A STUDY OF FREE RADICAL
1, 2-ACYL MIGRATIONS

hesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
E. JEROME MAAS
1970

 

7r“ “ "—1,;

't LIBRAAY
1 Michigan State
,1 University

fruit-bib

This is to certify that the

thesis entitled

A STUDY OF FREE RADICAL
1,2-ACYL MIGRATIONS

presented by
E. Jerome Maas

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Organic Chemistry

2% fl. (ELM/é

Major professor

 

Date August 25, 1970

 

0-169

   

 
  

V; i ’ IIN‘DIG BY
3; “DAG & SDIS'
'- “r 390K WIDE" INC. .

I! LIBRARY muons
8’]!!le Ila-nu ‘

 
   

  
    

ABSTRACT

A STUDY OF FREE RADICAL
1,2-ACYL MIGRATIONS

By E. Jerome Maas

The incomplete information pertaining to l,2-acyl shifts in free
radicals prompted this investigation of keto radicals III and IV; these
radicals are structually designed to permit l,2-acyl migration thus
generating radicals V and VI respectively. Experiments show that radical
IV undergoes this rearrangement to a much greater extent than radical
III; however, triphenyltin and tributyltin hydrides intercept most of

radical IV before rearrangement.

 

 

OR. 0 OR' I
I I u l Rearrangement I H
RC-CCH C-0-0C(CH ) -> RC-CCH ' t? 'CCH CR

2 3 3 2 2
I I I
R' RI
IR= =CD3,R'=D v
II R = = R‘ ’ CH3 VI
R' 0
III
CHBC- H?CH2CR
Rl
XI X = Br VII R = C03, R' = D IX
XII X = CI VIII R = R' ' CH X

2 E. Jerome Maas

t-Butylperlevulinate-g5 (I) and t-butyl-3,3-dimethylperlevulinate
(II) were decomposed in 0.25 M solutions of triglyme and phenyl ether at
l30-l32° for 5 1/2 hours. Greater than 97% decomposition occurred during
this time and the carbon dioxide evolution ranged from 60-90% of the
theoretical amount.

Decomposition of t-butylperlevulinate-g5 (I) afforded 45% (yield
based on mmoles of perester consumed) of a mixture of deuterated 2-butanones
(VII and IX) in triglyme and a 42% mixture in phenyl ether; however, the
rearranged ketone (IX) constituted no more than 10% of these mixtures.

Decomposition of t-butyl-3,3-dimethylperlevulinate (II) afforded a
23% yield of a mixture of unrearranged ketone [3,3-dimethyl-2-butanone
(VIII)] and rearranged ketone [4-methyl-2-pentanone (X)] in triglyme and
an ll% yield in phenyl ether; the ratio of rearranged ketone (X) to
unrearranged ketone (VIII) was l9 in both solvents. Reaction of radical
(IV) with the t-butoxy radical afforded an ll% yield (based on mmoles of
perester consumed) of 4-t-butoxy-3,3-dimethyl-2-butanone (XIII) in both
solvents.

0 CH

III3

CH3-C-CIICH2

CH

—0-C(CH3)3

3
XIII

Triphenyltin and tributyltin hydride reductions of 4-bromo-3,3-
dimethyl-Z-butanone (XI) and 4-chloro-3,3-dimethyl-2-butanone (XII) afforded
mixtures of unrearranged ketone [3,3-dimethyl-2-butanone (VIII)] and
rearranged ketone [4-methyl-2-pentanone (X)] whose ratio (VIII/X)

decreased with decreasing concentrations of the tin hydride.

A STUDY OF FREE RADICAL
1,2-ACYL MIGRATIONS

By

E. Jerome Maas

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

I970

797/00

To my wife, Peg

ii

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation to Professor
William H. Reusch for his invaluable guidance and encouragement during
this investigation.

Grateful acknowledgment is extended to the National Science

Foundation for providing financial assistance from Grant GP7807.

iii

TABLE OF CONTENTS

Page

HISTORICAL ............................ 1
INTRODUCTION ........................... 6
RESULTS ............................. 8
I. Syntheses ........................ 8

II. Perester Pyrolysis ................... 12

A. Analytical Methods ................ 12

B. Reaction Products ................. l3

1. Unreacted Perester and Products

from the t-Butoxy Radical .......... 13

2. Carbon Dioxide ............... l3

3. Cage Products ................ 15

4. Rearranged Products ............. 15

5. Solvent Products . . . . . . . . . . . . . . 22

III. Halide Decomposition .................. 23

A. Analytical Methods ................ 23

B. Reaction Products ................. 23

DISCUSSION ............................ 24
EXPERIMENTAL ........................... 29
I. General ......................... 29

II. Syntheses ........................ 30

A. General Perester Preparation ........... 30

B. Preparation of t-Butylperlevulinate ........ 30

C. Preparation of t-Butylperlevulinate-d ...... 33

1. Preparation of Levulinic Acid-d5 ...... 33

2. Preparation of t-Butylperlevu11 ate-g5 . . . 33

D. Preparation of t-Butyl—3,3-dimethylperlevu1inate, , 36
1. Preparation of 3,3-Dimethyllevu1inic Acid . . 36

2. Preparation of t-Buty1-3,3-dimethy1per-
levulinate ................. 36

iv

TABLE OF CONTENTS - Continued Page

F. Preparation of 4-Bromo-3,3-dimethyl-2-butanone . . 4]
1. Preparation of Ethyl-3-keto-2, 2-

dimethylbutanoate ...... . 4l
2. Preparation of Ethyl- 3, 3- (ethylenedioxy)-.

2, 2- -dimethylbutanoate . .......... 4]
3. Preparation of 3-Keto-2,2-dimethy1butanol . 45
4. Preparation of 3-Keto-2,2-dimethyl-

butyl tosylate ............... 45
5. Preparation of 4-Bromo-3,3-dimethyl-2-

butanone .................. SI

G. Preparation of 4-Chloro-3,3-dimethy1-2-butanone . 5]

H. Preparation of Triphenyltin Hydride ....... 56

. Thermal Decomposition Procedure ............ 57
A. In Phenyl Ether ................. 57

1. Purification of Phenyl Ether ........ 57

2. Reaction and Analysis Procedure ...... 57

B. In Triglyme ................... 59

1. Purification of Triglyme .......... 59

2. Reaction and Analysis Procedure ...... 59

C. In Sealed Ampoules ................ 59

. Identification of Reaction Products .......... 61
A. Identification of 2-Butanone-g5 . ........ 61

B. Identification of 3,3-Dimethy1-2-butanone . . . . 61

C. Identification of 4-Methy1-2-pentanone ..... . 61
D. Identification of 4-t-Butoxy-3,3-dimethyl-2—
butanone ..................... 62
LITERATURE CITED ........................ 63

TABLE

II
III
IV

LIST OF TABLES

Deuterium Distribution in the 2-Butanone from the
Pyrolysis of Isotopically Labeled t-Butylperlevu-

linate in Triglyme .................
Proton Distribution in Labeled 2-Butanone .....

Calculated Hydrogen-Deuterium Distributions . . . .

NMR Chemical Shifts (r) of CH3COCR1R2CH20C(CH

VI

3)3....

Page

..18
, .19
, .21

62

FIGURE
1.
2.

IO.
11.

12.

13.

14.

15.

16.
17.

LIST OF FIGURES

Page
Plots of perester concentrations versus time ....... 14
Infrared spectrum of 4-t-butoxy-3,3-dimethyl-2-butanone
(neat) .......................... 16
Nuclear magnetic resonance spectrum of 4-t-butoxy-3,3-
dimethyl-Z-butanone ................... l7
Infrared spectrum of t-butylperlevulinate (neat) ..... 31

Nuclear magnetic resonance spectrum of t-butylperlevulinate 32
Infrared spectrum of t-butylperlevulinatefhs (neat) . . . 34

Nuclear magnetic resonance spectrum of t-butylperlevulin-
35

Infrared spectrum of 3,3-dimethyllevulinic acid (neat) . . 37

Nuclear magnetic resonance spectrum of 3,3-dimethy11evu-
linic acid ........................ 38

Infrared spectrum of t-butyl-3,3-dimethylperlevulinate (neat) 39

Nuclear magnetic resonance spectrum of t-buty1-3,3-dimethy1-
perlevulinate ....................... 40

Infrared spectrum of ethyl-3-keto-2,2-dimethylbutanoate
(neat) .......................... 42

Nuclear magnetic resonance spectrum of ethyl-3-keto-2,2-
dimethylbutanoate ..................... 43

Infrared spectrum of ethyl-3,3-(ethylenedioxy)-2,2-dimethy1-
butanoate (neat) ..................... 44

Nuclear magnetic resonance spectrum of ethyl-3,3-(ethy1ene—
dioxy)-2,2-dimethylbutanoate ............... 45

Infrared spectrum of 3-keto-2,2-dimethy1butanol (neat) . . 47

Nuclear magnetic resonance spectrum of 3-keto-2,2-dimethyl-
butanol .......................... 48

vii

LIST OF FIGURES - Continued

FIGURE
18.

19.

20.
21.

22.
23.

Infraped spectrum of 3-keto-2,2-dimethylbutyl tosylate
CCl
4 oooooooooooooooooooooooooo

Nuclear magnetic resonance spectrum of 3-keto-2,2-dimethy1-
butyl tosylate (CC14) ...................

Infrared spectrum of 4-bromo-3,3-dimethy1-2-butanone (neat)

Nuclear magnetic resonance spectrum of 4-bromo-3,3-di-
methyl-Z-butanone .....................

Infrared spectrum of 4-chloro-3,3-dimethyl-2-butanone (neat)

Nuclear magnetic resonance spectrum of 4-chloro-3,3-di-
methyl-Z-butanone .....................

viii

Page

49

50
52

53
54

55

HISTORICAL

Free radicals are electron deficient, uncharged, trivalent Species
which for the most part exist only momentarily during the course of
chemical reactions. These reactive intermediates can react by a variety
of pathways (1), and in some cases may undergo molecular rearrangement
before suffering a terminal reaction. The simplest rearrangement is an
intramolecular 1,2-shift (equation A); and although such transformations
are well known for carbonium ion intermediates (* = +), the number of

authentic free radical examples (* = ') is relatively small. In most of

A R1R2R3CCH2 -———+- RIRZECH2R3
the known free radical rearrangements the moiety undergoing transfer
is an aryl group or a halogen atom, and the 1,2-shift usually proceeds
so as to generate a more stable radical.

