified
as 3,3-dimethyl-2—butanone (pinacolone) by comparison of

its vpc retention time, and ir and nmr spectra with those

of an authentic sample. The yield of pinacolone was 72 and

75% by procedures B1 and B2 respectively.

47

2. 2-Methyl-2-butene (21). The oxidation of 6.3 ml

 

(0.060 mole) of 2-methyl-2-butene (Aldrich Chemical Co.)

by procedure B1 resulted in a 53% yield of 3-methyl-2-
butanone. Vpc analysis was carried out on a 5-ft FFAP
column at 80°. The product was identified by comparison

of its vpc retention time and ir and nmr spectra with those

of an authentic sample.

3. cis-3-Methyl-2-pentene (26). The oxidation of 7.3

 

ml (0.060 mole) of this alkene (Aldrich Chemical Co.) by
procedure B1 resulted in a 63% yield of 3-methyl-2-pentanone
according to Vpc analysis on a 5—ft SE-30 column at 120°.
The product was identified by its nmr spectrum (triplet

at T 9.15 (3 H), doublet at T 8.97 (3 H), multiplet at T
8.57 (2 H), singlet at T 7.97 (3 H) and multiplet at T

7.63 (1 H)) and the identity of its ir spectrum with the

published spectrum (38).

4. trans-3—Methyl-2—pentene (22). This alkene (7.3

 

ml, 0.060 mole, Aldrich Chemical Co.) was oxidized in the
same manner as the gi§_isomer. The yield of 3-methyl-2-
pentanone was 70%.

If the boron fluoride etherate is excluded from the
oxidation, vpc analysis of the residue shows only small
amounts of ketone, the major volatile products being three
compounds with longer retention times which have the fol-

lowing ir bands in common: 3600, 1780 and multiple bands

48
between 1250 and 1050 cm-1. These products appear to be
hydroxytrifluoroacetates similar to the ones Emmons (3)
has obtained in his studies on peroxytrifluoroacetic acid
oxidations of alkenes. This residue was then dissolved in
100 ml of methylene chloride, cooled to 0° and 17 ml of
boron fluoride ethereate was added dropwise, with stirring.
After standing at 0° for 1 hr after addition was completed
the solution was worked up as described in procedure B1.
Vpc analysis of the residue now showed only one peak which

was identified as 3-methyl-2—pentanone.

5. l-Methylcyclohexene (32)u This alkene (7.0 ml,

 

0.060 mole) was oxidized by both procedure B1 and B2. Using
procedure B1 the yield of the only detectable oxidation
product, 2-methylcyclohexanone, was 8%. Vpc analysis was
carried out on a 10-ft Apiezon L column at 115° and this
product was identified by comparison of its ir and nmr
spectra and vpc retention time with those of an authentic
sample. The only other major volatile product detected

was 1-methylcyclohexyl trifluoroacetate (15%). This pro-
duct was identified by its nmr and ir Spectra (Figures 8

and 21 respectively). It was also prepared as the exclusive
product by mixing at 0°, 0.0041 mole each of l-methylcyclo-
hexene in 70 ml of methylene chloride, boron fluoride ether-
ate and trifluoroacetic acid. This solution was kept at

0° for 3 hrs, hydrolyzed with 10 ml of water and washed

with 10 ml of 5% aqueous sodium hydroxide followed by 10 ml

49
of water. After the product was dried over anhydrous mag-
nesium sulfate, filtered and the solvent evaporated, vpc
analysis showed one peak with the same retention time and
ir spectrum as the trifluoroacetate found in the oxidation.
Following procedure B2, the yield of 2—methylcyclo-
hexanone was 41%, whereas the yield of trifluoroacetate

was 9%.

6. 1,2-Dimethylcyclohexene (36). Procedure B1 was

 

followed using 7.1 g (0.060 mole) of 93% pure 1,2-dimethyl-
cyclohexene. One major product was detected by vpc analysis
on a 5-ft SE—30 column at 150° and it was assigned the
structure 1-acety1-1-methylcyclopentane on the following
evidence. The compound had a carbonyl band in the ir at
1700 cm-1; the nmr spectrum consisted of two singlets at

T 7.87 and 1 8.77 (3 H each) and a broad band at 1 7.93-
8.85 (8 H). The distilled material, bp7 50 (lit. 18,

bp11 50.2-50.9) gave a semicarbazone, recrystallized from
20% ethanol, with a mp 139.5-141.0° (lit. 18, 141°). The

yield of ketone was 76%.

7} Afi'1°-Octalin (41). Procedure B1 was followed, 8.7

 

g (0.060 mole) of 94% pure octalin gave a 79% yield (vpc
analysis on a 5—ft SE-30 column at 180°) of spiro[4.5]decan-
6-one (86% yield based on consumed starting material). The

structure assignment was based on the following evidence

50
The compound had a carbonyl band in the ir at 1700 cm-1
and the distilled product, bpl 45-55 (lit. 39, bpzo 105-

110), gave a semicarbazone which, recrystallized from

ethanol, had a mp of 187-190° (lit. 40, 188-190°).

8. 2—Methyl—3-phenyl-2-butene (42). A mixture (8.8 g,

 

 

0.060 mole) containing 80% 2-methyl-3-phenyl-2-butene and
20% 3-methyl-2-phenyl-1-butene was oxidized by procedure
B2. The single product (vpc analysis on a 5-ft SF-96
column at 170°) was 3-methyl-3-phenyl-2-butanone in 77%
yield based on the total amount of both alkenes and 96%
based on the tetra-substituted alkene. When the oxidation
was repeated using a mixture containing 46% 2-methyl-3-
phenyl-Z-butene and 54% of the other isomer, the yield of
3-methyl-3-phenyl-2-butanone was 73% based on both alkenes
and 117% based on the tetra-substituted alkene.

The product was assigned its structure on the follow-
ing evidence. It had a carbonyl band in the ir (liquid

KBr at 1701 cm-1) and an nmr

film) at 1701 cm'1 (lit. 41, v
spectrum consisting of three singlets at T 2.79 (5 H), T
8.08 (3 H) and T 8.58 (6 H). The semicarbazone, obtained

from vpc purified material, after recrystallization from

ethanol, had a mp of 185-1870 (lit. 42, 185-1870).

9. 3.4—Dimethyl-2,5-dihydrothiophene-1,1-dioxide (52).

 

This alkene [(43), 7.9 g, 0.054 mole] was oxidized by pro-
cedure-BZ but worked up as follows. The reaction mixture

was hydrolyzed with 35 ml of water, 20 min after the addition

51
of the reagents was completed. The organic layer was then
washed with two 35-ml portions of water followed by 35-ml
portions of 10% aqueous sodium hydroxide until the aqueous
layer was no longer yellow. The combined alkaline washes
were acidified with concentrated hydrochloric acid, saturated
with salt and extracted with four 50-ml portions of methylene
chloride. After the product was dried over anhydrous mag-
nesium sulfate and filtered the solvent was evaporated
leaving a residue (2.0 g) of viscous yellow oil. The ori-
ginal methylene chloride layer, after drying, filtering and
evaporating the solvent, yielded 0.4 g of off white solid
which decomposed at 158°. These products were not further

investigated due to the small amount of material recovered.

10. Isopropylidene Malononitrile (54). This compound

 

did not react under the reaction conditions of procedure
B2. The reaction mixture did not warm nor darken in color
during the addition of the reagents. After workup, the
colorless residue showed only one peak on the vpc (IO-ft
FFAP column at 225°) with the same retention time as the

starting material.