Diradicals should be considered as a separate topic, since in cases
where interaction of the reactions sites can occur (e.g. 1,3 diradicals),
a markedly different behavior is observed (2).

Hydrogen and alkyl groups do not normally migrate in free radical
1,2-shifts. Walling states "In most cases where they have been reported
alternate explanations are available for the observed results, or the
radical nature of the reaction is in doubt" (3). Nonetheless, an
apparently unambiguous alkyl group migration has been reported (B). A

fragmentation recombination mechanism was proposed for this reaction.

——» ch

(4)

Aryl group migrations take place in preference to any conceivable

alkyl shifts, as shown in the following examples:

I

H

L;

OCH(CH3) 2

WCH3)2 2

OZCHCHZ'

O2C(CH CH °

3) 2

DiCHZ'
lllilllill—a0(¢)20°

[lilr_1llil.»c(n)20'
N02 ©,C(CH3)2CH2' __. N02©JH2C(CH3)2 (ll)

 

CH ’

CH ‘

s: CH CHCHZO

3

4: (CH CCHZO

 

3)2

 

,; OCHCHZQ

 

4; OC(CH3)CH2O

‘sz

II

 

A

 

(10)

These rearrangements usually proceed so as to give the more stable

radical, however equations K and L illustrate two cases lacking this

driving force.

14
2

 

K OCH CH - :; p‘4CH2CH ° (12)

2 4% 2

(13)

 

L CZCHMCHC 14% = DCHMCDZH

The facility with which aryl groups undergo 1,2-shifts to radical
sites is apparently related to the unsaturation of these groups, since
other radicals having homoallylic unsaturation also rearrange if a more
stable radical can be formed. The following examples of vinyl and acyl

group migrations illustrate this fact.

)CH ° 4:— CH CH=CHCH CHCH

2 3 2 3 (‘4)

M CH3CH=CHC(CH

 

3

 

 

 

 

 

 

N (15)
0 (16)
P . (17)
Q CH3COCO(CH3)CH2' 9: CH3COCH2C(D)CH3 (18)

The last case (equation Q) is interesting in that three rearrangement
pathways are possible: 1,2-shift of a phenyl, methyl, or acyl group.
The preferential migration of the acyl group was unexpected.
Rearrangements involving 1,2-shifts of silicon, sulfur, bromine,

and chlorine atoms have also been reported. These atoms all have valence

4
shell d-orbitals which may be used to form a three membered cyclic trans-

ition state or intermediate. Some examples of these rearrangements are:

 

 

 

 

 

R RC(SR')2CH2' +~> RC(SR')CHZSR' (,9)
R = CH3,D R'= n-Bu, O, p-CH3O
s (CH3)3SiSi(CH3)2CH2"—-—————+-(CH3)3SiCH2Si(CH3)2 (20)
1 [(CH3)3Si]ZSi(CH3)S' =(CH3)35iSi(CH3)SSi(CH3)3 (21)
U (CH3)2CBrCH2° = (CH3)zéCHzBr (22)
v CH3CHBrCH2' :, CH3CHCHZBr (22)
w CC13CHCHzBr -———————> CC12CHC1CHZBr (23)
x CC13C(CH3)CHZBr tCClzCC1(CH3)CHzBr (24)

The absence of bonafide alkyl group 1,2-shifts in radical inter-
mediates can be explained if one considers the activation energy for
such a shift. Application of a truncated Huckel molecular orbital
theory to this system indicates that a cyclic transition state is
energetically unfavorable (25-27). The three atomic orbitals involved
in the migration are converted into three molecular orbitals; one of
these molecular orbitals is bonding and the other two are antibonding.
In carbonium ion rearrangements only two electrons are involved in the
bonding of the migrating group, and these electrons would occupy the

lowest available orbital, which is the bonding orbital. In mono-radicals,

5

however, there are three electrons which must be accomodated and the .
third electron must occupy the lowest anti-bonding orbital, thereby
destabilizing the transition state.

Rearrangements of unsaturated groups such as aryl, vinyl, and acyl
do not fall into this category, and can easily be rationalized by three

membered cyclic intermediates, e.g.:

@
A

aryl ' vinyl acyl

 

we undertook this study to determine whether a 1,2-acyl shift wouid
take place in a system incapable of generating a more favorable radical
by rearrangement and to obtain information concerning the rate of acyl

shift.

INTRODUCTION

Methods of generating suitably substituted 3-ketoradicals (e.g. I
and II) were required for this study of free radical acyl rearrangements

(equation Y).

 

ORI RI fl
III , |
Y R-C-(II-CH2 t C-CHZ-C-R
R' R'
: ': : I:
I R R CH3 III R R CH3
II R=CD3,R :0 IV R=CD3,R'"D

The choice of radicals I and II was based on the expectation that I
would have a moderate driving force for the rearrangement (i.e. a primary
radical going to a tertiary radical), while II would provide a reference
measurement in which this driving force is missing.

Common methods of generating radicals include thermal decomposition
of peresters (equation Z), thermal decomposition of azo compounds
(equation A'), radical initiated decarbonylation of aldehydes (equation 3'),
decomposition of tertiary alkoxy radicals (equation C') and radical

initiated reduction of alkyl halides by tin hydride derivatives (equation 0').

 

 

o
I) .m,

z R-C-O-O-R' A, R'O' + R- -0' R‘

A' R-N=N-R 4 N2 + 2R'

 

 

 

O O
H A -co
8' R-C-H + R'-0° '-—>- R'-OH + R-C' ; R'
I
C' R-C-O' 4: R' + R C=O
I 2
R
D R-X + R35n° at; RBSn-X + R

The present study of acyl rearrangements in radical intermediates
relied heavily on the first method of radical formation, since it is a
generally reliable and well-tested procedure and since the necessary

P-keto acid precursors (equation E') were known compounds.

0 R' o R' 0
. u I II I II
E R-C-C-CHZCOZH -———————4» R—C-C-CHZC-O-O-C(CH3)3
RI RI
V R = CH3, R' = H VIII
A.
VI R - CD3, R' - D IX
VII R = R' - CH3 x
O R' O ' H

 

R-C-C-CH2° «‘————— R-C- -CH C-O' + ‘O-C(CH3)

2
' XI

3

RESULTS

A. Syntheses

Preparation of a selectively deuterated levulinic acid was first
attempted by treating ethyl-4,4-(ethylenedioxy)-pentanoate with alkaline
deuterium oxide in order to form a levulinic acid-g12 having deuterium
at the 2-position. Several different bases were used in these reactions
but no deuterium was incorporated into the resultant levulinic acid at
reflux temperature. When deuterium exchange was attempted in a sealed
tube at 1500 some deuterium was incorporated into the 2-position, but
concomittant cleavage of the ketal occurred and the product was a
mixture of deuterated levulinc acids.

These results indicated that the 2-position of levulinic acid
was relatively resistant to base catalyzed deuterium exchange, so it
seemed that preparation of levulinic acid-g5 might be straightforward.
When ethyllevulinate was treated with alkaline deuterium oxide the
recovered levulinic acid was found to be largely the gs-compound;
however,significant amounts of g3-, g4-, 96-, and g7-labled material
was also present. Mass spectral analysis of the corresponding perester

2,4-dinitrophenylhydrazone derivative showed deuterium incorporation

as follows: 0.6% go; 1.3% g,; 2.9% g2; 7.6% as; 26.8% g4; 44.6% g5;
12.4% g6; 3.5% 97.

The preparation of t-butylperlevulinate and substituted perlevulinate

esters is difficult to accomplish cleanly due to facile cyclization to

pseudo-esters (equation F'). The presence of the 'Y-lactone was easily

0 O (CH ) CO-O O O
F CH3CCH2CH2C-O-OC(CH3)3 ,

detected by a strong band at 910cm'1 in the infrared spectrum; the strong

 

carbonyl stretching absorption at 1785cm'] overlapped the perester

absorption at l775cm'].

Several unsuccessful approaches were tried before
the synthesis of t-butylperlevulinate (VIII), t-butylperlevulinate-g5 (IX),
and t-butyl-3,3-dimethy1perlevulinate (X) were finally achieved by the

following reaction sequence:

 

o r' I§::;V”"*2"C=° o R‘
R I CCH CO H 1' *- R t tCH CO (CH )
| 2 2 2. (CH3)3C-O-OH | 2 3 3 3
R' R'
v R — CH3, R' = H VIII
VI R - CD3, R' D IX
VII R = R' = CH x

These reactions probably proceed by way of a mixed anhydride of
imidazole-N-carboxylic acid and the levulinic acid (equation G'). This
anhydride rapidly loses carbon dioxide to form the imidazolide (equation
H') which then reacts with t-butylhydroperoxide to give the perester.

Unreacted acid was easily recovered and could be recycled.

IO

N -———4§-C=O
6' R[;2? I. I R /’ Al YéiZH‘

RCOZH + N RC-O-C- + HN

1(1)
; RC_ J

L5}

The B-halo ketones 4-bromo-, and 4—chloro-3,3-dimethyl—2-butanone

were prepared by the following scheme:

 

 

 

l. NaH 1. NaH
CH COCH CO CH CH ; CH COCH(CH )CO C CH ——->
3 2223 2.MeLi 3 3 2H232.HeLi
XII XIII
1. (CHZOH)2
CH3COC(CH3)2C02CH2CH3 2 C H 50 H : CH3C(OCH2)2C(CH3)2C02CH2CH3
XIV 7 7 .3 xv
LAH H3O+ TsCl
--———* CH C(OCH ) C(CH ) CH OH —-—>CH COC(CH ) CH 0H———>
XVI XVII
) LiBr (C1) ) ( )
CH COC(CH CH OTs <: CH COC CH CH Br C1
3 3 2 2 2-butanone 3 ( 3 2 2
XVIII XIX (XX)

However, most of the 4-chloro-2-butanone was prepared by another route:

11

OAc

\II Pb(0AC)4 )1;
CH COCH:C(CH ) + H NNII —_. 4 I
3 3 2 2 2 NH N

OAc H
——> MeLi (CH3)3COC1
4 ' 4- CH3COC(CH3)2CH2CI

 

 

Plans to use 4-bromo-3,3-dimethy1-2-butanone as a precursor to
3,3-dimethy11evulinic acid were unsuccessful because the Grignard
reagent from the corresponding ketal bromide led only to the counled
bisketal shown in equation I'. Fortunately the dimethyl levulinic acid
could be prepared from 2,2-dimethylsuccinic acid, as shown in equation

J' (28).