11. 3-Ethy1-2-methyl-2-pentenenitrile (55). This

 

alkene [(44), 7.8 g, 0.060 mole] was oxidized by procedure
B2. This oxidation differed from most of the other alkene
oxidations studied in that the solution did not warm to re-

flux during the addition of the reagents until the addition

52

was almost complete and then continued to reflux, without
external heating, for about 10 min. After workup, the
residue Showed only one major peak in the VpC (5-ft SE-30
column at 150-240°) which had the same retention time and
ir spectrum as the starting material. Vacuum distillation,
bpz 40-1330, gave mostly starting material with consider-
able amounts of tar as the residue. Investigation of this

oxidation was stOpped at this point.

12. 2-Methyl-2-butenal [(52), tiglaldehyde]. This

 

compound (25.2 g, 0.30 mole) was oxidized using a variation
of procedure B2 and five times the quantity of reagents and
solvent. The addition time of the reagents was 55 min.

The reaction mixture was then refluxed for 3 hrs, hydrolyzed
‘with 125 ml of saturated aqueous sodium chloride and washed
consecutively with three 100-ml portions of the salt solu-
tion, three 100-ml portions of saturated aqueous sodium
bicarbonate and one 100-ml portion of salt solution. The
sodium bicarbonate washings were combined and extracted with
100 ml of methylene chloride. After the combined organic
layers were dried and filtered, the solution was distilled
until a head temperature of 43° was reached. The residue
was diluted to 100 ml with methylene chloride and vpc anal—
ysis on a 5-ft Carbowax column at 145° showed that 14.6 g

of starting material remained. The only other peaks on the
vpc trace were from two oxidation products with almost

identical retention times, which amounted to about 3.2 g.

53
These products were not fully characterized but showed
the following Spectral characteristics. A mixture of the
two compounds (vpc collected) showed strong carbonyl ab-
sorption in the ir at 1722 cm_1 and strong absorption be-
tween 1150 and 1250 cm-1, which could be associated with
an ester group. The peak with longest retention time was
collected and showed in its nmr Spectrum a quartet (J =
7 cps) at T 5.03 (1 H), a Singlet at T 7.90 (6 H) and a
doublet (J = 7 cps) at T 8.67 (3 H) which could possibly
be assigned to the structure shown on page 22 (’62) . The
other product showed in its nmr Spectrum a singlet at T
1.96 (1 H), a quartet (J = 7 cps) at T 4.89 (1 H), singlet
at T 7.87 (3 H) and a doublet (J = 7 cps) at T 8.62 (3 H).
This spectrum fits well for the formate ester of 3-hydroxy-
2-butanone (62). This compound could not be obtained pure
enough for a reliable mass Spectrum but the former compound
could and Showed a parent peak at mass 139, base peak at

43 and one prominent peak at 87. These peaks can’be readily

O
i. .’

explained with the assigned structure.

13. 3,4-Dimethyl-3—penten-2-one (64). Procedure B2

 

was followed in oxidizing this alkene (3.4 g, 0.028 mole)
using one half the amount of reagents and Solvent. Only
starting material and one product were detected by vpc anal-
ysis on a 10-ft FFAP column at 160°. The product had a
carbonyl band in its ir Spectrum at 1700 cm-1 and an nmr

Spectrum consisting of two singlets of equal intensity at

54

T 7.93 and T 8.70, which was consistent with the structure

3,3-dimethyl-2,4—pentanedione (nmr Spectrum, Figure 9).

In addition, the mp (21-220) was the same as the reported

value (45) for this compound. The yield of diketone was

61% (81% based on consumed starting material).

14. 4-Methyl-_d_3 -3-methyl-3-penten-2-one-1 , 1 , 1 , 5 , 5 , 5+_c_1_6

This compound was oxidized in the same manner as the

(2,1,).
The vpc-collected

non-deuterated compound described above.

 

product showed one Singlet in its nmr spectrum, at 'r 7.93
and a much smaller peak at 1' 8.70 (Figure 10) and was there-

fore assigned the structure 3,3-dimethyl-d6—2,4-pentanedione-

1,1,1-<_i_3 with 66% deuterium in the 3-position.

15. 2-Cyclopentylidenecyclopentanone (16). This alkene

(9.0 g, 0.060 mole) was oxidized by procedure B2. Only

starting material and one product were detected by vpc an-

alysis on a 10—ft FFAP column at 250°. The ir and nmr

spectra (Figures 22 and 11, respectively) of the product

were consistent with the structure spiro[4.5]decane-6,10-

dione. The'yield of diketone was 27% (43% based on consumed

starting material).
Calcd for C10H1402:

C, 72.12; H, 8.60.

Anal. C, 72.26; H, 8.49.

Found:

Further structure proof was obtained by converting 2 g.
of distilled product (bpo.9 100-1030, mp 11.5-12.50) to 5-
cyclopentyl-S-oxopentanoic acid by reflux. for 30 min with

12 m1 of 6N sodium hydroxide in 50 ml of ethanol. The

7 55
product, after dilution with 60 ml of water and acidifica-

tion with concentrated hydrochloric acid, was extracted

with two 30-ml portions of ether, dried over anhydrous mag-
nesium sulfate, filtered, and the solvent removed with a
The semicarbazone was then prepared

rotary evaporator .
The residue was

directly from the residue as follows.
Water was then added until

taken up in 20 ml of ethanol.
the solution became cloudy, followed by clarification with
a few drops of ethanol. Semicarbazide hydrochloride (2 g)

and sodium acetate (3 g) was then added, dissolved, and the

solution heated for a few seconds in a boiling water bath.
After being cooled to room temperature, the solution was

placed in an ice bath and the Sides of the flask were scratched

with a glass rod until white crystals formed (1.3 g, 45%
overall, mp 176-1780). Repeated crystallization from ethanol
gave a constant mp of 180—182° (lit. 46, 181-1820).

C, 54.76; H, 7.94; N, 17.41.

Calcd for C11H19N3033

Anal .
C, 54.62; H, 7.96; N,

Found: 17.34.

16. 3-Chloro-2-methyl-2-butene (’82.). Oxidation of

7.04 g (0.060 mole) of 90% pure alkene by procedure B2 gave

one major product (vpc analysis on a 10-ft FFAP column at

130°) assigned the structure 3-chloro—3-methyl-2-butanone
The ir Spectrum showed a carbonyl

on the following evidence.
and the nmr spectrum (Figure 12) consisted

band at 1712 cm.-1
The nmr

of two singlets at T 7.68 (3 H) and T 8.33 (6 H).

spectrum compared favorably with the published Spectrum

(lit. 47, singlets at T 7.65 and T 8.31).

56
17. 2-Chloro-3-methyl-2-butene-1,1,1113 (8,6,) . Pro-
cedure 32 was followed, as for the non-deuterated compound.

The vpc-collected product had an nmr spectrum (Figure 13)
consisting of a singlet at 'r 8.36 and a small absorption
The product was therefore assigned the structure

at T 7.68.
3-chloro-3-methyl-2—butanone-1,1,1-d3 with the one position

about 80% labeled with deuterium.

18. 3-Bromo-2-methyl-2-butene (’94,). 3—Bromo-2-

methyl-Z-butene [(48) , 8.9 g, 0.060 mole] was oxidized by

procedure B2. One major product was detected by Vpc analy-

sis on a 10-ft FFAP column at 110° and was assigned the
structure 3-bromo-3-methy1-2-butanone on the following evi-

The nmr Spectrum consisted of two singlets at T

dence.
(6 H) and the ir Spectrum Showed a

7.63 (3 H) and T 8.18
therefore,

carbonyl band at 1710 cm"1 Both spectra were,

very similar to the analogous 3-chloro—3-methyl-2-butanone.