O M 0 A
II
B. 9 W
Ether Y I

 

V

O O
I 0 0 II
- H Ac20 EtOH HOWOR
. HO ‘p —-> ‘
J II
0 0
II I
SOCl2 ’/,\:><:,\\\r/0Et Malonic Et
__> C1 ‘ W (EtOZC)2
0
I I
+ l
H30 WET. ”30+ WH
-C0 0 0

2

12

II. Perester Pyrolysis

A. Analyticai Methods

The carbon dioxide evolved during perester decomposition was collected
in an ascarite trap after volatile organic compounds had been condensed
from the effluent gases. A preliminary wash of the perester pyrolysis
mixture by aqueous NaHCO3 failed to disclose any acidic products.
Unreacted perester was then determined by the method of Silbert and Swern
(29), whereby perester is decomposed by a catalytic amount of ferric ion
in the presence of excess iodide ion; the iodine liberated in this
reaction is titrated with a standard thiosulfate solution. The following
mechanism has been proposed:

(l)Fe(III) +1 —» Fe(II) + I.

(2) Fe(II) + RCO3C(CH3)3 -——-v-(CH3)3CO"+ RCO; + Fe (III)
(3) (C 3)3C0 + H20---—-t-(CH3)3COH + H0“
(4)HO +I ——>HO_ + I'
(5) 21' ----*~IZ

In all cases the amount of unreacted perester was less than 1%. The
presence of triglyme or phenyl ether in the mixture necessitated
mechanical stirring during the titrations; however,the accuracy of
the measurements was apparently not affected by either solvent.

The absence of carboxylic acids and unreacted perester permitted
the work-up procedure to be simplified. Immediately after pyrolysis the
solvent was distilled through a spinning band column and the early

fractions and pot residue were analyzed by vapor phase chromatography

and/or column chromatography.

13

B. Reaction Products

1. Unreacted Perester and Products from the t-Butoxy Radical

 

Pyrolysis of a 0.25M solution of perester in phenyl ether or
triglyme at 130-1320 for 5 1/2 hours resulted in greater than 97%
perester decomposition. The t-butoxy radical gave t-butanol and
acetone which were swept from the pyrolysis flask into the dry ice
traps. The ratio of t-butanol to acetone was 252 in triglyme and

decreased to 1.1 in phenyl ether.

2. Carbon Dioxide

 

The amount of carbon dioxide collected in the ascarite traps
ranged from 60% to 90% of the theoretical quantity, and the rate of gas
evolution was often sporadic. Because of this variance, the original
plan for monitoring the course of the reaction by carbon dioxide evolution
was abandoned, and kinetic measurements were made by determining the
unreacted perester remaining in aliquots of a 0.25M phenyl ether solution
being heated at 130-132°.

A graph of (total perester)/(unreacted perester) versus time in
minutes was plotted for t-butylperlevulinate, t-butyl-3,3-dimethyl-
perlevulinate, and t-buty1-3-phenyl-3-methylperlevu1inate (Figure l).

4

Only t-butylperlevulinate (k = 2.75xlO' sec-A) showed simple first order

kinetic behavior for most of the reaction. Both t-buty1-3,3-dimethylper-
levulinate (k = 3.46x10'4sec-1) and t-butyl-3-phenyl-3-methy1perlevulinate

4

(k = 2.71x10- sec-A) appear to suffer induced decomposition in the early

stages of pyrolysis (see dotted lines Figure l).

14

ON—

32;33522353.?33.3. C .

#32:..628533 3

.mum:w~=>mpgmnpxa=nuu Ac "wavy mzmsm> mcoppusucmucou swpmogoa yo mpopm .p ogsmwm

co

 

 

 

 

\
s
so
\
\
LN
1
1‘
Lo
1
an
E
L.

3352 E 05:.

A3

4.1

 

 

 

ON. 00 O?

N

V

0

 

 

 

 

(10$an popoeiun) / (unwed [0;01)

15
3. Cage Products

 

A previous study of the pyrolysis of t-butyl-3-phenyl-3-methy1-
perlevulinate showed that 23% to 26% of the recombination product
4-t-butoxy-3-phenyl—3-methyl-2-butanone was formed in the course of the
reaction (18). Similar recombination products were anticipated in this
work; however, only the dimethyl perlevulinate ester yielded such a
compound (4-t-butoxy-3,3-dimethyl-2-butanone). The keto ether 4-t-butoxy-
3,3-dimethyl-2-butanone (XXIV) (11% yield based on mmoles perester) was
identified from its ir (Figure 2) and nmr (Figure 3) spectra and by a
carbon-hydrogen-nitrogen microanalysis of its 2,4-dinitrophenylhydrazone
derivative.

The absence of 4-t-butoxy-2-butanone (XXV) among the pyrolysis products
from t-butylperlevulinate was unexpected. It is possible that this keto
ether may have undergone B-elimination to methyl vinyl ketone. No
methyl vinyl ketone was found among the volatile products; however, methyl
vinyl ketone polymerizes readily and could be removed by reaction with

other radicals present in the reaction mixture.

4. Rearranged Products

 

The monomeric rearranged products formed in these pyrolyses
experiments were recovered from the dry ice traps along with acetone and
t-butanol, and the yields (based on mmoles starting perester) of these
simple ketones were obtained by vapor phase chromatography analysis.
Pyrolysis of t-butylperlevulinate yielded 2-butanone in 46.4% yield when
triglyme was the solvent and 42.5% yield when phenyl ether was the solvent.

Pyrolysis of t-butyl-3,3-dimethylperlevulinate (X) in triglyme

resulted in a 23.4% yield (based on mmoles perester) of a ketone mixture

16

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ w m o
wevmuz<rcznz<uh

 

4000

"'00qu (up.

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mun nu tun '1

Infrared spectrum of 4-t-butoxy-3,3-dimethyl-
butanone (neat).

2-

Figure 2.

17

o—

.mcocmpznumuFxcpmewuum.mnxx0p:auuue to Eagpumam mucmCOmmL ovumcmms Cmm_u:z

.m mg:m_d

 

‘

 

 

 

 

 

18

consisting of 95% rearranged ketone (4-methy1-2-pentanone) and 5% 3,3-
dimethy1-2-butanone. When this perester was pyrolysed in phenyl ether,
an 11.7% yield of a ketone mixture composed of 95% 4-methyl-2-pentanone
and 5% 3,3-dimethyl-2-butanone was obtained.

Efforts to determine the extent of acetyl rearrangement in the
2-butanone obtained from the pyrolysis of the deuterium labeled t-butyl-
perlevulinate were complicated by the rather broad distribution of
deuterium in the perester itself. Mass spectral analysis of the ketone
showed a range of deuterium incorporation (Table I) very similar to that
observed for the perester, and it is clear that over 30% of the sample
has less deuterium than desired (d5) while 20% has more deuterium. The
ratio of hydrogen atoms to deuterium atoms in this mixture is 0.67, while
pure 2-butanone-g5 has a 0.60 ratio. Despite this difficulty, the nmr
spectrum of the labeled 2-butanone can be interpreted so as to set an

upper limit on the amount of rearrangement.

 

TABLE I

 

Deuterium Distribution in the 2-Butanone From
the Pyrolysis of Isotopically Labeled
t-Butylperlevulinate in Triglyme

 

 

 

Deuterium (Per Cent) Deuterium and Hydrogen
Incorporation Concentration Units (Based on 800 Total)
Deuterium Hydrogen

d2 1.6 3.2 9.6

d3 7.4 22.2 37.0

d4 24.3 97.2 97.2

d5 46.3 231.5 138.9

d6 16 5 99.0 33.0

d7 3.4 23.8 3.4

99.5 476.9 319.1

 

 

 

 

 

 

19

 

 

 

 

 

TABLE II
Proton Distribution in Labeled 2-Butanone
Source of 2-Butanone Area of Proton Resonance Signal
CH3 —- CO — CH2 — CH3
(a) (b) (C)
Perester Pyrolysis in Triglyme 0.60 t .08 0.22 t .01 3.0
Perester Pyrolysis in Diphenyl Ether 0.69 i .06 0.25 t .02 3.0

 

 

 

 

 

Table II shows the relative areas of the three distinct proton
resonance signals in the nmr spectrum of the labeled 2-butanone. The
chemical shifts of these signals are: (a)‘r 7.94, (b)‘7'7.6l, and (C)
rr'8.98. Now, if the 2-butanone consisted entirely of gs-labeled molecules
with hydrogen located only on groups b and c, the ratio of b/c would be
zero in the absence of any rearrangement and would rise to two if the

substrate were completely rearranged (see equation K'). In the likely

 

O 0
|| Rearrangement
K CD3-C-CD2-CH2 : COB-C-CHz-CD2
IH-abstraction IH-abstraction
0 0
II II
CD3-C-CDZ-CH3 CD3-C-CH2-CD2H
b/c = 0 b/c = 2

event that rearrangement proceeds through a symmetrical intermediate or

causes complete equilibration of the two radicals the maximum b/c ratio

20

would be unity. From these considerations it is a simple matter to
relate the b/c ratio to the mole fraction of rearrangement (x) as follows:
b/c = 2x/3-x. The b/c ratios from the two experiments listed in

Table 11 thus point to 11.9% rearrangement in diphenyl ether and 10.6%
rearrangement in triglyme.