The yield of ketone was 79%.

19. 2,3-Dibromo-2-butene (93). This alkene (6.4 g,

0.030 mole, K 8c K Laboratories) was oxidized by procedure
BZ using half the amount of reagents and solvent. After

workup the residue was distilled through a Short path dis-
tillation apparatus (bp 60-130°, 5.3 g; pot residue 0.5 g).

Vpc analysis on a 10-ft SE-30 column at 125° showed one

major product, comprising 80% of the total peak areas and

starting material comprising 14%. This product (61% yield;

69% based on unrecovered starting material) was assigned

57
‘meSUMcture 3,3—dibromo-2-butanone on the following evi-
dence. The ir spectrum (liquid film) showed a carbonyl band

and the nmr spectrum consisted of two singlets

at 1715 cm"-1
T 7.04

ofemml intensity at T 7.33 and T 7.50 (lit. 49,

am 77222). This product was also independently synthe-

sflmdlw brominating 2—butanone in the presence of phos-

gmonxsfollowing the procedure which Schotte (50) used to

Inomnmte 3-pentanone. Vpc analysis showed three products.
lme flust product to appear on the vpc Showed, in its nmr

‘ 7 cps) at T 5.65 (1 H), a singlet

spectrum, a quartet (J -

at'r7270 (3 H) and a doublet (J ‘ 7 cps) at T 8.32 (3 H). ,

lflds compound was assigned the structure 3-bromo-2-butanone

(lit. 51, T 5.40, T 7.74, and 1 8.40).

The second product had a Vpc retention time,
oxidation product.

and nmr

and ir spectra identical with those of the

Calcd for C4H6Br20: C, 20.90; H, 2.64; Br, 69.51.

Anal.
C, 21.01; H, 2.76; Br, 69.66.

Found:
The third product showed in its nmr spectrum a quartet

'7 cps) at T 5.10 (1 H), a singlet at T 5.65 (2 H) and

(J:
(lit. 51, T 4.88, T

7 cps) at T 8.22 (3 H)

a doublet (I
5.40 and T 8.00) . The difference in chemical shifts between

these compounds and the reported values is probably due to

the fact that my nmr spectra were obtained from samples

dissolved in carbon tetrachloride, containing TMS, whereas

the reported Spectra were of neat samples, using an external

sample of TMS in carbon tetrachloride as the reference.

 

SUMMARY

1. A.number of alkenes were oxidized to ketones in one

st>by peroxytrifluoroacetic acid - boron fluoride.

2. Eflectron-withdrawing substituents or the absence of

electron-donating groups about the double bond tended to
JINEI the yield of ketone or stop the oxidation altogether.

3. Migratory aptitudes were found to parallel those ob-

served in other Wagner-Meerwein rearrangments.

4. The products were shown to be the result of hydrogen,

methyl, acyl, chlorine, or bromine migration, as well as

ring contraction and eXpansion.

5. The products 3-bromo-3-methyl-2-butanone (92) and 3,3-

dibromo-2-butanone (95) from the oxidation of 3-bromo-2-
methyl-2-butene (24) and 2,3-dibromo-2-butene (93), respec-
tively, are believed to be the result of the first re-

ported examples of bromine migration in a pinacol-type

rearrangement.

6. The oxidation of 2-cyclopentylidenecyclopentanone (16)

:nesulted.in the first reported synthesis (43% yield) of

spiro[4 .5] decane-6, 10-dione (2,8) .

58

PART II

THE ACID-CATALYZED REARRANGEMENT OF

OCTAMETHYL-B-OXATETRACYCLO[4.3.0.02I5.07r9]NON-3-ENE

59

INTRODUCTION

The chemistry and photochemistry of hexaalkyl-2,4-
cyclohexadienones has been extensively investigated in this
laboratory (13,52). We believed that a similar study of
the next higher, fully methylated, homolog, octamethyl-2,4,6-
cyclooctatrienone (92) might be of interest.

An attempted synthesis of this compound, or one of its
valence tautomers, by the oxidation of octamethyljgygrtri-
cyclo[4.2.0.02'5]octa-3,7-diene 22'with peroxytrifluoro-
acetic acid proved unsuccessful. A number of products were
obtained, however. One of these, isolated in 6% yield,

was assigned the structure 100.

CH 0 CH3

 

 

 

 

 

 

 

 

 

 

CH CH3
i
I
C 3 3 CH3 CH3
100
CH3 CH3 CH3 CH3CH3
CH3 0 <9
CHb or
CH3
CH3 /’ .
H3 CH3 c1.3 cH3 H3
3 CH3 )
2§.

On searching the literature, it was discovered that
Maier (53) had prepared what was undoubtedly the same com-
rxmind by the epoxidation of 92, followed by acid catalyzed

60

61
rearrangement, but his structure assignment, 103, was dif-

ferent from ours.

 

 

 

 

CH3
CH
CH 3 CH3
¢C03H x H2501 , IO 1
99 / acetone _\

CH3 CH3 CH3

CH3 CH3

103

 

Part II of this thesis is concerned with the structure

determination and Chemistry of the oxidation product of 92,

RESULTS AND DISCUSSION

xx. The Acid-catalyzed Rearrangement of Octamethyl-S-oxa-

tetracyclo[4.3.0.02.5.07I9]non-3-ene (102)

1. Nuclear_Magnetic Resonance and Infrared Spectra.
The nmr and ir Spectra of the product from the peroxytri-
fluoroacetic acid oxidation of 92 and the acid-catalyzed
rearrangement of 422 were identical in every respect.
Maier's assignment (53) of structure 422 was based on the
following evidence. The ir Spectrum Showed a Sharp band
at 1695 cm.1 assigned to the disubstituted double bond of

the four-membered ring. The nmr spectrum Showed four

 

CH3
CH3 CH3 CH3
l 0 l
CH3 CH
CH3 3
H3
103

equally intense peaks; quartets (J - 1 cps) at T 8.46 and

1 8.54 assigned to the methyl groups at the double bonds,
and two sharp singlets at r 8.93 and T 8.98 assigned to the
remaining methyl groups. Since the molecule has a plane
cfifsymmetry (from the center of the double bonds through
oxygen), all that Should be seen in the nmr are four equally
intense singlets. To explain the splitting between the

62

63
olefinic methyl groups, Maier claims that this is due to
the special geometry of the system with a strained 1,4-
bridged Six-membered ring. To support this claim, Maier
points out that Criegee (54) found Similar results with the

endo peroxide 105. But, upon examination of Criegee's

 

 

 

CH3
CH (ii/CH3
lo/ I
CH3 CH3 CH3
H3 122

article, we found that he states that the nmr spectrum con-
sisted of four slightly split signals of equal intensity.
This could be explained by twisting of the molecule, due

to strain (causing all the expected equivalent methyls to
be in slightly different environments), or by the possi-
bility that the peroxide bridge may not be linear, which
would also make the environment of each methyl Slightly
different. In addition, it is possible that Criegee ob-
tained a mixture of peroxides (4922 and 4222). These iso-
mers might have almost identical nmr Spectra and the simi-
larity of their chemical shifts would also account for the
slight Splitting of the singlets observed by Criegee. It
should also be noted that the peroxide has a different num-

ber of atoms in its ring system from the number in 103.

64

CH3

 

 

CH3 CH3
H3
CH3 CH3
CH3
105a 105b

 

In any case, if the non-equivalence of the olefinic

methyls in 103 was caused by strain, then the other methyls

would be expected to have different environments also, and

appear as two slightly Split singlets. In fact, these

signals are Sharp.

The structural assignment, 100, made in this thesis
does not require any such dubious rationalization of the

spectral evidence. (This structure was apparently not

—1
considered by Maier.) The ir band at 1695 cm would a-

rise from the two disubstituted double bonds in the four-

membered.rings (Figure 23). The olefinic methyls would not

and therefore would be expected to couple,

be equivalent,
Each

thus appearing as two quartets in the nmr spectrum.

(pair'cxf aliphatic methyls (cc'; dd') should appear as a

sharp Singlet (Figure 15).
{Ehe actual geometry, syn (100a) or anti (100b) could

not be deduced from these data.

65

 

3?
II
R.
O‘
H
CZ
0
II
n-
0.;
ll
0:
U‘
s,
n)
x
U‘
‘tk
O
‘H\
3:

100a 100b

(For clarity, the methyl roups will, from now on,
be represented by lines.)

Further reactions were carried out on 100 to test the
validity of the structural assignment and to more clearly

determine its geometry.

2. Hydrogenation. Compound 422 smoothly took up one
mole of hydrogen after being Shaken overnight in a Parr
hydrogenator at 4 atm using 5% Pd/C in ethanol with Slight
heating. The compound would not take up an additional mole
of hydrogen even after 72 hours. This fact would tend to

support assignment of the syn structure 100a, where addition

66
of a second mole of hydrogen from the least hindered side
would be extremely unfavorable, due to the interaction of

the endo methyls. This problem would be less serious in

H
100a , 2

w

 

 

106a 107a

 

 

tthe case of the anti isomer, 100b, going to 106b, and

finally to 108 with an additional mole of hydrogen.

H
H
100b 7 —-——-) H
H .

106b 108

m

 

Examination of the nmr spectrum of 422 (Figure 16)
showed a multiplet for the methine protons (T 7.6 - 8.4),
two doublets for the methyl groups adjacent to the protons
(T 8.91 and 9.21, J = 7 cps), two quartets for the olefinic
methyls (T 8.45 and 8.57, J = 1 cps) and four singlets for
the remaining methyl groups, (1 8.80, 8.92, 9.00 and 9.02).
This spectrum is readily rationalized by structure 4222 or
4222, On the other hand, according to Maier's structure,

103, the expected hydrogenation product 109 or 110 should

67
Show only two singlets and a doublet for the methyl groups
other than the olefinic ones, and a quartet for the methine
protons. Once again, a special rationalization would be
required to explain the two quartets, rather than one Sing-

let for the vinyl methyls.

Ii—m . ‘ H

H _
103 ___£L_:> __J O or 0
"VV H

 

 

 

 

 

 

 

 

 

222, 1,12.
The infrared spectrum of the hydrogenation product is,
perhaps, even more revealing. The band at 1695 cm.1 (Fig-
ure 24) is still present, though qualitatively diminished
in size relative to the other bands. If 422 were the pro-
duct, then the 1695 cgm-1 band would be expected, but of
approximately the same intensity as in 422” the starting
material. On the other hand, if 442’were the product, then
this band Should be completely absent. For our postulated

product, 106, one cyclobutene ring would remain and a dimin-

ished band at 1695 cm-1 would be expected.

3. gromination. Compound 422 readily took up two
moles of bromine in carbon tetrachloride, but the complex
mixture of products obtained indicated that rearrangement
was taking place. This assertion is further supported by
the fact that the hydrogenation product, 422/ was inert

under the same conditions. This would also favor the syn

68
structure 106a (and thus 100a), Since addition of bromine
to 106a would result in the formation of the highly strained

molecule 111a. Such strain would not be present were bromine

 

 

 

 

 

Br2 \
7' H
B ‘3(
r) (
106a 111a
to add to the anti isomer (106b > 111b).
H
H
Br
0 2 > Br
Br
106b 111b

Because of the complexity of the bromination reaction,
no data could be obtained to further clarify the structure

of 100.

4. Hydroxylation. Maier (53) had reported that ESE
(or 422) reacted with one mole of osmium tetraoxide upon
remaining overnight in an ether solution containing pyridine,
and with two moles (after isolation of the initial glycol)
after four weeks. The fact that two moles were taken up

might Speak in favor of the anti structure 100b, but the

69
slowness of the addition of the second mole might indicate
rearrangment prior to addition; alternatively, very Slow
addition without rearrangement may have occurred to produce

the highly strained molecule 113.

 

 

 

 

 

 

Via 1 .0504
100a, 2. OH“)
HO HO (3‘ OH
HO HO )( H
112a 113
1. OSO
100 _4
“““’ 2. OH
OH OH
OH OH
via 1. 0504
‘ 3 2. OH 4
100b Ho
HO
112b 114

Since Maier did not report any Spectral data for either
compound, it was decided to repeat his work, at least to
the diol stage. Following his procedure (53), one mole of
osmium tetraoxide was added to 422; The nmr and ir spectra
of the product were compatible with structure 4422 or 4422’
but not with structures 442 or 442) if Maier's compound had
the structure (422) which he assigned to it.

The ir Spectrum (Figure 25) showed strong hydroxyl

bands at 3405, 3560 and 3720 cm-1, as well as a weak band

70

F—-"—-OH

1. OSO4AHO
103 's _ O or I O '
" " " 2 0 OH IHO ' r—L\ n—Jr— OH

 

 

 

 

 

 

115 116

rwv

 

at 1695 cm-1. The nmr Spectrum (Figure 17) showed two
broad singlets (T 6.86 and T 7.40) for the hydroxyl pro-
tons (these signals disappeared upon adding a few drops of
deuterium oxide), two quartets (T 8.43 and T 8.53, J = 1
cps) for the olefinic methyl groups and five singlets

[T 8.63, 8.77, 8.97, 9.00 (3 H each) and 1 8.98 (6 H)] for
the remaining methyls. If Maier's structure 422 had been
correct, then the glycol 442 or 442 would be expected to
Show, in its nmr spectrum, three singlets for the quarter-
nary methyls and a singlet for the hydroxyl protons (as
well as for the olefinic methyls). TherefOre, the Spectra
of both the hydroxylation and hydrogenation products provide
further evidence in support of structure 422'and against

structure 103.

5. Attempted Photolysip, If 100 had the syn config-
uration (100a), then it was thought that it might possibly

undergo photolytic intramolecular cyclization to 117.

100a h” 4}

 

117

71
Under a variety of conditions (see Experimental Section),
either no reaction occurred or the solutions turned yel-
low with no volatile products being obtained. These results
do not necessarily rule out structure 4222) Criegee has
reported (55) that he could not prepare octamethylcubane
/

(118) by the photolysis of 92 (these compounds are closely

related to 100a and 117, respectively).

 

 

 

 

1'» “xx
/ V/ / k
22, 11%

Criegee (55) (1962).

6. Mechanism. A discussion of the possible mechanisms

 

which may account for the transformation of 422 to 422 must
be only Speculative at this time, since the exact geometries
of 422 and 422 are not known with certainty. A number of
mechanisms for converting either possible epoxide (4222 or
4222) to either the gyg_(4222) or §££4_(4222) product can
be postulated, depending upon the role of the acid in the
rearrangement. Only a few possibilities are Shown here.