The assumption that all the labeled butanone is g5 is of course not
correct, and the major error in the previous calculations is that
hydrogen atoms remaining at b due to incomplete exchange are counted as
hydrogen atoms from rearrangement. This clearly leads to an exaggerated
value for the mole fraction of rearrangement, and the actual amount of
rearrangement must be less than 11% (perhaps even zero).

The next step in this data analysis is to ascertain whether the mass
spectral data (Table I) and the nmr intensities (Table II) for the 2-
butanone obtained from labeled perester pyrolysis in triglyme are
mutually consistent with a degree of rearrangement ranging from 0-11%.
In order to do this, the isotopic composition at each position in the
2-butanone (i.e. a, b, and c) is given in terms of the hydrogen and
deuterium units defined in Table I (i.e. positions a and c will have
three hundred units each, while b has two hundred units). The distribu-
tions given in Table III depend upon the following assumptions:

1) If there is no rearrangement, all the deuterium units in
2-butanone-g5 are at positions a and b. The extra hydrogen in d2, d3,
and d4-labeled ketone will be at positions a and b, while the extra
deuterium in the d6 and d7 compounds is at c.

2) If 10% of the methylene hydrogen units at c rearrange to b, a
reasonable a/b hydrogen ratio is obtained only if'gll_the extra hydrogen

in the sample (i.e. the 44 hydrogen units associated with the d2, d3, and

21

d4 labeled ketone) is located at position a.
3) The distribution of hydrogen and deuterium for a 5% rearrangement

of methylene groups lies midway between the 0% and 10% extremes.

 

TABLE III

Calculated Hydrogen-Deuterium Distributions

 

 

 

ReaEFERnggAt Hydrogen and Deuterium Distribution
2. 2 2.

0% 32H, 2680 12H, 1880 277H, 23D

5% 38H, 2620 15H, 1850 268H, 320

10% 44H, 256D 18H, 1820 259H, 4lD

 

 

 

 

 

 

Since the total number of hydrogen and deuterium units remains
unchanged for the three cases listed in Table III, we can judge their
relative merits by comparing the ratio of hydrogen atoms in positions
a, b, and c with the experimental values given in Table II. These are
0.35 : 0.13 : 3.0 for 0% rearrangement; 0.43 : 0.15 : 3.0 for 5%
rearrangement and 0.51 : 0.21 : 3.0 for 10% rearrangement. Although
none of these examples fit the observed ratio (0.60 : 0.22 : 3.0)
exactly, it is clear that the best agreement is for 10% rearrangement.
The unique assumption that isotopic exchange in the levulinic ester
precursor is complete at the methylene group (b) but incomplete at the
acetyl methyl group (a) becomes less outrageous if one assumes that the
inductive effect of the carboxyl group greatly enhances the acidity of

the methylene protons.

22

5. Solvent Products

 

Several solvent incorporated products were obtained from the
reactions in phenyl ether. These products appear to arise from a
combination of a methyl radical (from the t-butoxy radical) and/or a
keto radical (from perester decompostion) with a phenyl ether
molecule. This combination produced ortho, meta, and para substituted
products. Both methylated and higher alkylated phenyl ethers were
separated by vapor phase chromatography and identified by spectroscopic
analysis. The ir and nmr spectra were consistent with the structural
assignments; however, it could not be determined whether rearrangement
of the 3-keto radical had occurred before combination with the solvent.

Several solvent derived products were also formed when triglyme was
the pyrolysis solvent. These compounds could be separated by vapor
phase chromatography, and each proved to have some carbonyl incorpora-
tion. One of the products also had a hydroxyl function. These compounds

were not studied further.

23

III. Halide Decomposition

A. Analytical Methods
The triphenyltin hydride reductions were carried out by heating
the reactants in sealed ampoules. These were then cooled to 0° overnight
before being opened. The cooling served to precipitate most of the
triphenyltin compounds, and the remaining solution was then analyzed by

vapor phase chromatography.

8. Reaction Products

Reaction of 4-bromo-3,3-dimethyl-2-butanone with 0.55M triphenyltin
hydride at 132° resulted in formation of a ketone mixture composed of
96% 3,3-dimethy1-2-butanone and 4% 4-methy1-2-pentanone. No starting
keto halide was evident in the mixture. A similar reaction of 4-chloro-
3,3-dimethyl-2-butanone with 0.43M triphenyltin hydride gave a ketone
mixture composed of 90% of 3,3-dimethy1-2-butanone and 10% of 4-methy1-
2-pentanone. Again no starting material remained. The decrease in the
ratio of unrearranged to rearranged ketone from 24 to 9 in these two
experiments corresponds to the decrease in tin hydride concentration.
Reduction of 4-bromo-3,3-dimethy1-2-butanone with a 0.36M concentration
of the less reactive tri-n-butyltin hydride (30) gave an even smaller
ratio (5.5) of unrearranged ketone (3,3-dimethy1-2-butanone) to rearranged

ketone (4-methyl-2-pentanone).

DISCUSSION

The question of whether intramolecular 1,2-acy1 shifts can occur
in free radicals can no longer be argued. Extensive rearrangement of the
3—methy1-3-pheny1-2-ketobutyl radical (XXI) was reported by Curtis
Karl (18); and in this work, rearrangement of the 3,3-dimethyl-2-ketobutyl
radical (I) is clearly well established. These two studies have several

similar features and one important difference.

 

o R R o
. [l | . .' N
L CH3C-lc-CH2 .. f-CHz-CCH3
R' R'
xx1 R - CH3, R' p XXII R CH3, R‘ = 9
I R - R - CH3 III R = R = CH3

The product distribution from radicals I or XXI, generated by tert-
butyl perester decomposition at l30-l32°, was remarkably insensitive to
changes in the solvent, although the rate of carbon dioxide evolution
from I was faster in triglyme than in diphenyl ether. Of course, the
ratio of t-butyl alcohol to acetone (both from the t-butoxyradical) did
rise in a striking fashion when a perester decomposition in diphenyl
ether solution was repeated in p-cymene or triglyme (these latter solvents
provide a much better source of abstractable hydrogen than diphenyl
ether); however, the relative amounts of rearranged ketones and unre-
arranged t-butoxyketones were not significantly affected. Apparently the

initially formed primary radicals undergo cage recombination with a

24

25
t-butoxy radical (equationrfl) or rearrangement to a longer-lived tertiary
radical (equationlj) more rapidly than they participate in hydrogen

abstraction from nearby solvent molecules.

 

O R 0
H I fast I I
M' CH3C-C-CH5 + (CH3)3CO' =~ CH3C-lC-CH2-0C(CH3)3
R' Rl
I R=R'=CH3 XXIV R=R'=CH3
XXIII R = R' = H XXV R = RI = H

The major difference in the fate of these two radicals (I and XXI)
is that XXI leads to considerable dimer formation (coupling of rearranged
radical XXII) while I does not. This is not surprising in View of the
relatively long life expected for the tertiary benzylic radical XXII.

Experiments in which radical I was generated by triphenyl tin
hydride reduction of 4-chloro,and 4-bromo-3,3-dimethy1-2-butanone (31)
established that a 0.55M concentration of the stannane was sufficient
to intercept over 95% of I before rearrangement to 111. This result was
not completely unexpected since Wenkert gt_al;_(32) recently reported
that tri-n-butyltin hydride reacted with 2-dichlormethyl-2-methylcyclo-
hexanone to give 2-ch1oromethyl-2-methylcyclohexanone.

The rate of acetyl rearrangement (i.e. I-¢~III) can be roughly
estimated from the known absolute rate constant for hydrogen abstraction

from triphenyl tin hydride by t-butyl radicals at 25° (k = 5x106M'15ec'1).
Since the present work involves hydrogen abstraction by primary alkyl

radicals at 132° this rate value represents a lower limit for hydrogen

abstraction and consequently a probable upper limit for acyl rearrangement.

26

Thermal decomposition of t-butyl perlevulinate (V) in diphenyl
ether or triglyme proceeded smoothly at 130-132° and yielded methyl ethyl
ketone as one of the major volatile products. Unexpectedly, no trace
of the corresponding cage recombination product [4-t-butoxy-2-butanone
(XXV)] could be found. The addition of authentic keto ether to the
product mixture clearly established that this compound would have been
detected had it been present. The elimination product, methyl vinyl
ketone, was also absent, but would probably have polymerized under the
reaction conditions.

The almost negligible rearrangement observed during decomposition
of deuterium labeled t-butylperlevulinate constitutes one of the most
important facts revealed by this study. In order to graSp the implications
of this result, it is necessary to first consider three plausible

mechanisms for 1,2-acy1 shifts to radicals (Scheme 1). The first

 

 

 

A CH3CO'
R1\\ -_1
C=CH
R”’ 2
CH3 2 CH
I I
" CH 0. C=0
I B 3 I
RIRZC-CHZ ~R] »--——.- R1R2C-CH2
R2
— 5'
CH3"7sr-51
JLc-b 80”, “‘ 80 -—
L31 R'2 _I

 

 

Scheme 1

27

mechanism (A) is a two step process involving fragmentation to an olefin
and an acetyl radical followed by a fast recombination step; the second
mechanism (B) proceeds in two steps through a cyclopropoxy radical
intermediate; and the third mechanism (C) is a single step process,
having extensive odd electron delocalization in the transition state.
Fragmentation-recombination mechanisms have been proposed for some
1,2-bromo shifts (22b) and for the rearrangement of the camphenyl
radical (4). The latter rearrangement required a high temperature
(255-290°), and was aided by the bicyclic nature of the substrate
which prevented the fragments from drifting apart. Because of the lower
temperature employed in the present study and because no evidence of
fragment separation was obtained, it is unlikely that the acyl rearrange-
ments proceed by mechanism A. The most damaging argument against this
mechanism, however, is that if such a fragmentation-recombination
process were to operate when R] and R2 are alkyl or aryl substituents
it should continue to function when R1 and R2 are hydrogen or deuterium
atoms. Thus, the very limited rearrangement observed in the latter
case is inconsistent with mechanism A.
Mechanism 8 appeared at first to be the most reasonable pathway
for the acyl rearrangements studied here, since it parallels similar
mechanisms proposed for the rearrangement of aryl and vinyl groups
(e.g. 9, 12, 13). However, an argument related to that used against
mechanism A also damages the credibility of mechanism B. The formation
of the cyclopropoxy radical intermediate is the key step in mechanism B.
Cyclopropoxy radicals are known to undergo rapid (almost concerted)
ring opening to ‘Y—keto radicals (33); consequently, once the cyclopropoxy

radical has formed rearrangement is assured. This first step should not,

28

however, be seriously affected by variations in R1 and R2. In other
words the increase in radical stability, occurring as a result of
rearrangement when R1 and R2 are alkyl or aryl substituents, should
not greatly influence the formation of the cyclopropoxy intermediate.
As noted previously, the experimental results do not agree with this
conclusion.