It might be reasonable to assume that the epoxide ring
is first broken to give a stable tertiary carbonium ion
4422'or 4422) which could rearrange by a number of bond

shifts to the syn product 100a. In addition, the endo

 

 

 

 

100a

 

epoxide 4222 could rearrangerig and allyliccation (422)

to either the §y3.(4222) or 3334'(4222) product. This

type of rearrangement is not possible for the 3&3 epoxide
44222). The OH group would always be on the side away from
the positive charge and could not close to the tetrahydro-
furan ring. The possibility also exists that the proton
could add to the double bond of 422 and cause rearrangement
through a 10w oxacyclononatetraene structure 424 (424 would
probably not be aromatic Since it cannot assume a planar

conformation) to either 100a or 100b.

73

 

 

119b 120 100a
rotation
.w

 

 

 

 

 

 

 

 

 

100b

 

 

 

1223—) (”\f

 

 

 

 

 

100a or 100b

74
Structure 424'might also be formed by attack of the
proton on the carbon-carbon bond of the epoxide ring.
This type of ring opening might occur due to the relief

of strain, which would result, in going from the tetra-

cyclic structure 102 to the tricyclic intermediate 122.

or

 

 

 

 

122 H

More work will have to be done in order to determine

which, if any, of these mechanisms is correct.

B. Thermal Rearrangement of Octamethyl—S-oxatetracyclo-

[4.3.0.05I5.o7l91non-3-ene,(102).

 

Maier (53) reported that upon melting, 422 rearranged
to a new isomer, 424; Maier stated that the nmr spectrum
of 424 consisted of four equally intense signals at r 8.28,
1 8.43, T 8.49 and T 8.91. Later in the same article, (in
reference to compound 422) he stated that in all the pre-

viously mentioned compounds the methyl groups gave very

75

 

 

 

 

 

 

 

 

 

/ 0 180° / / ——-«éo
\ L—<
1.9.2. 22::

Maier (53) (1963).

sharp signals in their nmr Spectra. From this, we assume
that the four signals in the nmr spectrum of 424 were sing-
lets. If this is so, then it would appear that the struc-
tural assignment was incorrect. The methyl groups attached
to the double bonds would not be equivalent and the nmr
spectrum should consist of two quartets for these methyls
and two sharp singlets for the remaining methyls (there is
a plane of symmetry going through the oxygen and between
the double bonds).

This thermal rearrangement was repeated. The nmr
spectrum of the crude material showed that it was a mixture
containing 43% starting material and 57% of a product having
four equally intense peaks at T 8.28, T 8.42, T 8.48 and
T 8.91. These values correspond well with Maier's values,
but the peaks at T 8.42 and 1 8.48 were quartets (J pg,

0.8 cps). It is concluded that Maier's structural assign-
ment, 424) was correct, but the splitting that we found was

either missed or incorrectly reported by Maier.

EXPERIMENTAL

A. Preparation of Octamethyl-syn—tricyclo[4.2.0.02 5]octa-

 

3 , 7-diene (22)

 

This compound was most conveniently prepared by fol-
lowing a Slight modification of the procedure of Rosenburg
and Eimutis (56). A cold solution of 2-butyne (12 g, 0.22
mole) in 20 ml of cyclohexane was added, by means of an
ice-jacketed addition funnel equipped with a drying tube
containing drierite, to a magnetically stirred suspension
of 14.8 g (0.11 mole) of anhydrous aluminum chloride in 30
ml of cyclohexane prepared in an erlenmeyer flask cooled
by means of an ice bath. The addition was started when the
aluminum chloride suSpension started to become viscous
(cyclohexane melts at 6.50). The 2—butyne solution was
added dropwise over a period of 40 min during which time
the reaction mixture slowly turned red-brown. After addi-
tion was completed, the reaction mixture was allowed to warm
to room temperature, at which time magnetic stirring be-
came difficult. The solution was then stirred intermittant-
ly for 3 hrs. The reaction mixture was poured onto crushed
ice and the gummy material in the reaction flask was scraped
onto the ice. This mixture was stirred for about 0.5 hr
until all the red-brown complex was destroyed (white solid
formed in the upper layer). The layers were separated and
the organic layer was washed with water. The aqueous layer

76

77
was extracted with cyclohexane and the combined cyclohexane
layers were dried over anhydrous sodium sulfate, filtered,
and the solvent removed with a rotary evaporator. The resi-
due was suspended in a small amount of acetone at 0° and
filtered to give 5.0 g (42% yield) of white solid mp 188-
192° (lit. 57, 198°). The nmr Spectrum showed two singlets
of equal intensity at T 8.60 and T 9.08 (lit. 56, T 8.58

and T 9.07).

B. Prpparation of 1,2,3,4,5,6,7,9-Octamethyl-8-oxatetra-

 

cyclo[4.3.0.02:5.07I9]non-3-ene (422)

 

Octamethyl—Syn-tricyclo[4.2.0.02I5]octa-3,7—diene
(7.1 g, 0.033 mole) was dissolved in 100 ml of benzene in
a 300 ml erlenmeyer flask cooled by means of an ice bath.

The solution was magnetically stirred and 6.5 g (0.033 mole)
of 87% meta-chlorobenzoic acid (Research Organic/Inorganic

Chemical Co.) dissolved in 130 ml of benzene was added

dropwise. The temperature of the reaction mixture was kept

below 12°. A white solid (mgggfchlorobenzoic acid) formed
during the addition. After addition was completed, the
solution was stirred at room temperature for 1.5 hrs and
then washed with three 50-ml portions of 5% aqueous sodium
hydroxide and two 50-ml portions of water. After the pro-
duct was dried over anhydrous magnesium sulfate and filtered,
the solvent was removed with a rotary evaporator. The
residue (7.3 g, 96% yield) was a white solid, mp 180—183°

(lit. 53, 180°).

78
This product had an nmr spectrum consisting of four
singlets of equal intensity at T 8.38, T 8.78, T 9.13 and

T 9.22 (lit. 53, T 8.38, T 8.79, T 9.11 and T 9.21).

C. Preparation of 1,2,3,4,5,6,7,9-Octamethyl-8-oxatri-

cyclo[4.3.0.07I9]nona-2,4-diene (104)

 

The previously prepared epoxide (55 mg) was sealed in
a vial and placed in an oil bath maintained at 185°. When
the compound had just melted, the vial was removed from the
bath and allowed to cool. The vial was then opened and the
crude product taken up in carbon tetrachloride. The nmr
spectrum of this solution (Figure 14) showed it to be a
mixture of starting material (43%) and the reported product

[ (53), 57%] .

D. Prgparation of 1,3,4,5,6,7,8,9-Octamethyl-2—oxatri-

cyclo[5.2.0.03:6]nona-4,8-diene (100)

 

1. Octamethyl-gygftricyclo[4.2.0.02I5]octa-3,7-diene
(3.4 g, 0.015 mole) was oxidized following procedure B1 in
the first part of this thesis except that 98% hydrogen per-
oxide (0.4 ml, 0.017 mole) was substituted for 90% hydrogen
peroxide, no boron fluoride was used and cooling was car-
ried out with an acetone-carbon tetrachloride-dry ice bath
adjusted so that the temperature of the alkene solution
(200 ml of methylene chloride) was at -25°. After workup,

vpc analysis on a 5-ft SE-30 column at 170° showed a 6%

79
yield of the desired product (mp 67-68.5°, Vpc collected)
which showed the peak of shortest retention time. A compli-
cated set of other peaks was also observed at longer reten-
tion times. The nmr and ir spectra of the product are

shown in Figures 15 and 23.

2. This procedure, which is the method of choice, is
a modification of the procedure of Maier (53). The epoxide
(7.3 9) prepared by procedure B was dissolved in 200 ml of
acetone to which was then added 10 ml of 2N sulfuric acid.
The solution was stirred for 10 hrs, then diluted with 1
liter of water. The solution was extracted with four 75-ml
portions of pentane, and the combined pentane extracts were
washed with two 60-ml portions of saturated aqueous sodium
bicarbonate and one 50-ml portion of water. After the pro-
duct was dried over anhydrous magnesium sulfate and filtered,
the solvent was removed with a rotary evaporator. The resi-
due was chromatographed through 180 g of alumina using hex-
ane as the eluant. The solid collected (3.9 g, 53%, mp
56—61°) was off—white (the solid formed upon standing or
scratching from a yellow oil) but showed an nmr Spectrum
identical to the vpc-collected material prepared by procedure

D1.

E. Reactions of 1,3,4,5,6,7,8,9-Octamethyl-2-oxatricyclo—

[5.2.0.03'6]nona-4,8-diene (422)

 

1. Hydrogenation. To 200 mg of 5% Pd/C in a pressure

 

80

bottle, was added 2.0 g of alkene dissolved in 25 ml of
absolute ethanol, followed by four drops of concentrated
hydrochloric acid. The bottle was then attached to a Paar
hydrogenator and shaken overnight under 4 atm of hydrogen,
with heating (external temperature 35°). Vpc analysis of
the crude product (after filtering and evaporating off the
solvent) on a 5-ft SE-30 column at 174° showed a single peak
with a longer retention time than that of the starting
material. The nmr Spectrum (Figure 16) of the crude and
vpc collected product were essentially the same; the pro-
duct was assigned the structure 1,3,4,5,6,7,8,9-octamethyl-
2-oxatricyclo[5.2.0.03:6]non-4-ene (ir spectrum, Figure 24).

£9334. Calcd for C16H260: C, 81.99; H, 6.83.

Found: C, 81.97; H, 6.82.

Hydrogenation for an additional 48 hrs showed no further
change. In addition, this product would not react with
bromine following procedure E4 for the unhydrogenated com-

pound.

2. Hydroxylation. Following the procedure of Maier

 

(53), 1.0 9 (0.0043 mole) of the alkene dissolved in 10 ml
of anhydrous ether was added to a solution of osmium tetra-
oxide (1 g, 0.0039 mole) and pyridine (0.63 ml, 0.0079 mole)
in 10 ml of anhydrous ether. The solution was allowed to
stand overnight and then filtered to isolate the brown solid
(2.0 g, 80%).

The solid brown osmate-pyridine complex was dissolved

in 40 ml of methylene chloride and placed in a ZOO-ml flask.

81

To the solution was added 2 g of mannitol and 0.6 g of
potassium hydroxide, dissolved in 50 ml of water. The mix-
ture was Shaken for 2 days, then filtered through Celite
to break up the resulting emulsion. The layers were separ-
ated and the aqueous layer was extracted with four 25-ml
portions of methylene chloride. The combined organic layers
were dried over anhydrous magnesium sulfate, filtered and
the solvent removed with a rotary evaporator. The result-
ing yellow oil was sublimed (100°, 2 mm) and the slightly
brown solid which resulted (0.3 g, 27%) was recrystallized
from pentane (mp 116-120°, lit. 53, 128°), resublimed and
recrystallized once again from pentane (mp 119—122°, white
solid). The nmr and ir spectra are shown in Figures 17 and
25 respectively and the glycol was assigned the structure
1,3,4,5,6,7,8,9-octamethyl-2-oxatricyclo[5.2.0.03I°]non-8--,
ene—4,5-diol.

5224, Calcd for C16H2603: C, 72.14; H, 9.84.

Found: C, 72.03; H, 9.75.

3. Attempted Photolysis. No reaction was observed
when 34 mg of alkene dissolved in 6 ml of acetone was ir-
radiated using a 450W Hanovia lamp through Pyrex for 2.5
hrs, or when the alkene dissolved in 3 ml of benzene con-
taining 93. 35 mg of acetophenone was irradiated for 4.5 hrs.

When 85 mg of alkene dissolved in 2 ml of acetone in
an all quartz apparatus was irradiated for 9.5 hrs, the

solution turned yellow. Vpc analysis on a 5-ft SE—30 column

82
at 220° showed a decrease in peak area of the starting
material, but no volatile products were detected. When 6
ml of benzene was substituted for the acetone, the same
results were found except that a yellow polymeric material

collected on the walls of the quartz tube.

4. Bromination. The alkene (0.1 g, 4.3 x 10-4 mole)

 

was dissolved in 1 ml of carbon tetrachloride. The solution
was cooled in an ice bath and a 5% bromine solution in car-
bon tetrachloride was added dropwise. The solution immedi-
ately turned pale yellow and remained so until 0.9 ml of
bromine solution (8.3 x 10-4 mole) had been added. At this
point the solution turned orange. An additional 0.3 ml of
bromine solution was added and the solvent was removed

with a rotary evaporator. Vpc analysis on a 5-ft SE-30
column at 175° showed no starting material but instead a
very broad complex of poorly resolved peaks of longer re—
tention time. This crude material was refluxed with 3 ml
of methanol and 75 mg of potassium hydroxide for 10 min
(white solid formed). Vpc analysis of the product at 105°
showed at least four peaks. No further analysis was car-

ried out.

S UMMARY

1. The product of the acid-catalyzed rearrangement of

octamethyl-B-oxatetracyclo[4.3.0.0215.07'9]non—3-ene (422)
was shown to be octamethyl-Z-oxatricyclo[5.2.0.03I°]nona-
4,8-diene (422) and not the previously reported structure,

103.

2. The assignment of structure 1001was based on its nmr
and ir spectra, and the spectra of its hydrogenation product

106 and its hydroxylation product 112.

3. The rearrangement product probably has the syn config-
uration (100a). This assignment was based on the high re-

activity of only one of its two double bonds.

4. The thermal rearrangement product of 422 was shown to
be octamethyl-S-oxatricyclo[4.3.0.07'9]nona-2,4-diene

(424) based on its nmr spectrum. The structural assign-
ment previously reported in the literature was correct, but

the reported nmr spectrum was incorrect in some details.

83

S PECTRA

84

NUCLEAR MAGNETIC RESONANCE SPECTRA

85

86

 

 

 

 

 

.37
l l 811 J l

'—'

7'5— 3‘0 V 805' 700 ?’§ I

  

Figure 1. Nmr spectrum of 1,2—dimethylcyclohexene (9,6).

87

.deV msmudnlalamcmnmlmlawzuoe

In tam dmwv osousnlmlawsonmlm1H>£uoEIN mo onsuxfle hmumo m mo Eduuommm HEZ .N onsmflm

WW.

2.
3‘ ”Ca 3 us . q 3 use ox: .3 9% he
. _ a __ 1. _ 1 _

...)..i: _ .... ...... i. s .1. A.
€335: . DE;
_ r. f; C m c .
__ . I

, m

1
_
.

i _._.
. I.) )l 2
. «\l 1x): ii

I

- ...— ro—
_. ..
.t —- 9 -~ .... ---. -..—...
...—— ___

_
i
‘

\I\

 

 

... .258 _

 

M
d

m i
\L4 .n

Y

 

mswd.