The only mechanism (Scheme 1) that clearly predicts the influence
of R1 and R2 on the extent of acyl rearrangement is C. Because of the
delocalization of the unpaired electron in the transition state, it is
possible for the stability of the rearranged radical to modulate the
activation energy of the rearrangement. When R1 and R2 are hydrogen
or deuterium, the rate of rearrangement is so slow that it cannot compete
with hydrogen abstraction from solvent or substrate. When R1 and R2
are alkyl or aryl groups, the rate of rearrangement is increased to suchaui
extent that it occurs to the exclusion of any direct hydrogen abstraction,

except for the special case in which triphenyl tin hydride was added.

EXPERIMENTAL

I. General

Melting_Points

 

Melting points were determined with a Hoover "Uni-Melt" melting
point apparatus and are uncorrected. All melting points were obtained

in open capillaries and are recorded in degrees centigrade.

Spectra

Infrared spectra were obtained with a Perkin-Elmer 237B grating
spectrophotometer. Proton magnetic resonance spectra were measured
with a Varian Model A-6O or a Varian Model HA-lOO high resolution
spectrometer. All nmr spectra were recorded using tetramethylsilane as
an internal standard. The mass spectra were obtained with a Hitachi

Perkin-Elmer Model RMU-6 spectrometer.

Gas Chromatography

 

Vapor phase chromatographic analyses were performed with a Varian
Aerograph A-90P3 chromatograph or a Varian Aerograph 1200 chromatograph.
Preparative work was carried out with the Varian Aerograph A-90P3
chromatograph. Relative peak areas were determined by the triangulation

method.

Elemental'AnalySis

 

All elemental analyses were performed by the Spang Microanalytical

Laboratory, Ann Arbor, Michigan.

29

30

II. Syntheses

A. General Perester Preparation Method

The 1,l'-carbony1diimidazole was dissolved in dry tetrahydrofuran,
distilled directly into the reaction flask, and an equimolar amount of
the levulinic acid in tetrahydrofuran was added dropwise to the stirred
solution. Gas evolution ceased after ca. thirty minutes, and following
a one hour interval, an equimolar amount of freshly distilled t-butyl
hydroperoxide was added dropwise with stirring. The reaction mixture
was stirred at ambient temperature for two days, following which the
solvent was removed under vacuum, keeping the temperature below 50°.

The residue was taken up in ethyl ether and washed successively
with ice cold 10% H2504, ice cold 10% NaZCO3, and water. The ether
solution was dried (M9504) and concentrated under vacuum, and the residue
was held under vacuum overnight to remove any residual t-butyl hydro-
peroxide. Distillation under high vacuum yielded the perester in yields

ranging from 40% to 60%.

B. Preparation of t-Butylperlevulinate
1.89 (11.1 mmole) of 1,1'-Carbonyldiimidazole, 1.39 (11.2 mmole)
of freshly distilled levulinic acid, and 1.09 (11.1 mmole) of freshly
distilled t-butyl hydroperoxide were combined according to the general
procedure outlined above. Distillation afforded 0.999 (53%) of t-butylper-
levulinate: bp 79-80O (0.001mm); ir (Figure 4) 2970, 1775, 1720, 1368,

1185, 1110, and 825m"1

; nmr (Figure 5) r 8.73 (s, 9H), 7.88 (s, 3H),
T 7.42 (q, 4H); 2,4-dinitrophenylhydrazone, mp 135°, parent ion at

m/e 368 in the mass spectrum.

31

(5.1‘ .

MHRONS 50

‘1.(1

71.f1

25
100'

 

 

 

 

 

 

 

 

 

 

 

 

 

7000

 

 

 

 

 

UZ<E§2mZ<mk

:Hmor

F)()(l

71

-‘II)I)II

$01011“va 11 »'

10.0

MICRONS

(‘1

11.0 12.0

.auu..
Efimwu:a

 

 

 

 

 

 

 

 

 

 

 

 

800

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

£3. E E. . E: .2 ~ .3 a M
.1 .3..... L . .uMHM. ... ___. .2. .w_.. IPC . .. ..IIIC. . a .
A t—r... . . For. ..nM. «I. ... u. .
.- .., _ J 1: _._: .2. . T . g I
A . . ,...—”u_..—4 .n 4“: _ . . I... .
.. .21: _ IUI » ._~, _ . .. .,.,
d— . . o. .c. m . u. ~u~I .
. w I . 2 RM 1. .p b . w p
.m; .L J p; . T“. y” 14%
.P .1 . .4 1.1“ .L
.. I .J . f I.
I.» _ __ . . .» :»
I 71 . M: w. _. C 57. .3 ..
..\.\\.. I. 1.. .. L . m .4” . . "I" 2. r 9.
[It M a l. , t. ...., T E. E:
2 u WI. f. .3 s i . L :3.
AIIHIIIPI lull-191.. w .H .- u
L C _ v p... .
L _ I r MI
. IIJIA

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4 I‘llll .
a H . u 2 WV» F.
I N _ ”II“ p; "L . L
. . . h n. H L H . .. .
I. 1.11 I .F .2... 7: . a _. 4

who

1400

who

1000

1600

 

 

5.0

2000

"(QUINCY ICM'I

butylperlevulinate (neat).

Infrared spectrum of t-

Figure 4.

32

 

.mpmcwpz>mpcwa_>p:nuu mo Echumam mocchmmL uwpmcmme cmmpuaz .m mcsmwm
o. a h m n
_

 

 

 

 

33

Anal. Calcd for C15H20N405: C, 48.91; H, 5.47; N, 15.21
Found: C, 48.84, H, 5.43; N, 15.17

C. Preparation of t-Butylperlevulinate-g5

1. Preparation of Levulinic Acid-g5

 

28.89 (0.2 mole) Ethyl levulinate was mixed with 309 (0.22 mole)
potassium carbonate dissolved in 100ml 020 and refluxed for forty-eight
hours under a nitrogen atmOSphere. The mixture was then concentrated
under vacuum and fresh 020 was added to the residue. After refluxing
an additional twelve hours methylene chloride was added to the cooled
mixture which was then carefully neutralized (concentrated HCl) with
stirring and cooling. The aqueous layer was extracted several more times
with methylene chloride and the combined extracts were dried (M9804)

and concentrated, affording 24.39 (100%) of levulinic acid-95.

2. Preparation of t-Butylperlevulinate-g5
Reaction of 23.79 (0.193 mole) levulinic acid-95, 329 (0.197 mole)

1,1'-carbony1diimidazole, and 17.49 (0.193 mole) of t-butylhydrOperoxide
was accomplished as described and afforded 229 (59%) t-butylperlevulinate-QS:
bp 84—85o (.OOlmm); ir (Figure 6) 2985, 1775, 1705, 1370, 1105, and
850cm"; nmr (Figure 7) I 7.56 (s, 1.95H), and T 8.76 (s, 9H); 2,4-
dinitrophenylhydrazone mp 1340 and parent ions at m/e 368 (0.5%). 359
(1.2%), 370 (2.6%), 371 (6.7%), 372 (23.5%), 373 (41.6%), 374 (17.9%),

and 375 (5.9%) in the mass spectrum.

34

n
v

 

 

 

 

 

 

 

 

1500

 

 

 

 

 

2000

 

 

 

 

00

 

I

25

 

 

 

6; 4 . _. filI
”I WWIII— . _ U ,.
aux... J.”
% Hx ‘ n
m. \L
x ., ,
OI._ w n L
4 . _ U ,
H.; w _
...= r.:h

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25'

 

0 0
.0 4
Lawmuz<Azzmz<an

20

 

moumcv (cu-I

 

1L0 120

, I

10.0

       

 

 

 

 

 

 

800

 

 

8.0 MILRQNS

5.1

 

 

 

1000

 

1400

 

1600

 

 

 

 

1800

r

 

 

 

 

2000

muumn 104')

pectrum of t-butylperlevulinate195

Infrared s
(neat).

Figure 6.

mepmcIEmZmapbanuu mo 53:83 8:28me Suwanee .3332 .u 953“.

o .h m

 

 

35

 

 

 

36

0. Preparation of t-Butyl-3,3-dimethylperlevulinate

1. Preparation of 3,3-Dimethyllevulinic Acid

 

Employing the procedure of Baumgarten and Gleason (28) 1,1-dimethy1-
succinic acid was successively converted to its anhydride; l,l-dimethy1-
2-carbethoxypropionic acid; 1,l-dimethyl-2-carbethoxypropionyl chloride;
ethyl-S-(bis-carbethoxy)-3,3-dimethy1levulinate; ethyl-3,3-dimethy11ev-

ulinate; and finally to 3,3-dimethyllevulinic acid.

2. Preparation of t-Butyl-3,3-dimethylperlevulinate

 

Reaction of 16.69 (0.115 mole) of 3,3-dimethy11evulinic acid, 10.79
(0.119 mole) of t-butylhydroperoxide, and 18.659 (0.115 mole) of 1,1'-
carbonyldiimidazole was effected in 200ml of tetrahydrofuran and gave
10.89 (40%) of t-buty1-3,3-dimethylperlevulinate: bp 93-940 (200p);
ir (Figure 10) 2980, 1770, 1700, 1355, 1080, and 850cm']; nmr (Figure 11)
T 7.43 (s, 2H), 7.88 (s, 3H), 8.73 (s, 9H), and r 8.79 (s, 6H); 2,4-
dinitrophenylhydrazone, mp 121-1220, parent ion at m/e 396 in the mass
spectrum.

Anal. Calcd for C17H24N4O7: C, 51.51; H, 6.10; N, 14.13

Found: C, 51.64; H, 6.13; N, 14.20.

E. Preparation of 2-Butanone-g5
7.29 (0.1 mole) 2-Butanone was mixed with 12ml 020 and 0.19 K2C03,
refluxed for seven days under a nitrogen blanket, and then concentrated
under vacuum. The residue was taken up in 12ml 020 and refluxed two
more days under nitrogen. A portion of the mixture was separated on

the vapor phase chromatograph: parent ions at m/e 75 (2.82%), 76 (11.27%),

0.
8

m m

60

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m o
wists... . _ . Au. . d
usrtetfiFuIIL -. . . LI I I“,
M . - . . I
1.1.11 ..IJL.I...II.'II-..L. 3.1.1. . cl 0
1 4.1. h _ 0
.. mu
0
2
O
0
5
2
O
0
O
3
0
O
5
3
0
0
O
.4

 

ICM'I

‘DtOIIINCV

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0
w 2 0
l . z s
I w n I II...
. . .
I . . I I- I...
I: ..lwa-I..- - - _..I
o . . I m
. .1 .. r1.II-A.....21. III II .
2|.-. .w ._ . h I 111 I .I 8
0. . . . ..
0 . 0
m w
5
N
O
R
k
M .m
0 w.
8
m
0.. 4
7 II
. 0
.0
I ..m
0:,
6
.. 0
0
mm
-2. .. , w L
o _. z -s. .I I .. -.m
5 O O
0 O O 02

(04")

1.1001ij

inic

ectrum of 3,3-dimethyllevul

D)
St
a
de
en
r(
a
rd
f..l
.nc
.13
8
e
r
u
g
.1:
F

38

O—

.A¢_uuv uwum owcw_=>mpngpmswu1m.m we Ezgpumam mucmcomwg uwpmcmms Cmmpuzz .m msamwu

o h m

.41-
«III-

 

 

39

6.0

MI(_RON5 5.0

 

 

:10

 

 

 

 

 

 

 

 

 

 

 

 

 

It..-
8 0 Ml(.l<()N5

ill/4m m v

16.0

20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

muumcv (cs-'1

Infrared spectrum of t-butyl-3,3-dimethy1-

perlevulinate (neat).

Figure 10.

4O

 

.mpmcw_=>mpgmachawswcum.mupxu:nup mo Escuuwam mocm:0mwc uvacmwe campusz .FP mezmwu
Op 0 h m k

 

 

 

 

 

41

77 (81.69%), and 78 (4.23%); 2,4-dinitrophenylhydrazone, mp 112-ll3°.

F. Preparation of 4-Bromo-3,3-dimethy1-2-butanone

1. Preparation of Ethyl-3-keto-2,2-dimethylbutanoate

 

Employing the procedure of Marshall and Cannon (34), 2.679 (0.11
mole) of NaH was mixed with 40ml dry benzene and 40ml azeotropically
dried dimethylformamide. Then 17.49 (0.11 mole) of ethylacetoacetate
was added dr0pwise and stirred one hour. 13.29 (0.093 mole) of
Methyl iodide was introduced dr0pwise and the reaction mixture was
heated on the steam bath for three hours and stirred at ambient
temperature overnight. Ethanol was added to destroy excess sodium
hydride and was followed by addition of 50ml water. The two layers
were separated, the water layer was acidified, and the aqueous layer
was extracted several times with ethyl ether. The ether extracts were
combined with the benzene, water washed, dried (M9504), and concentrated.
The residue was reacted as before with the exception that the aqueous
layer was adjusted to a basic pH. Distillation of the reaction product
afforded 9.79 (56%) of ethyl-3-keto-2,2-dimethy1butanoate: bp 34-35°
(0.5mm); ir (Figure 12) 1715, 1265, 1120, and 1155cm"; nmr (Figure 13)
r 5.87 (q, 2H), 7.9 (s, 3H), 8.78 (t, 3H), and r 8.70 (s, 6H).

2. Preparation of Ethyl-3,3-(ethylenedioxy)-2,2-dimethy1butanoate
1309 Ethyl-3-keto-2,2-dimethy1butanoate was mixed with 300ml

benzene, 100ml ethylene glycol, and 0.19 p-toluene sulfonic acid,

and then refluxed for thirty-six hours. Water was removed from the

reaction mixture by means of a Dean-Stark trap. Distillation of the

 

4.0 MICRONS 5.0

 

 

 

 

 

 

 

 

1500 ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

'Illlllil

 

 

 

 

 

u
. «I e .
P .I
. I
. .
.. w. _ . . ‘ a
co 1 14 rrIOI. d
e- .. .: _ ..
_ . . _ _ _
101 .. ll 10' 1 ll Ill till-la III I l
_. H. . . _. ... . .
11 I .-II -I I.I 111 III

 

 

 

 

 

2500

ICM'I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I n m . .
I . #1 . _ ..II M
. _ _
o .-.. . . - .. -IalIIIIIIIIIII
QmLm ITS/F. .:...._. —.: .m . .m-
.. L W . _ _ _
i .1. p I. . _ A _, _ m
.1“: ..M. I..._ .. .- .....-._ .. H 15V
J . .. . . _
11., . .. . H I I. I I 1..--..
,-ss _ "1+..s»s -“
. LN ... _ ., 4 w F -..I_iI
(I L\ ....o.«#¢. -.Ib... ~ .- 1.1-:4 . ...il I .
I Q. r C I»
. I" 4- . .. “INN. . u
.3... *F.. .. H M "
0 0
8 6

a:

u.)

UZ<:._<<mZ<~:.

20
0

IthUINCV

 

16.0
.I

 

 

 

 

 

L
A
.. III I I II I II I 1.1.-
.
11 I1 I II
.

        

1
I

 

 

11.0 12.0

 

 

10.0

 

 

 

 

MILRUNb

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.0 I.
E. muz<E2mz<E

800

1000

1400

1800 1600 I200
moumu 104 '1

2000

Infrared spectrum of ethyl-3-keto-2,2-dimethy1-

butanoate (neat).

Figure 12.

43

.mpmocmusnFxgmewU1N.Nuopwx1mu”>5uw mo Ezguomam mocmcommg ovumcmms Lampuaz

 

 

 

.m_ 0030?;

4 o MICRONS 5

 

S?»

III!”IIL.LJ._.;;.

 

 

! I 1' 1
‘ : Q 3 I I
80 :H—r—w- "T I 'I III.‘
-__- _ . 3 I II:
__._M_... _ -, I Le _'_I, I..I A
1 g '.1 I I i;
60;__ ___ I: ij:‘. _
. ~ 1 1.1 I '
I. 1 I I“

TRANSMITTANCE (7.1)
8
g 1

 

 

 

 

 

 

I ,
I
. _ i

 

I 1
1,

g;
I
I
I.
I

 

 

1
I
I

 

 

I
I I
I

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(“Z11

3500

1800

Figure 14.

 

1 I - ,
2500 2000

"£001ch ICAI‘I

8.0 MICRONS 10.0

I
1
I
1

1__

 

 

1600 1400 1200 1000 800

moumcv (on

Infrared Spectrum of ethyl-3,3-(ethylenedioxy)-
2,2-dimethy1butanoate (neat).

 
  
 

45

.mgmocmu: » asp I . I .
3F :9 .c N N Azxowumcmpxgpmv-m mIFzgum yo Ezguumam mucmcommg uppmcmms mepuaz

 

 

 

 

.m_ mgzmwu

46
reaction mixture afforded 159.99 (91%) of ethy1-3,3-(ethy1enedioxy)-
2,2-dimethy1butan0ate: bp 98-100° (14mm); ir (Figure 14) 1720, 1270,
1125, and 1045cm'I; nmr (Figure 15)‘7’5.95 (q, 2H), 6.12 (s, 4H),
8.72 (s, 3H), 8.80 (t, 3H), and 7'8.82 (s, 9H).

3. Preparation of 3-Keto-2,2-dimethy1butan01

 

6.459 (0.27 mo1e) Lithium a1uminum hydride was mixed with 150 m1 of
tetrahydrofuran that had been fresh1y disti11ed. The mixture was chi11ed,
62.59 (0.31 mo1e) of ethy1-3,3-(ethy1enedioxy)-2,2-dimethy1butanoate was
added dropwise, and then the reaction was ref1uxed for two days. After
coo1in9, 7 ml water was added foT1owed by 7 m1 15% NaOH and 21 m1 water.
The resu1ting white soTid was fi1tered and washed severa1 times with ethy1
ether. The ether was combined with the tetrahydrofuran fi1trate, dried
(M9504), and concentrated on the rotary evaporator. The residue was
stirred overnight in a 50:50 mixture of acetone and water containing a
cataTytic amount of acid, and then this soTution was extracted several
times with methyTene ch1oride. The extracts were washed successiveTy with
water, 10%NaC03, and water, dried (M9504), and concentrated. DistiTTation
of the residue afforded 32.19 (88%) of 3-keto-2,2-dimethy1butano1: bp
78-79° (14mm); ir (Figure 16) 3400, 2950, 1710, 1475, 1375, 1120, and
Iosscm'I; nmr (Figure 17) —r 5.82 (s, 1H), 6.53 (s, 2H), 7.91 (s, 3H), and
'r 8.92 (s, 6H).

4. Preparation of 3-Keto-2,2-dimethy1-buty1 tosylate

 

31.09 (0.267 mo1e) 3-Keto-2,Z-dimethyTbutanol was added dr0pwise as
a so1ution in 60 m1 pyridine to a so1ution of 679 (0.351 moTe) p-toluene-

su1fony1 chToride in 60 m1 pyridine. The mixture was stirred for forty-

47

,0 Mmmows

   
 

 

40

[RANbMHTANCE (7:1

3500 3000 2500 2000 ' 1500

IIIOVINCV [CM '1

I , 6.0 7 MICRONS ".0

I ;

 

I%I

1800 1600 1400 1200 1000 - 800

Figure 16. (Infrared spectrum of 3-keto-2,2-dimethy1butano1
neat .

48

o—

.Focmpznchuwswcum.muopmxum mo Ezcgumam mucmcommg ovumcmme me_u:z
o m .h h o

 

_ _ _

 

 

 

 

 

 

.m_ magmwa

49

8.0

6.0

MICRONS 5.0

4.0

3.5

3.0

2.5

2500 2000 1500

((M'

3000

 

 

 

3500

a. ... l I. “ n I . 'Ivnl
. . . . , . . . . . .
0 4

INQUINCV

16.0

11.0 12.0

MICRONS

 

 

10.0

8.0

 

.15.. ..z. ~...-Iu...v... . .
m m w 4 2

$3 muz<E2mz<fi

 

 

 

 

 

 

 

l .

 

 

a I

 

 

 

1
A

 

l

 

 

LA_L
....

 

 

 

 

 

 

I
,A. . ....;.__.._- V
.
.

1200

In at if NC I! (04"!

ACA—-5.r‘
I
I
I
I
I
I
I
' ' i
l
1
1
I

 

 

 

 

who-

 

 

 

 

 

 

1600

 

 

 

 

 

mmo—

 

 

 

 

 

 

 

 

 

 

 

 

2000

thyTbuty1

11116

pectrum of 3-keto-2,2-d

Infrared s
tosyTate (neat)

Figure 18.

50

.mpMFzmou Fxpaapxspmewuum.muopmxum $0 Ezguumam mucmcommg owpocmms Lompuaz .m_ mgzmwm

o— o m h b 0 n V m

 

 

 

 

 

.11...

51
eight hours, poured into a mixture of ice and water, and extracted

severa1 times with methyTene ch1oride. The combined extracts were
washed successiveiy with c010 10% H2S04, c01d 10% NaC03, and water,
dried (M9504), and concentrated. The resulting oi1 crysta11ized from
pentane affording 43.39 (82.5%) 3-keto-2,2-dimethy1~buty1 tosyTate:
mp 56°; ir (CC14) (Figure 18) 2950, 2800, 1710, 1600, 1360, 1245,

1175, 1100, 975, 850, 745, and 730cm'I

; nmr (C614) (Figure 19) 1 2.50
(q, 4H), 6.09 (s, 3H), 7.53 (s, 2H), 7.96 (s, 3H), and 1 8.91 (s, 6H).
Ana]. CaICd for C13H1804S: C, 57.76; H, 6.71

Found: C, 57.72; H, 6.53

5. Preparation of 4-Bromo-3,3-dimethy1-2-butanone

In the manner of Leriverend and Conia (35), 35.69 (0.132 mo1e) of
3-keto-2,2-dimethy1buty1 tosyTate, 739 (0.84 mole) of 1ithium bromide,
and 500 m1 of 2-butanone are ref1uxed for forty-eight hours. The
mixture is cooTed, fi1tered, poured into wateg,and extracted severa1
times with ethyl ether. The combined ether extracts are dried (M9504)
and concentrated. Disti11ation of the residue afforded 23.59 (95%)
of 4-bromo-3,3-dimethy1-2-butanone: bp 79° (18mm); ir (Figure 20)
2980, 1705, 1470, 1350, 1248, and 1150cm'I; nmr (Figure 21) T 6.48
(s, 2H), 7.84 (s, 3H) and r 8.76 (s, 6H); parent ions at m/e 178 and 180

in the mass spectrum.

G. Preparation of 4-Ch1oro-3,3-dimethy1-2-butanone
Emp1oying the procedure of DePuy, e: '31; (36), mesity1 oxide was
reacted with hydrazine hydrate to give 3,5,5-trimethy1-2-pyrazo1ine.
This pyraonine was reacted with 1ead tetraacetate to afford 3-acetoxy-

3,5,5-trimethy1-1-pyra201ine, which was thermaTTy decomposed to 1,2,2-

52

 

 

4o

 

 

 

 

 

 

 

 

 

L'xlCR()NS

 

 

 

 

 

 

 

obvvmlNCV 'Cu')

16.0
. 5 .J. ,3...
5 ~11 100

 

 

 

 

 

L.-.

 

11.0 12.0

 

10.0

MILRUNS

11.0

 

 

 

 

 

 

 

 

 

 

 

, I cm

41 11. .w

.. .

. .
0.1.?! ..I In 1 I lo ..04. ..1 1h
6 m . II III I [In

9 U .

..c .....s...

 

 

 

 

 

a .
VIII I.s.-.II-.-II-.-i. .... WII-... I I I I“
w . .. ........ . a
s . , _
....-I 1-. 1 III II-.. 2
o. _.. I - _m
5 0 0 0
W ow w 2 2

'(M'l

(an. mu! v

thy1-2-

1me

bromo-3,3-d

Infrared spectrum of 4

butanone (neat)

Figure 20.

53

.mcocmpaniwuFxgpmewuum.muosocnuv $0 Ezcuumnm mucmcommg owumcmms campus: ._N mesmwu

o— o h m h

 

 

 

 

 

 

 

70

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.5

80..

w :8 .;o

navmuz<fiqzn

4
24¢...

 

I 500

KM.)

OHOUHK'

 

 

 

 

 

 

 

 

 

 

10.0

 

 

 

 

MlCl!(.)1‘~I5

 

 

 

 

8.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w w
.5 32<c3mz<fi

 

 

 

 

7000‘

"will no It.“ ‘1

Infrared spectrum of 4-ch10ro-3,3-dimethy1-2-

butanone (neat)

Figure 22.

55

.mcocmuznumupxspmewuum.muogo_;uue mo Eaguumqm mocmcommg ovumcmms Lampozz .mN mesmwm

Op 0 .P m n

 

 

 

 

 

 

 

 

56
trimethyI-cyc1opr0py1 acetate in quantitative yie1d. This cycTopropyT
acetate Ted to 1,2,2-trimethy1-cyc1opropano1 which was c1eaved to the
desired 4-ch10ro-3,3-dimethy1-2-butanone by the action of t-butyT-
hypoch1orite: bp 25° (0.5mm); ir (Figure 22) 2980, 1710, 1470, 1355,
1105, and 765cm'I; nmr (Figure 23) i 8.78 (s, 9H), 7.84 (s, 3H), and

r 6.37 (s, 2H); parent ions at m/e 134 and 136 in the mass spectrum.

H. Preparation of Triphenyltin Hydride

Uti1izin9 the method of Kuivi1a and Beume1 (37), 1.569 (40.9 mo1es)
of Tithium a1uminum hydride and 38.59 (100 mmoTes) of triphenyTtin
ch10ride were added to 150m1 of co1d,anhydrous ether. The mixture was
stirred at ice bath temperatures for 15 minutes, at room temperature for
three hours, and then was sIowTy hydronzed with 100m] of water. The
ether 1ayer was separated, washed with two 100m] portions of ice-water,
dried (M9304), and disti11ed. The triphenyTtin hydride was disti11ed,

using an oi1 bath preheated to 200°, with varying yields (77-85%).

57

III. Thermal Decomposition Procedure

A. In Phenyl Ether

1. Purification of Phenyl Ether

 

Phenyl ether was washed with 10% KOH, followed by a saturated NaCl
solution, and then dried over anhydrous sodium sulfate. Distillation
afforded 99.9% pure ether (v.p.c.). When this solvent was heated to
200° for five hours while being flushed by a slow nitrogen stream,

no volatile substances were collected in dry ice-cooled traps.

2. Reaction and Analysis Procedure

 

The reaction apparatus consisted of a three-necked flask equipped
with a magnetic stirring bar, a serum cap, and gas entrance and exit
tubes. The exit tube was connected to two dry ice-isopropanol traps in
series by tygon tubing. The cold traps were in turn connected to a
tared, ascarite filled "U" tube which was equipped with ground glass
valves. A gas flow detector terminated the exit stream.

The entrance tube to the round-bottom flask was connected to a
source of dry, C02 free, nitrogen. This consisted of a tank of purified
nitrogen, a drying tower filled with ascarite, a scrubber unit filled
with concentrated H2504, and another drying tower filled with alternating
layers of KOH and indicating Drierite.

Most of the phenyl ether was added to the reaction flask which was
then lowered into a preheated oil bath. The system was purged with
nitrogen and then connected to the weighed ascarite tube. The perester

was disso1ved in the remainder of phenyl ether and rapidly introduced

58

into the flask by injection through a serum cap.

After several hours reaction time, the oil bath was removed and
the reaction vessel was immersed in a dry ice-isopropanol bath to quench
the pyrolysis. After warming to ambient temperature, the reaction
mixture was then analyzed for unreacted perester by the method of
Silbert and Swern (29). In this procedure an a1iquot of the pyrolysis
solution was dissolved in chloroform, saturated sodium iodide solution
was added along with 0.002% FeCl3°6H20 in glacial acetic acid, and the
mixture was kept in the dark for ten minutes. Water was then added to
the flask and the mixture was titrated to a starch endpoint using
standard thiosulfate solution. Magnetic stirring was necessary in this
titration.

The carbon dioxide evolved during the perester pyrolysis was
determined by weighing the ascarite trap.

Volatile materials condensed in the dry ice traps were weighed
and analyzed by vapor phase chromatography, 20% SE-30, six foot column.
Most of the acetone and t-butyl alcohol formed in the reaction were
found here.

Fractional distillation of the pyrolysis mixture at reduced pressure
(ca. 1mm) using a spinning band column was continued until most of the
solvent had distilled. The first two fractions (ca. 3ml) contained
additional volatile reaction products (v.p.c. analysis), but the remaining
fractions were greater than 99% pure phenyl ether. These volatile
products were identified by comparing their retention times with those
of known compounds and from the ir, nmr, and mass spectra of samples

trapped by vapor phase chromatography.

59

The residue from the distillation was examined by vapor phase
chromatography and partition chromatography on silica gel. The major
components other than phenyl ether proved to be substituted phenyl

ether. These were collected and examined by spectroscopic methods.

8. In Triglyme

1. Purification of Triglyme

 

Triglyme was dried over magnesium sulfate and distilled from lithium
aluminum hydride. The first fraction contained glyme and diglyme and
was discarded; however, the higher fractions were greater than 99% pure
by v.p.c. analysis. When the purified triglyme was heated to 150° for
several hours in a nitrogen stream, only a trace of volatile material

was collected in a dry ice-cooled trap.

2. Analysis and Reaction Procedure

 

The reaction apparatus and procedure was the same as that for phenyl

ether.

C. In Sealed Ampoules
A benzene solution containing 1.5 equivalents triphenyltin hydride,
1.0 equivalents 4-halo-3,3-dimethyl-2-butanone, and 3-5 mole percent
di-t-butylperoxide was placed in an ampoule made by drawing out a four
inch Pyrex test tube. The ampoule was attached to a vacuum system and
degassed as follows: (1) The ampoule was slowly inserted into a liquid
nitrogen bath. (2) The cooled ampoule was evacuated. (3) The ampoule

was allowed to warm to ambient temperature. This process was repeated

60

three times following which the ampoule was sealed while under vacuum.
An oil bath was preheated to 130° and the sealed ampoule was suspended
in the oil bath for thirty-six to forty-eight hours. At the end

of this time the ampoule was removed and cooled. After the cooled
ampoule was opened, the contents were analyzed by v.p.c The v.p.c.
column was disconnected from the detector immediately after being used,
steam washed several times at 150°, and left overnight with gas flowing
through the column. If this procedure was not followed, the detector

became contaminated.

61

IV. Identification of Reaction Products

A. Identification of 2-Butanone-_d_5

The 2-butanone-g5 separated from the t-butylperlevulinate-d5

pyrolysis products by v.p.c. was collected as a CCl4 solution in
an nmr tube. This solution was compared with a 13614 solution of
synthetically prepared 2-butanone-g5 by nmr spectroscopy and parent ions

at m/e 74 (1.55%), 75 (7.22%), 76 (23.71%), 77 (45.36%), 78 (18.04%),

and 79 (4.12%) were present in the mass spectrum.

8. Identification of 3,3-Dimethy1-2-butanone
The major product from the reaction of 4—bromo-3,3-dimethyl-2-
butanone with triphenyltin hydride was analyzed by nmr and ir spectroscopy.
These spectra were identical with those from authentic 3,3-dimethyl-2-

butanone.

C. Identification of 4-Methyl—2-pentanone

The volatile products from thermal decomposition of t-buty1-3,3-
dimethylperlevulinate were separated by v.p.c. on a 20% SE-30 column.
The major product, with a retention time identical to 4-methyl-2-
pentanone, was passed into a 2,4-dinitrophenylhydrazine solution, and
the solid derivative was isolated and purified. A mixed melting point
with the 2,4-dinitrophenylhydrazone of authentic 4-methyl-2-pentanone
(mp 93-94°) was not depressed.

62
0. Identification of 4-t-Butoxy-3,3-
dimethyl-Z-butanone (XXIV)

The largest concentration of this compound was found in the first
fraction from the Spinning band distillation. This fraction was
separated by v.p.c. on a 20% SE-30 column and the very volatile keto
ether was collected by passing into CCl4 or a 2,4-dinitrophenylhydrazine
solution. The properties of this substance--ir (Figure 2) 2960, 1710,
1365, 1195, 1080, and 980cm'I; nmr (0014) (Figure 3) 16.75 (s, 2H),
7.95 (s, 3H), 8.84 (s, 9H), and -r8.91 (s, 6H); and its 2,4-dinitro-
phenylhydrazone derivative, mp ll5-116°, parent ion at m/e 352 in the
mass spectrum--are consistent with the assigned structure.

Anal. Calcd. for C16H24N405: C, 54.54; H, 6.90; N, 15.90

Found: c, 54.74; H, 6.97; N, 16.06.

An nmr comparison (Table IV) of 4-t-butoxy-2-butanone (XXV) and

4—t-butoxy-3-phenyl-3-methy1-2-butanone (18) supports the assignment of

resonance signals made in this case.

 

 

 

 

TABLE IV
NMR Chemical Shifts (70 of CH3-C0-CR1R2-CH2-OC(CH3)3
(a) (b) (c) (d)

P051t10n R1 = R2 = H R1 = R2 = CH3 R1 = CH3, R2 = 0

(a) 7.92 7.95 8.10

(b) 7.50 8.91 8.53 2.77

(c) 6.44 6.75 6.25

(d) 8.85 8.84 8.87

 

 

 

 

 

(11)
(12)

(13)
(14)

(15)
(16)

(17)
(18)

LITERATURE CITED

a) C. Walling, Molecular Rearrangements , P. deMayo, Ed., John Wiley
and Sons Inc., New York, N. Y., pp. 407-455; b) W. A. Pryor, Free
Radicals , McGraw-Hill Inc., New York, N. Y., 1966.

 

F. D. Greene, W. Adam, and G. A. Knudsen, Jr., J. Org. Chem., 81,
2087-90 (1966).

 

C. Walling, Molecular Rearranggments , P. deMayo, Ed., John Wiley
and Sons Inc., New York, N. Y., 1963, pp. 416-417.

 

J. A. Berson, C. J. Olsen, and J. S. Walia, J. Am. Chem. Soc., 88,
5000-1 (1960); 88, 3337-48 (1962).

 

W. Rickatson and T. S. Stevens, J. Chem. Soc., 1963, 3960.

 

a) C. G. Overberger and H. Gainer, J. Am. Chem. Soc., 80, 4561-5,
(1958); b) F. H. Seubold, ibid., 28, 2532-3 (1953).

D. Y. Curtin and M. G. Hurwitz, ibid., 28, 5381 (1952).
J. Weinstock and S. N. Lewis, ibid., 18, 6243-7 (1957).
J. W. Wilt and H. PhiIip, J. Org, Chem., 88, 891-9 (1960).

 

M. S. Kharasch, A. C. Poshkus, A. Fono, and W. Nudenberg, ibid.,
1_§_, 1458-70 (1951).

C. Ruchardt and R. Hecht, Tet. Letters, 1962, 957-64.

 

a) L. H. Slaugh, J. Am. Chem. Soc., 81, 2262-6 (1959); b) W. B.
Smith and J. D. Anderson, ibid., 88, 656 (1960).

 

 

W. A. Bonner and F. 0. Mayo, J. Org, Chem., 88, 29-34 (1964).

 

L. K. Montgomery, J. W. Matt, and R. Webster, J. Am. Chem. Soc., 88,
923-41 (1967).

 

L. H. Slaugh, ibid., 81, 1522-6 (1965).

H. Reimann, A. S. Capomaggi, T. Strauss, E. P. Oliveto, and D. H. R.
Barton, ibid., 88, 4481-2 (1961).

W. Reusch, C. K. Johnson, and J. A. Manner, ibid., 88, 2803-10 (1966).

C. Karl, Ph. D. Thesis, Michigan State University, East Lansing,
Michigan, 1967.

63

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)
(28)
(29)
(30)
(31)

(32)

(37)

64

R. Kh. Freidlina, A. B. Terentiev, and R. G. Petrova, Doklady Akad.
Nauk S.S.S.R., 119, 860 (1963);'1§1, 866 (1963).

 

H. Sakurai, R. Koh, A. Hasomi, and M. Kumada, Bull. Chem. Soc. Jap.,
88, 2050-1 (1966).

C. G. Pitt and M. S. Fowler, J. Am. Chem. Soc., 88, 1928-30 (1968).

 

a) P. S. Skell, R. G. Allen, and N. G. Gilmour, ibid., 83, 504-5
E19613; b) W. 0. Haag and E. I. Heiba, Tet. Letters, 81: 3683-7
1965 .

 

R. Kh. Freidlina, V. N. Kost, M. Ya. Khorlina, and A. N. Nesmeyanov,
Doklady Akad. Nauk S.S.S.R., 188, 316 (1959).

 

A. N. Nesmeyanov, R. Kh. Freidlina, and A. B. Belyavsky, Izvest.
Akad. Nauk S.S.S.R., Odtel Khim. Nauk, 1959, 1028.

 

 

N. F.)Phe1an, H. H. Jaffe, and M. Orchin, J. Chem. Ed., 88, 626-35
1967 .

 

A. Streitweiser, Molecular Orbital Theory for Organic Chemists, John
Wiley and Sons, Inc., New York, N. Y., 1961, p. 380.

H. E. Zimmerman and A. Zweig, J. Am. Chem. Soc., 88, 1196-1212 (1961).

 

H. E. Baumgarten and D. C. Gleason, J. Org. Chem., 18, 1658-68 (1964).

 

L. S. Silbert and D. Swern, Anal. Chem., 88, 385-7 (1958).

 

D. J. Carlsson and K. U. Ingold, J. Am. Chem. Soc., 88, 1055-6 (1968).

a) L. W. Menapace and H. G. Kuivi1a, ibid., 88, 3047-51 (1964);
b) L. Kap1an, ibid., 88, 4531 (1966)."“‘

 

E. Wenkert, P. Bazukis, R. J. Baumgarten, D. Doddrell, P. W. Jeffs,
C. L. Leicht, R. A. Mueller, and A. Yoshikoshi, ibid., 92, 1617- 24
(1970).

C. H. dePuy, Acct. Chem. Res., 1 33-41 (1968).

 

F. J. Marshall and W. N. Cannon, J. Org. Chem., 81, 245-7 (1956).

 

P. Leriverend and J. M. Conia, Bull. Soc. Chim., 1966, 116-20.

 

C. H. dePuy, W. C. Arney, Jr. , and D. H. Gilson, J. Am. Chem. Soc. ,
90,1830-40 (1968).

 

H. G. Kuivi1a and O. F. Beume1, Jr., ibid., 88, 1246-50 (1964).

   

MICIHIGAN STATE UNIVERSITY LIBRARIES
JIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
3 1293 03145 0319