88

5

.3» Cam )mxw mo musuxfi: omuow no mo Enuuoomm .22 .m wHDmHm

......» ea Ms 0.“ h.» s.» .3 av. 13¢ o6 M.» Rm . .3.
_ A _ . .. a 1|. _ _ q a _ _ H

 

 

 

_i
if 22 w
mi... mvnm ,
m :__: mans W
H _
m M:
g
i.
mam
é:
W. ..
¢ H

89

.355
wCOINICODCOQIMlahfiuwfiflvlv.m

 

Uwumumusmp mo Ednuommw HEZ .m oudmflm
Q1 .3 3. u o... u.»
a J a

_ _
111,332;
.

d
_
r

.,wdv wCOINIGmqumImI

thumeflplv.m mo Eduuoomm HEZ .v wusmflm

.4».
a

.3 es
_ _(

bk

a)
fill

11 ___ “P.

 

 

L

90

.Ame wamusnlmlamnumEINIOHOHno .dmwv wamusnlmlamsumfilm

 

 

    

 

 

In Umumumpswv mo Esuuommm MEZ .b mudmflm Ionoanolm mo Esuuowmm HEZ .m musmwm
ko\ .ms 9» \%a 0% us k_\ kw as Md 0% .%5
fl fl fl 4 q _ .1 w l
L. 93:}: a. iii
. .¢c H " ::-;;:;x‘xl ‘
. Ju ...... L ,
:1, E. m

 

 

91

 

 

 

 

- 8.41
J l J l . L 1 ; _
7.5” 8,0 3'5 7-0 ‘35 My

Figure 8. Nmr spectrum of l-methylcyclohexyl
trifluoroacetate (ii).

 

92

.GNMV m:0fl©mcmucmml¢.«lawsuwfiflw .dmwv macavwcmgcmmuv.ml
I .m vmumumusmo mo Enuuommm HEZ .OH musmflm Hmsumaflolm~m mo Ednpommm “82 .m musmflm
\o ...; 0a b4“ ca .3. .3 3. r bfi. om hm”
— _ _ . _ _ _ ‘ a _

 

 

 

 

 

 

 

 

 

 

 

 

93

an;

EEE I.rL .1.

can. 5... ‘
an 8

_ . an;

    

 

 

 

 

 

 

Nmr spéctrum of spiro[4.5]decane-6,10-dione

(78

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 11.

m)

94

.‘llr lbs ,I' "3‘

.dmmv mcocmusnlmlamnumelmlonoHno
In Uwumuwusmw m0 Esuuowmm HEZ

i: .3 3 .3 _ 0.»

.mH wudmflm
mm

 

- n,

   

14

.h

  

.dwwv mcocmusnlmlamnume
Imlouoanolm mo Esuuuwmm H82

.3 3 .3

 

 

        

.NH mudmflm
A. ..2
q ,

)5?

.chmv aceflwuv.mumcocHm.bo.o.m.wa

\/ ‘I )x

IOHuwofl.mm%OImIH%£pQEmuuo 12m ANOHV uculmlcocflm~ho.msno.o.m.va

 

leachumu;34mm0ImIHm£ucSm;Do mo musuxfle unumv M MD Esuuommm uEz .vH wusmflm
T3 :s 2” .m "r5 m; 5mm «8 m3
= _ d. _ d

_

—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VOH

u
m
d

u
<

NOH

 

 

 

 

 

 

 

 

---

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l l J ' J l
8.0 8,6“ 9.0 $6 IOT

Nmr Spectrum of octamethyl—Z-oxatricyclo—

' 15. -
Flgure [5.2.0.03:6]nona—4,8-d1ene (122)-

97

x. Cu.»

m;

03 «S. .23 one

 

 

.dd‘w mo AWWMV HUDUOHQ coaumcmmanwwn 93 mo Esnuommm HEZ .mH musmflm

35m .32. no.»

7fl_4 __ L.
illHllllnl

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

III"! III' I'll | ' I I

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l L J I
no to ma m1

Figure 17. Nmr spectrum of the hydroxylation product
(112) of 100.

INFRARED SPECTRA

99

100

.Amlv aflamuUOIoa.m< mo AvHUUv Enuuowmm UmnmumcH .mH musmflm

tunnctazriiii...

-—..._.-_._

 

—— ..‘gw.-_.-

 

-4»- —-—O—‘——-M .
.

 

 

 

 

 

M

 

 

 

 

 

101

 

“3.4m. . L . _n_._...L,..s..wa.f

up... ,. . I ,.
ilk”: . est»;

)2

.Aomv mcmxmnoaumoamsumfiflplm.a mo AVHUUV Esnuommm pmumumaH .ma mudmflm

tgsnuzco>asr

..Jx .. -L ;.L _)L... 3):)». L._ I

R.

 

 

 

 

 

102

I\v

nun O.- .‘u.

.5.

 

.35
mcocmucmmoHumumcmpflamucmmoHomolm mo AEHHM Ufldvflav Esuuommm UmumnmcH

....L y ..., .n . .. n.

.ON mudmflm

tumwzzce>a>r

11

 

 

‘ L‘Itlllv"

.I\ 'll'lltl‘ III.
'II. I ' i II ‘II III !I I I All: 1 114‘. Ill 14
f L 1.!

103

 

.Aamv mumumUMOHOSHMHHu.EkmonomoamLumEIH mo Avaouv Eduuommm UmumumcH .HN musmflm

 

. w:
_:
fi-_w .. n3—
." . . _
L“ we r 4. .
_fi ” . . _ _
g , A _ ,3...
g _ .,. 1 T _ .t.
o .. .. ,‘ + I”
H _ m r m
, . M x . m
._. r,” w. H
. .. .,
. i p _ W I
_ .__ .. _ . _ . .
. . : _ . y
a ..M .._ _ . ...
«Fa: .. . p. r .L x
2 j a” __ a, _. . g
.... __ .. H __. x u r
L“; _ : h ;
2.x. : A T . h _;
Ho . o No I 0 —~
:. m . _ M. r
_L_. c .. 3 V. t
L .— h . _, ..
t c H , _u
J . H r I.
. f ,
# H a m mi
a _ .L . .. A;
m. m a g . ..
p . ..

'

.ij

'-

“
€-
1
I;
Mil- -—.--
C
,_,
.a’
{,‘k- _
4.-
a
v ‘ ' ' l
k“ «....--__ -

pf3§.£:\\}>p\ra§.fip\r\..

r H . ...?.r\#z.\.\1!wfi .\ r

m _ -
.. \ ... {It I .
) J \ $. il‘ 3"\ l5

(‘1, k

 

I’D ‘1'.
111
III"

104

 

.. 20L I? v garlfiflnu. firs»,

.AWLMV mcoflploaélmcmomcflm.193mm no 2403 Esuuommm UmumumcH .NN musmflm
k... :3... . 7..

LR§
5. c

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

105

.Aoofiv mcmflcnm.w
IMGOCHo.no.0.N.mHanmofluumeImlamnuwfimuuo mo AVHUUV Ennuommm vmumuwcH .mm wusmflm

00v—

 

00m 000. 00m_ 000—

T.

' . . . . 0

com. oooN Com m . Doom” Down 009.

o M

 
   
  
 

        
 

    

,1‘

   
  

.
. .31

-.w.

 

.
.
.

Mp

L
“
1

L. )lII ‘Ll|+|0||.. 1

 

b
JLWH -
'A

------~.c.‘ . .

om

 

 

LLLLLL

l

.— v Fr—H*mr-—d—O-"——,r

 

 

3
l
L
J-
7
<3

 

.......

 

- It‘ll! ’ ..III’L l...l

LLLLL

 

'ull’lnlviL‘Ltl'IOI 'I

I
é
o

 

O
O
8

.....

"/) EDNVUIWSNVHL

I
\

l.
a
V
W
n 1.
V
N
3
3
%

o . v 4 L. L a

 

 

:2

 

 

.
9
0
L
.
\
—

09 -g-.._,L.“ uL LerL m ML M. 4% ”Ans. . M w w.” W,m WW. -u M .
0.2 , 0.: 0.: 0.9 mzoaui 9m ox 0.0 L. o.n o... 2” o5 ma

 

 

 

 

 

 

106

    
 

.OOH mo “moHv uosuoum qofiumcwmouwan wflu mo Avaoov Esnuowmm pmumumCH .vN musmflm

8o 08— oofi 8: 000— 89 8mm Doom

'1

1..

(fié)___mvmws~va1

n.

mzoxui L J

   

(“H aanqusuvm

107

cu

0V

8—

0.0—

   

.OOH mo ANHHV uoSUOHm coaumHhxouchn ms“ mo Avauov Eduuowmm UmumnmcH

.mN musmflm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

com 000— 009 003 000— com. oowum oonw 000m Down 009.
h . L _ , _ U L _ u u U . . , . L _ O

a LLLLL LLLLLL _ 2
LL?» L L. LLLLLL LL11 LLL ..
wH . M H L .m . . w
OVNMILI -m u r H “ ovN
MH . . . w . x: ...... ..... M
uT- _ . L M u
w s i_- L w

8 8

...... LLLJL
. 00— W r c r ._ . u L " oo—

LITERATURE CITED

1. For reviews, see D. Swern, Org. Reaction, 1, 378 (1958),
and J. B. Lee and B. C. Uff, Quart. Rev. (London), 21,
479 (1967).

2. D. Swern, J. Am. Chem. Soc., 69, 1692 (1947).

3. W. D. Emmons, A. S. Pagano, and J. P. Freeman, J. Am.
Chem. Soc., 16, 3472 (1954); W. D. Emmons and A. S.
Pagano, ibid.,lll, 89 (1955). -

4. For a review, see R. B. Parker and N. S. Issacs, Chem.‘
Rev., 59, 737 (1959).

5. A. C. Cope, P. A. Trumbull, and E. R. Trumbull, J. Am.
Chem. Soc., 8Q, 2844 (1958).

6. D. J. Reif and H. 0. House, Org. Syn., Coll. Vol. 4,
375 (1963).

7. C. L. Stevens and S. J. Dykstra, J. Am. Chem. Soc.,
‘16, 4402 (1954).

8. E. A. Braude, A. A. Webb and M. U. S. Sultanbawa, J.
Chem. Soc., 3328 (1958).

9. H. 0. House and R. L. Wasson, J. Am. Chem. Soc., 18,
4394 (1956).

10. For example, see H. 0. House and D. J. Reif, J. Am.
Chem. Soc., 11” 6525 (1955).

11. For a reView, see C. J. Collins, Quart. Rev. (London)
12, 357 (1960).

12. H. Hart and R. K. Murray, Jr., J. Org. Chem., 32, 2448
(1967) and referencestherein.

13. H. Hart, P. M. Collins, and A. J. Waring, J. Am. Chem.
Soc.,.éé. 1005 (1966).

14. H. Hart, C. A. Buehler and A. J. Waring, Advances in
Chemistry Series, 51, 1 (1965).

15. H. Hart and L. Lerner, J. Org. Chem., 32” 2669 (1967).
16. W. B. Smith, R. E. Bowman and T. J. Kmet, J. Am. Chem.
Soc., 81, 997 (1959).

108

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

109

R. Filler, B. P. Camara and S. M. Naqvi, J. Am. Chem.
Soc., 81, 658 (1959).

H. Meerwein, Ann., 542, 123 (1939).

S. Nametktorsky, Chem. Ber., 57B, 583 (1924).

C. A. Bunton and M. D. Carr, J. Chem. Soc., 5854 (1963).

E. L. Eliel, "Stereochemistry of Carbon Compounds,"
McGraw-Hill Book Company, Inc., New York, N. Y., 1962,
pp 234-239.

W. Hfickel, R. Danneel, A. Schwartz, and A. Gucke, Ann.,
474, 121 (1929).

C. H. Hassall, Org. Reactions,_g, 73 (1957).

G. B. Payne and P. H. Williams, J. Org. Chem., 24,
284 (1959).

H. 0. House and G. D. Ryerson, J. Am. Chem. Soc., 8;,
979 (1961) and references therein.

H. 0. House and R. L. Wasson, J. Am. Chem. Soc., 18,
4394 (1956).

R. N. McDonald and T. E. Tabor, J. Am. Chem. Soc., §g,
6573 (1967) and references therein. -

For example, see H. L. Goering and L. L. Sims, J. Am.
Chem. Soc., 19, 6270 (1957).

Y. Pocker, "Molecular Rearrangements," P. de Mayo, Ed.,
Interscience Publishers, Inc., New York, N. Y., 1963.

G. A. Olah and J. M. Bollinger, J. Am. Chem. Soc., 89
4744 (1967).

W. D. Emmons, J. Am. Chem. Soc., ZQJ 3468 (1954).

W. P. Campbell and G. C. Harris, J. Am. Chem. Soc.,
9;. 2721 (1941).

G. S. Hammond and T. D. Nevitt, J. Am. Chem. Soc., 16,
4121 (1954).

H. Christol, A. Laurent and M. Mousseron, Bull. Soc.
Chim. France, 2319 (1961).

J. Colonge and K. Mostafari, Bull. Soc. Chim. France,
.Q, 342 (1939).

, ' Gun-i

36.
37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.
49.
50.
51.

52.

53.
54.
55.

56.

57.

110
R. Mayer, Chem. Ber., 89, 1443 (1956).
A. Behal, Ann. Chim. Phys., 15, 282 (1888).

"Sadtler Standard Spectra," Sadtler Research Laboratories
Philadelphia, Pa., 1962.

D. J. Cram and H. Steinberg, J. Am. Chem. Soc., :13,
2753 (1954).

M. Quadrat-i-Khuda and A. K. Ray, J. Indian Chem. Soc.,
.19. 525 (1939).

E. Campai ne and D. R. Maulding, J. Org. Chem., 28,
391 (1963 .

J. W. Huffman and T. W. Bethea, J. Org. Chem., 39”
2956 (1965).

O. Grummitt and A. L. Endrey, J. Am. Chem. Soc., 82,

.3614 (1960).

A. C. Cope, C. M. Hofmann, C. Wyckoff and E. Harden-
bergh, J. Am. Chem. Soc., 63, 3452 (1941).

K. von Auwers, Ann., 415, 169 (1918).

E. W. Barber and L. Lapides, J. Am. Chem. Soc., 19”
2859 (1948).

D. P. Wyman and R. K. Kaufman, J. Org. Chem., 29,
1956 (1964).

E. A. Braude and E. A. Evans, J. Chem. Soc., 3331 (1955).
C. Rappe and R. Kumar, Arkiv Kemi, 23“ 475 (1965).

L. Schotte, Acta Chem. Scand., 5“ 969 (1951).

C. Rappe, Arkiv Kemi, El, 503 (1964).

H. Hart and D. W. Swatton, J. Am. Chem. Soc., §§J 1874
(1967).

G. Maier, Chem. Ber., 26, 2238 (1963).

R. Criegee et.a1., Chem. Ber., 26, 2230 (1963).

R. Criegee, Angew. Chem. Intern. Ed. Engl., l, 519 (1962).

H. M. Rosenberg and E. C. Eimutis, Can. J. Chem., 45,
2263 (1967).

R. Criegee, G. Schrbder, G. Maier and H. Fischer, Chem.
Ber., 93” 1553 (1963).

,-

xEtfinthEEa .... 4. LE: