FIJCH'GAN STATE UNVERRITY
EASI LANSING, MICHIGAN

MlCHIGAN STE‘TE UNIVERSITY

 

 

I.
II.

THE DECOMPOSITION OF CYCLOPROPANEACETYL PEROXIDE
THE DECOMPOSITION OF SOME t-BUTYL PERESTERS DERIVED
FROM ALICYCLIC ACIDS

By

Romeo A. Cipriani

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1961

ACKNOWLEDGMENT

The author wishes to express
his appreciation to Professor Hart,
for his assistance throughout this
investigation, and to the Monsanto
Chemical Company, for its financial
aid during the academic year of
1958-1959 .

ii

I. THE DECOMPOSITION OF CYCLOPROPANEACETYL PEROXIDE

II. THE DECOMPOSITION OF SOME t-BUTYL PERESTERS DERIVED
FROM ALICYCLIC ACIDS

By

Romeo A. Cipriani

AN ABSTRACT

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

DOCTOR OF PHILOSOPHY
Department of Chemistry

1961

Approved

AN ABSTRACT

The purpose of this investigation was two-fold:
to obtain more information about the deCOmposition of
cyclopropaneacetyl peroxide and to study the de30¢position
of t-butyl peresters of cycloalkanecarboxylic acids. In
connection with the work on cyclopropaneacetyl peroxide,
it was necessary to develop a reliable method for the
preparation of cyclopropaneacetic acid in quantity.

A two-step procedure for the preparation of
cycloprOpaneacetic acid was developed. Cyclopropyl
chloride was converted to cyclOpropyllithium using lithium
sand in ether and was then treated with ethylene oxide to
produce fi'-cyc10propylethanol in 66% yield. Alternatively
cyCIOprOpylmagnesium chloride, prepared from cycloprOpyl
chloride and magnesium in tetrahydrofuran and entrained
with benzyl bromide, when treated with ethylene oxide
produced this alcohol; however, the procedure was lengthier
and the yield somewhat lower (42»). Oxidation of
fe-cyclopropylethanol with sulfuric acid-chromium trioxide \
in aqueous acetone provided cyclOpropaneacetic acid in
57% yield.

The decomposition of cycioprOpaneacety‘ peroxide
in carbon tetrachloride at various temperatures and
concentrations was found to follow first-order Pinetics.
The principal products of the reaction were carbon

dioxide and an ester, probably cyclOprOpylcarbinyl

iv

cyclopropaneacetate, along with lesser amounts of
cyclopropaneacetic acid and a substance of undetermined
structure derived from the alkyl part of the peroxide.
Phosgene and hexachloroethane were also produced, but in
undetermined quantities. Whereas the carboxyl group was
quantitatively accounted for, only three-fourths of the
alkyl portion was detected.

The rate of decomposition of cyclopropaneacetyl
peroxide was unaffected by the presence of either acetic
or trimethylacetic acid. However, trichloroacetic acid
accelerated the decomposition, the rate increase being
proportional to the concentration of acid. A similar,
though not quite so large, effect was obtained in the
case of cyclohexaneacetyl peroxide. Accompanying a
decrease in yield of carbon dioxide was the appearance of
an ester of trichloroacetic acid, the nature of the alkyl
group not determined. An accelerated decomposition was
also noted in the presence of pyridine, but here a
comparable effect was obtained with cyclohexaneacetyl
peroxide. The implication of these results on the
mechanism of decomposition of cyclopropaneacetyl peroxide
is discussed.

The decomposition rates of t—buty: peresters of
cyclOprOpanecarboxylic acid and cyclohexanecarboxylic acid
were measured in carbon tetrachloride. At 110*, t-butyl
cyclohexanepercarboxylate dec0mposed three times faster

than t-butyl cyclopropanepercarboxylate; extrapolated to

'(00, the difference is a factor of five. The corresponding
rate difference for the acyl peroxides at 70° is about 140.
The enthalpies of activation, £iH*, for the reactions

were 28.7 and 30.7 kcal./mole respectively, a

somewhat smaller difference than in acyl peroxides. The
results are consistent with some carbon-carbon bond
stretching in the transition state leading to decomposition

of the t-butyl peresters.

vi

TABLE OF C ONTENTS

Page

INTRODUCTION ........ ........ .............. ............ 1
RESULTS AND DISCUSSION...... ........ . ....... .......... 3
I. The Preparation of CyclOprOpaneacetic Acid ..... 5
II. The Deco position of Acyl Peroxides ............ 7

III. The Deco position of t—Butyl Peresters of Some
Alicyclic Acids................................ 36

SUMMARY.......................................... ..... 46
EXPERIMENTAL.......................................... A8
I. The Preparation of CyclOprOpaneacetic Acid..... 4o

A. Displacement Reaction of Cyclopropylcarbinyl
Benzenesulfonate with Potassium yanidc..... 40

B. Attempted Condensation of Cyclopropane-
carboxaldehyde with Rhodanine............... ;

l. The Preparation of a Copper-Zinc Catalyst 49
2. Dehydrogenation of CycloprOpylcarbinol... 51
3. Reaction of Cyclopropanecarboxaldehyde
WithRIIOdaninC...oo...................... .L
C. Preparation of fS-CycloprOpylethanol........ 55

1. Using CyclopropyLnagnesiun Chloride...... 53
2. Using Cyclopropyllithium....... .......... 54

D. Oxidation of xg-CyCIOprOpylethanol.......... 56

1. With Potassium Pernanganate.............. 56
2. With Chromium Trioxide-Sulfuric Acid..... 58

II. The Preparation of Acyl Peroxides.............. 59
A. Preparation of the Acid Chlorides........... 53

1. Preparation of Cyclopropanecarooxy yl

Chloride O O O O O O O 0 I 00000 O O I O O l O 0 O O O O O o O O O O O 5;:
2. Preparation of CycloprOpaneacetyl
Chloride.................. ..... .. ....... . 6}

vii

TABLE OF CONTENTS - Continued

III.

VI.

a. By Exchange with Benzoyl Chloride.....
b. By Exchange with Thionyl Chloride .....
3. Preparation of Cyclohexanecarboxylyl
Chloride........... .....................
A. Preparation of Cyclohexaneacetyl Chloride
5. Preparation of trans- -4- t —Butylcyclo-
hexanecarboxylyl Chloride... .............

B. Preparation of Acyl Peroxides ...............
The Decomposition of Acyl Peroxides............
A. Kinetics of Decomposition.. .............. ...
B. Products of Decomposition...................
l. Trans-A—t—Butylcyclohexanecarboxylyl
Peroxide in Carbon Tetrachloride... ......
2. CyclogrOpanecarboxylyl Peroxide in
t-Butylbenzene...........................
3. CyCTOprOpaneacetyl Peroxide in Carbon
TetraChlorideOl.........OOOOOOIOOOOOOOOOO
A. Cyclopropaneacetyl Peroxide in Carbon
Te trachloride Containing Trichloroacetic
ACid. 0.0.00... OOOOIOOOOOOOOOOOOO 0.0000
5. Initiation of the Polymerization of
Qtyrene by Cycl opropaneacetyl Peroxide...

Preparation of t-Butyl Peresters of Some
Alicyclic Acids.. ........ . .....................

A. Purification of t-Butyl HydrOperoxide.. .....
B. Preparation of the t—Butyl Peresters........
C. Method of Titration of the Peresters........
The Decomposition of Peresters.................
Miscellaneous Experiments ....... ........ ..... ..

A. Preparation of Cyclopropylcarbinyl Cyclo—
propaneacetate. ........... .. ...... ... .......

B. Preparation of Cyclopropylcarb nyl
Trichloroacetate...... ................. .....

C. Reaction of Cyclopropaneacetic Acid with
Sulfuric Acid..... ................ ..........

.viii

74
74

82

TABLE OF CONTENTS - Continued

LITERATURE CITED .............. . .......................
APPENDIX ....... . . . . . . .................................

ix

LIST OF TABLES
Table Page

i. The Decomposition of ncyl Peroxides in Carbon
TetraChloride at rKOOOOIOOOOOOO0.0.0.0....OI0...... 12

2. Effect of Temperature upon the Decomposition of
Cycloalkaneacetyl Peroxides....................... i/

9. Rate Constants for the Decomposition of Cyclo-
propaneacetyl Peroxide Determined by Infrared and Q
Titrinetric Techniques............................ lb

4. Effect of Iodine on the Decomposition Rate of
Cycloa kaneacetyl Peroxides in Carbon Tetrachloride 21

5. Effect of Concentration upon the Rate of
Deconposition of Cyclopropaneacetyl Peroxide in
carbon TetraChlorideOOOO0.00.0.0........OOOOOOOOOO 22

6. Effect of Acids upon the Decomposition of Cyclo-
all{aneacet:‘]rl PerOXideS.o.............o.........c.. 2""

7. Effect of Base upon the Decomposition of
Cycloalkaneacetyl Peroxides....................... 27

8. Products of Decomposition of CycloprOpaneacetyl
Peroxide in Carbon Tetrachloride.................. 2;

9. Products of Decomposition of Certain Acyl
Peroxides in Carbon Tetrachloride................. 50

10. Products of Deco_position of CycloprOpaneacetyl
Peroxide in Carbon Tetrachloride in the Presence
of Trichloroacetic Acid........................... 32

11- Decomposition of t-Butyl Peresters in Chlorobenzene 37

12. DeCOnposition of t-Butyl Cycloalkanepercarboxylates
in Carbon Tetrachloride........................... p)

13. Comparison of the Decomposition of t-Butyl
PereSterS and Acyl PerOXideSooooooooo00.00.0000... E“)

14. Decomposition of Cyclopropaneacetyl Peroxide in
Carbon Tetrachloride at 44.00..................... E;

15- Dec0nposition of CycloprOpaneacetyl Peroxide in ‘
Carbon Tetrachloride at 44.00..................... b;

LIST‘OF TABLES - Continued

21.

22.

24.

25.

27.

28.

DeCOIposition of Cyc10pr0paneacetyl Peroxide in
Carbon Tetrachloride at A9.80.....................

Decomposition of CyCIOpropaneacetyl Peroxide in
Carbon Tetrachloride at 50.80.....................

Deconposition of Cyclopronaneacetyl Peroxide in
Carbon Tetrachloride at 50.80.....................

Deconposition of CyCIOprOpaneacetyl Peroxide in
Carbon Tetrachloride at 50.8“.....................

Decomposition of Cyclopropaneacetyl Peroxide in
carbon TetraChj-Oride at 56.170000000000009.00......

Decomposition of Cyclopropaneacetyl Peroxide in
carbon TetI‘aChlorj-de at 5L)O(}IOOOOOOOOOOOOOOOOO0....

Decom,osition of CyclOprOpaneacetyl Peroxide in
Carbon Tetrachloride at 56.50.....................

Decomposition of CyclOpropaneacetyi Peroxide in
Carbon Tetrachloride in the Presence of Iodine,

56.50.............................................

Decomposition of Cyclopropaneacetyl Peroxide in
Carbon Tetrachloride in the Presence of Iodine,

56.50.............................................

Decomposition of Cyclopropaneacetyl Peroxide in
Carbon Tetrachloride, Followed Titrimetrically,

56.50.............................................

Decomposition of CycloprOpaneacetyl Peroxide in
Carbon Tetrachloride, Fo7lowed Titrinetrically,

44.50.............................................

Decomposition of CycloprOpaneacetyl Peroxide in
Carbon Tetrachloride, Followed Titrinetrically,

44.50.............................................

Decomposition of CycloprOpaneacetyl Peroxide in
Carbon Tetrachloride after Washing with sodium
carbonate sojution! 44 .50 I o a o o o o o o o o o o o o o o o I o o o o o .

Decomposition of 0.386 N CyCIOpropaneacetyl
Peroxide in Carbon Tetrachloride, #4.5°...........

Decomposition of CyClODPOPaneacetyl Peroxide in
Carbon Tetrachloride Containing ACEtiC *Cid’ ”H'SO

xi

Gk

LIST OF TABLES — Continued

#0.

Al.

42.

43.

in.

45.

Decomporitian of CycloprOpaneacetyl Jerodee

in Carbon Tetrachloride Containing Tri ethyl—
acetic [AC—id;41+.500coo-00.000.000.0000000.00.00.00

Decouposition of CyCIOprOpaneacetyl Peroxide
in Carbox Tetrachloride Containing Pyridine, 44.50

Deco po;ition of CycloprOpaneacetyl Peroxide

in Carbon Tetrachloride Conta;ning Tricnloro-
acetic Acid, 44.50................................
Deco posi ion of CyCIOpropaneacetyl Peroxide

in Carbon Tetrachloride Containing Trichloro-
acetic Acid. 44.50................................

Deco position of Cyclopropaneacetyl Peroxide
in Carbon Tetrachloride Containing Trichloro-
acetic .“.‘3id} 44050000000000.00000000000.00.000.000

Decomposition of Cyclohexaneacetyl Peroxide in
Carbon Tetrachloride. 4A.5°.......................

Decomposition of Cyclohexaneacetyl Peroxide in
Carbon Tetrachloride. 64.30.......................

Decomposition of Cyclohexaneacetyl Peroxide in
Carbon Tetrachloride Containing Iodine, 64.30.....

Decomposition of Cyclohexaneacetyl leroxide in
carbon TetraC£1:Oride, Ff:080000000000000000.0000...

Decomposition of Cyclohexaneacetyl Peroxide in
Carbon Tetrachloride Containing Pyridine, 71.80...

Deco position of Cyclohexaneacetyl Peroxide in
Carbon Tetrachloride Containing Trichloroacetic
IACid: C'uOBOOOOOOOOOOOIOOOOO......OOOOOOOO00.00....

Decomposition of t-Butyl Cyclohexanepercarboxyiate
in Carbon Tetrachloride, 8).6°....................

Deconposition of t-Butyl Cyclohexanepercarboxylate
in Carbon Tetrachloride, 89.60....................

Deco position of t-Butyl Cyclohexanepercarboxylate
in Carbon Tetrachloride, 1000.....................

Decomposition of t-Butyl Cyclohexanepercarboxylate
in Carbon Tetrachloride, 100°.....................

co
m

(‘3')

I '\
\i,’

\

lOO

LIST OF TABLES - Continued

#6.

Deconposition of t—Butyl
in Carbon Tetrachloride,
47. DECOPPOSIEIOU of t-Butyl
in Carbon Tetrachloride,
48. Deco position of t—Butyl
in Carbon Tetrachloride,
Aj. Dec0nposit;on of t—Butyl
in Carbon Tetrachloride,
50. Deco:position of t-Butyl
in Carbon Tetrachloride,
51. Deco position of t—Butyl
in Carbon Tetrachloride,
52. Deco position of t-Butyl
in Carbon Tetrachloride,

Decomposition of t—Butyl
in Carbon Tetrachloride,

Page

Cyclohexanepercarboxylate

llO“................ 105

Cyclohexanepercarboxylate

1100 0 O O O O O O O O O O O O I 0 I O O I O O 105
CyclOpropanepercarboxyiatc ,
1100 O O O O O O O O O O O O O O O O O O I I O 10';

Cyclopropanepercarboxyl te
3.100.000.0000.000.000....10L)

Cyclopropanepercarboxylate
1200......

CycloprOpanepercarboxylate
1200.00.00.00000000000...1-07
CycloprOpanepercarboxylate

129° 0 10C‘

Cyclopropanepercarboxylate

12900ooooooooooooOOooo'ooo108

xiii

F Igure P:

Arrhenius Plot for the Decomposition of
CyCIOprOpaneacetyl Peroxide.......................

2. First-order Rate Curves for the Deco position of
Cyclopropaneacetyl Peroxide at Various Temperatures

3. Arrhenius Plot for the Decowposition of t-Butyl
Cyclohexanepercarboxylate.........................

4. First—order Rate Curves for the Deco:position of
t-Butyl Cyclohexanepercarboxylate at Various
Ternt)era tures . . 0 Q . . 0 U . . . . O O C O O C O O . O 0 O . O O O C . O O O O O C C O

5. Arrhenius Plot for the Decomposition of t—Butyl
Cyclopropanepercarboxylate........................

6. First-order Rate Curves for the Decomposition of
t-Butyl Cyciopropanepercarboxylate at VEPIOIS
Terwl-{jer‘ajcureSOOOOOOI....0.........OOOOOOOOOOOOOOOIO

1- Infrared Spectrum of Acid Obtained fro. Reaction
Product of CyCLOpropylcarbinyl Benzenesulfonate
al‘ld POtaSSiu'.-: Cb'alhlideOO... O O... 0 O. 0 00.00.00... 0 O. 0

8. Infrared Spectrum of 5 -Cyclopr0py1ethanol in
Carbon Tetrachloride..............................

9. Infrared Spectrum of CyclOprOpaneacetic ACid in
Carbon Tetrachloride..............................

10. Infrared Spectrum of Cyclopropaneacetyl Chloride
in Carbon Tetrachloride...........................

ll. Infrared Spectrum of CycloprOpaneacetyl Peroxide
in Carbon Tetrachloride...........................

l2. Infrared Spectrum of Cyclopropaneacetyl Peroxide
in Carbon Disulfide...............................

13. Diagram of the Apparatus Used for Decomposition

RLIZIS C O C O C I O O O C O O C . O O O I O O O O O C O O O C O I O O O O I O O O O O I I O O O C

$-

Infrared Spectrum of the High Boiling Material
Obtained from the Decomposition of CycloprOpane-
acetyl Peroxide in Carbon Tetrachloride...........

X 3

}-
<1

LIST OF FIGURES - Continued

15. Infrared Spectrum of t-Butyl Cyclohexane-
percarboxylate in Caroon Tetrachloride............ 7

l6. Infrared Spectrun of t—Butyl Cyclopropane-
percarboxylate in Carbon Tetrachloride............ Yb

1?. Infrared Spectrum of Cyclopropylcarbinyl
Cyclopropaneacetate in Carbon Tetrachloride....... 81

18. Infrared Spectrum of CycloprOpylcarbinyl
Trichloroacetate in Carbon Tetrachloride..........

0’)
\JJ

XV

INTRODUCTION

Tnis thesis deals with the preparation of cyclo-
prOpaneacetic acid, the decomposition of cyclopropane-
acetyl peroxide and the decomposition of t-butyl
peresters of certain alicyclic acids.

In the decomposition of peroxides derived irom
alicyclic acids, Hart and wyman (i) noted that the
behavior of cycloprOpaneacetyl peroxide was incongruous,
both in rapid rate of decomposition and in the high yield
of ester, with other cycloalkaneacetyl peroxides studied.
Furthermore, the rates were erratic, suggesting that the
peroxide contained either an accelerator or an inhibitor
of an undetermined nature. The purpose of this investiga-
tion then was to re-exanine the decomposition to better
understand the role of the cyclopropylcarbinyl system in
free radical reactions. Rate constants and products were
to be re-determined, paying special heed to the purity
of the peroxide, p ssible catalysis by acid or base,
induced dec0mposition and the energetics of the deconposi—
tion.

The relatively large amounts of cyclopropaneacetic
acid needed for preparation of cycloprOpaneacetyl peroxide
in sufficient quantity for rate and product studies
required the development of a reliable synthetic procedure
adaptable to large quantities; initial efforts were

directed towards this goal and a method superior to those

‘in the literature was developed. Study of the peroxide
decomposition then became possible.

Acyl peroxides derived from cycloalkanecarboxylic
acids decomposed at different rates, depending upon the
size of the ring (1). This suggests that the stability
of the alkyl radical which is produced when an acyl
peroxide deco poses influences the rate. The effect was
extensively studied by Bartlett and co-workers (2, 3, 4)
using a wide variety of peresters. It was of interest
to determine whether peresters were more or less sensitive
to changes in structure of the alkyl group than were acyl
peroxides. Towards this end, t-butyl peresters of
cyc10propanecarboxylic and cyclohexanecarboxylic acids
were synthesized and their rates of decomposition conpared

with those of the corresponding peroxides.

RESULTS AND DISCUSSION

A. The Preparation of CyCIOprOpaneacetic Acid

CycloprOpaneacetic acid is perhaps one of the host
difficult of the simple acids to prepare in good yield
and iarge quantity, for most of the conventional methods
fail. Because of the unreactivity of cycloprOpyl halides
in displacement reactions, procedures such as condensation
of cycloprOpyl chloride with malonic ester do not work.
Grignard reagents are easily formed from cyclOprOpyl
bronide and iodide, but these are not readily available;
cyclOpropyl chloride, which is readily available, could
be converted to the Grignard reagent only in small yield,
using ether as the solvent (5). Carbonation of the
Grignard reagent obtained from cyclopropylcarbinyl bromide
produces allylacetic acid exclusively (6). .

The first unequivocal synthesis of cyclopropaneacetic
acid was that of Smith and MacKenzie (6), who prepared it

according to equation-l; the overall yield was low. Two

 

O 0 N:
H N2CH-002C2H5 g H fl A820
C—Cl % [> -C- OC2H5 _________>,

C2H OH 1) KOH
Dian) —~—-—-5:————> C>—cs(cogczss)2 —- —>

2) H+
/
o/'\002H5

O

[:>*CH(002H)2 -EEEE¥> [:>»CH2gOH (1)

subsequent nethods, though useful, possessed undesirable
features. Wallis and Turnbull (7) used the Arndt-Eistert

reaction (equation 2), which is good for only small

 

\\\> B CH2H2 . 8 Aseo
. /—-C-Cl --> [::\—CCHN2 -——-—?
t/ CH30H
[::>-CH 8cc 1) KOH b\\ 0 @0H (2)
20 H3 —————--+ " H2
2) H+ '///_

quantities because of the hazards involved in using
diazomethane. To obtain cyclopropaneacetic acid, Hart
and wyman (1) treated cycloprOpyllithium with ethylene
oxide, denydrogenated the resulting alcohol and oxidized
the corresponding aldehyde (equation 3). Although this

Chg-CH2
L i 0/ '\
[:>~c: -———-——e Li -———-w-9 L//~CH2CH20H

Pentane

0
Ag H As 0 @
-——_—.—") D'CHECH 4") [>’CH2 OH (
Pumice

route is adaptable to a large scale, the preparation of

)

\1.’

cyclopropyllithiuh, which was done in pentane, was found
to be erratic. Furthermore, the procedure used to prepare
the dehydrogenation catalyst lacked sufficient detail for
successful reproduction of the yield.

In the attempt to find another practical procedure,
cycloprOpylcarbinyl benzenesulfonate was treated with
potassium cyanide in order to obtain cyclopropane-

acetonitrile and, subsequently, the acid; this was

5
abandoned, for only a small amount of crude acidic
haterial was obtained. Similar treatment of the cyclo-
propylcarbinyl tosylate also was unsuccessful because the
tosylate was unstable (46). Extending the side chain of
cyclopropanecarboxaldehyde using rhodanine, an excellent
method for aromatic aldenydes, also was futile because
the intermediates were ill-defined and obtained in poor
yield.

The success in preparing the Grignard reagent of vinyl
chloride using tetrahydrofuran as the solvent (8)
encouraged work on the preparation of cyclopropylmagnesium
chloride. It was found that cyclOpropyl chloride,
entrained with one-fifth of a mole of benzyl bronide,
reacted with magnesium in tetrahydrofuran to give the
combined Grignard reagents in 76% yield. Treatment with
ethylene oxide produced (9 -cyc10pr0pylethanol in ”(27:3
yield, based on cyclopropyl chloride. Ethylene
chlorohydrin was forhed as a co-product, and, because
of the closeness of the boiling point to that of the
desired product, had to be removed by treatment with
sodium hydroxide. The discovery that ether could be used
as a solvent to prepare cycloprOpyllithium (9)
consistently eliminated the problem with ethylene
chlorohydrin and @i-cyclopropylethanol could be obtained
in 66p yield.

Oxidation of the alcohol to the acid presented other

difficulties, for acidic oxidizing agents attack the

6

cyclopropane ring, resulting in ring opening. Oxidation
with basic potassium permanganate produced a mixture of
cyclopropanecarboxylic and cycloprOpaneacetic acids.
However, a chronic oxide and sulfuric acid solution in
water added slowly to an acetone solution of fi3-cyclopr0pyl-
ethanol, a method previously used for the oxidation of
unsaturated alconols (10), produced cyclopropaneacetic
acid in 57% yield. The total reaction is shown in
equations 4-6.
CH2;pH2
1) ‘\o P\

Mg \\\.,- L. \
l:>”01 -% PMOCl .7 -~CH20H20H 4
THF, CbHSCHaBP c// b 2) HOH, H+ L/’ ( )

 

 

CHa-CH2

. 1) \0/
LL [:>~ [:5» i “a CH (5)
_" i ————-—~-.---_..__> L» .3612 f
[::>Cl Ether L 2) HOH, H+ i

 

° : CrOe. H2804 CH fiOH (,)
>CH2CH20H Acetone —) [>— 2, D

7
B. The Deco p sition of Acyl Peroxides
Detailed discussions of acyl peroxide decompositions
can be found in several textbooks (11, 12, 13, 14). In

order to intrOduce the reader to certain phases of the

field, the mechanisms of these reactions will be briefly

discussed.

The decomposition of acyl peroxides can occur in a
variety of ways. Primarily, there are the spontaneous
reactions, in wh ch the oxygen-oxygen bond breaks

honolytically to form radicals

P u
Rdoocn -————+ 2 RCO' (r)
or heterolytically to form ions (15).
O O
l I II a -ll
stoogs' -—————> RCO + OCR' (8)
III addition to these modes of decomposition, the acyl
pexroxide can undergo induced decomposition either by
:raciicals (l6, 1!) or by ions (18).
Spontaneous homolytic cleavage prevails when a
:syhs etrical peroxide undergoes deco position in a
:reliatively non-polar solvent, such as benzene or carbon
'teinrachloride. This is probably the result of the small
tNDnCl dissociation energy of the oxygen-oxygen linkage.
For example, the bond dissociation energies of acetyl and
benlzoyfl.peroxides have been estimated to be 27-35
KCal./r..ole (19). This type of dec0mposition will be

Qixscussed further in later sections.

8

Often accompanying this reaction is induced
decohposition caused by the presence of radicals in
solution. The relative contribution depends upon the
temperature (20) and the initial peroxide concentration,
as well as on the solvent employed (17). Since this gives
rise to higher order reactions, thereby conplicating
kinetics, it must be eliminated in order to simplify the
study of the spontaneous deCOmposition. Either kinetic
analysis (17) or the addition of a radical trap such as
styrene or iodine (21), which intercepts the radicals, is
effective. The point of attack of such a decomposition
does vary: in the decomposition of benzoyl peroxide in
'benzene containing the triphenylmethyl radical, attack is

'both on the ring and on the oxygen (22)

W +012—aic; 1 r +[Olc

(9)
vhuereas in ethyl ether, attack occurs at the peroxide
oxygen (23) .

13 IO:L8 CH3 0

0
II H
[.COJ2 + CH30HOCH2CH3 -—>(/__ \,>C' -OCHOCH20H3 +</ \)CO'

(10)

A decrease in symmetry of the molecule is accompanied
by EH1 increasing tendency for the reaction to proceed
hetercdytically. This type of decomposition may run the
gamut from couplete dissociation into ions, through

intermediate stages, to a slight polarization of the

oxygen—oxygen bond (15).

o o 92'

RCIOOICIR' ———> Rc'ofiP + “001 (11)
g 5+ a’fi

——) Rc-o—o-CR' (12)

As would be expected, solvents of high dielectric constant
favor this mode of reaction. An example is the
decomposition of p-methoxy-pknitrobenzoyl peroxide studied
by Leffler (18). In benzene, this peroxide decomposes at
the same rate as benzoyl peroxide; in nitrobenzene,
however, the decomposition proceeds eight times faster,
the increase being attributed to an ionic mechanism
operating in polar solvents. The postulated mechanism is

shown in equations 13 to 15.

CH30©>ICOOCQN02 ——-> CHaoOc-fia-o
P. e
CH30<_\>-'C -o@ —-—> CHsofl—o-cm (11+)
0 O 0
e9 ‘3‘ ., ll
CHSO -c::o + o-BQNoz ———> caaoQ o'éoc®—Noa (15)

(I)

TTLis was substantiated by isolation of (I) in 38% yield

(5:0

0(3~{:>4‘102 (13)

When thionyl chloride was used as the solvent. The

pernaxide is also susceptible to induced decomposition by
acixis. Whereas benzoyl peroxide is virtually insensitive
to Tune presence of all but strong mineral acids, the rate
0f Checomposition of p-methoxy-p'-nitrobenzoyl peroxide is

proportional to the acidity constant of the acid. Similar

10
catalysis occurs with phenylacetyl peroxide, the
spontaneous decomposition being autocatalytic because of
the acid formed during the reaction (24).

In the spontaneous houolytic cleavage of acyl
peroxides in non-polar solvents, a najor problem is the
number of bonds broken in the rate-determining step. The
decomposition can proceed by one- or by two-bond scission,

as shown in equations 16 and 17.
O

n H H
R-COOC-R ~——~> 2 R-CO° (16)

O
1 l n H u
R-ClOOC-R -—) R-CO----O-=C----R ———> R-CO- + 002 + R- (1?)

Hammond and Soffer (21) demonstrated that the
deconposition of benzoyl peroxide results in the formation
of two benzoyloxy radicals which, in the presence of
ixxiine and water, quantitatively produce benzoic acid.
Imuase results demonstrate the complete capture of the
bernzoyloxy radical as the hypoiodite; one concludes that
injgtial reaction involves the rupture solely of the
Oxygen—oxygen bond (equation 18). The decomposition of
acetxyl peroxide in moist carbon tetrachloride containing

iodjnie, however, produced principally methyl iodide,

H
-/ \\COH + 2HI

2
O O Q.& \—
H ”
[©—c-oJa-+2< / Vol-o ——+2 / lCOI }
...... \2 // \31 + 2002

.\ —“ .’

(18)

11

possibly indicating multiple scission (25). But Szwarc
found that the activation energies of propionyl and
butyryl peroxides are, within experimental error, the
same as for acetyl peroxide and this is assumed to be the
oxygen—oxygen bond energy in acyl peroxides (26). The
absence of acetic acid in the reaction products is
attributed to the extreme instability of the resulting

acetoxy radical, the decomposition being strongly

exothermic (25).

A very detailed study of the products of decomposition
of delta-phenylvaleryl peroxide in carbon tetrachloride
was made by DeTar and Weis (27). The major products were
carbon dioxide (84%), acid (hm), ester (17.5%),
1,8-diphenyloctane (21.5%) and l-chloro-fl-phenylbutane
(41%”. They concluded that two acyloxy groups were
iJLitially formed and subsequent loss of carbon dioxide
was; very rapid. With the exception of l-chloro-n-phenyl-
butxane, which resulted from chlorine abstraction frox
the! solvent by the 8-pneny1buty1 radical, the products
werwe formed entirely within the solvent cage.

(jther cases show the carbon-carbon cleavage to be
impcnatant, especially where the resulting radical is
stablxa. Bartlett and Leffler (24) found that the
deconixosition of phenylacetyl peroxide proceeds more
rapidly’at.oo than benzoyl peroxide at 80° and attributed
the ixuxrease to the formation of the benzyl radical in the

rat - ,' .' - o 1
e detuermlnlng step. The deCOmpOSltion scne e is snown

12

in equation 19. The relationship between the stability

0
II
OCH? + 002 + 'OCCHa/ )

'I- -‘

/ \ CH2€O2 a / O (19)

....— \ - I
<:\>CH2® + 002 + OICCHZ/ \

—
h-

of the resulting radical and the ease of deco;..position
has been demonstrated in the study of the decomposition
of‘ t-butyl peresters (2). This effect has also been
noticed by Hart and wyman (1), who found that the
peroxides derived from cycloal‘xanecarboxylic acids
decomposed at a rate which depended on the size of the
ring. Furthermore, acyl peroxides which gave secondary
alkyl radicals decoz.;posed more rapidly than those which
gave primary radicals (acyl peroxides derived from

CYC loalkaneacetic acids). Their results are listed in

 

Table l. The rate of decomposition and the heats of
TABLE 1

THE DECOMPOSITION OF ACYL PEROXIDES, (Rcoe)2. IN CARBON
TETRACHLORIDE AT 70° (1)

‘

 

 

Relative A H*
_\R Rate Ideal/mole
Cb’c lOprOpyl (abs. rate const. - 1 28-3

4.8 x 10 4880.1)

C‘Jc lobutyl 16.8 26.6
CU‘C ] opentyl 72 .3 25 0
C3“: 7 Ohexyl 157 . 2 26. 1
eye lohepty’l 161 21+ .8
CyClObutylcarbinyl 6.1} 25 7
Cb’cloioentylcarbinyl 4 . 1 26.3
wexylcarbinyl 5.9 25 .7

 

13
activation of the cycloalkanecarboxylyl peroxides
decrease with increasing stability of the cycloalZ-zyl
radical. In the cycloalkaneacetyl peroxides,
(( H2) CH-CH2002)2, however, an anamolOi-s effect was
obtained. Where n = 3, 1+ or 5, the c3-;clobutyl,
cyclopentyl or. the c3,cl ohe:-:3,l derivative, the peroxide:
deconposed at essentially the same rate; wnere n = 2,
the cyoloprOpylcarbinyl syste:.., extrei..el3; idsc rates
were obtained. It is th._s case which is of pr is interest
in this thesis.

There are many reactions in wnich rates are greatly
(accelerated by the presence of a cyclOpropane ring. Under
ionizing conditions, the cyclOprOpylcarbinyl system often
undergoes rearrangement to allylcarbinyl and cyc’iobutyl
derivatives (28) and these reactions proceed faster than
the correSponding allyl co;..pounds (29). In the
diazotization of cyclopropylcarbinylamine, Roberts et. al.
(30) showed that the cycloprOpylcarbinyl cation rearranges

eXtGeldsively to prodqu a 1:;ixture of products (equation 20).

 

O - “...,OH
HN '5: , | ‘
D CHgNHz 2 ’ 3 {> CH20H + + CH23CH-CH2CH20H
8s 47% 5%

(20)
However, in reversible reactions, :1 ultiplicl y of

DI‘C>Cil.‘1cts is not obtained; instead, the 1ost stable

derivative is produced, as in the treatment of cyclo-
DPC31—h/Iicarbinol wl t;1 Lucas reagent, where M-chloro-l-butene

is Obtained exclusively .

11+
TAE presence of such an equilibrium in free radical
reactions has not been established, a7though several
reactions involving; the cycloprOpylcarbinyl sy
been reported. Overberger and Lebovitz (51), in a study
of the deco:.position of several alicycch substituted

azo—bisnitriles, found results si..ilar to those reported
by Hart and Wyman (l) on the decor-..position of analogous

acyl peroxides .

as as as
R-(|3-N=N-<|3-R ———-> 2 R-clz. + N2 (21)
ON ON on

When R was cyclobutyl, cyclopentyl, cyclohexyl or
isopropyl, the decouposition proceeded at approxiately
the same rate; when R was cyclopropyl, a 15-fold increase
in rate was obtained. This acceleration was ascribed to

resonance stabilization of the radical produced.

0
CH3 ..CH3 H3
6;. <———-> i <————) — H etc.
' \
CN 01! ° CN
(22)
Um Ortunately, products were not determined and the
qllestion of rearrangement was unanswered. However, in the
pnc>1:olytic chlorination of :2ethylcycloprOpane, much
n‘ck‘lloro-l-butene was produced along with the expected
CYC lopropylcarbinyl chloride, indicating that the
\ ‘I 1 O I O
CJQ 10propvicarblnyl radical does rearrange considerably
(52) . During the course of the present work, Roberts and

C‘
uChuster (33) noted that in the photolysis of cyclo-

15

propaneacetaldehyde, the cycLOpropylcarbinyl radical

rearranged to produce exclusively l-butene.

I
F HagH —h—‘—’—a CH2L‘CHCH20H3 + co (25)

Notably absent from both reactions was the cyciobutyl

derivative, which indicates that an equilibrium similar

to that in the ionic series does not exist.

The fast rate for the decouposition of cyclopropane-

acetyl peroxide was approximately 400 times larger than

the other cycloalkaneacetyl peroxides; the amount of ester

produced also was abnormally high. accounting for 85p of
the peroxide, as Opposed to 30% for cyclohexaneacetyl

peroxide. Later, Lau (34), who carefully washed the

peroxide witi aqueous sodiuh carbonate, stated that the

.rate of decomposition was less than that reported by Wyman,

lout still ten times faster than the henologous peroxides.

8

lfilese high ester yields and rapid rate: indicate
pcotential ionic node of decomposition, such as that for
p-nnethoxy-p'-nitrobenzoyl peroxide and phenylacetyl

iDeImoxide, where high ester yields and rapid rates were

iobtxained when the peroxides were deco posed in the

prwasence of acids. Possibly, the cycloprOpaneacetyl

IDeIWlxide used for earlier kinetics and products
exmniriments was contaminated with cyclopropaneacetic acid.

:[t h”is therefore of interest to re-investigate the kinetics

a . .
“d EDroducts of deconposltion.

16
The acyl peroxides used in this work were prepared
by treating the corresponding acid chloride with sodium

peroxide in anhydrous ether. CycloprOpaneacetyl and

2 R-BCl + NazOz ———$ R-SOOg-R + 2 NaCl (24)
cyclohexaneacetyl peroxides, which were liquids at roo;
temperature. were purified by crystallization fron pentane
at dry—ice tenperatures; the other peroxides were
crystallized from pentane or hexane. The purities of the
peroxides, as determined by titration of liberated iodine,
were 85—100fl (see page 67).

The kinetics experiments were performed in dilute
guirified (35) carbon tetrachloride solutions, the
ccn1centration of peroxide ranging from 0.02-0.09 N. Most
of‘ the decompositions were followed by measuring the rate
of' disappearance of the 5.62/4 peroxide band in the
iniirared. The remainder were followed by determining,
usiJIg the method of Silbert and Swern (36), the amount of
lexiine liberated upon treatment of the peroxide solution
Witki sodium iodide in acetic acid. When the spectrOphoto-
metrtic method was used, the rate constants were obtained
froi. the slope of the equation listed on page 87; in the
titPjJfietPiC procedures, the log of the titer was plotted
direcrtly against time, and tJG rate constants were
Calculzated fro; the lepe of the resulting line.

CEWBIOpropaneacetyl peroxide was decomposed between 44°

l?
and 56°; as can be seen in Table 2, the rate d ubles for
approximately every 6° increase in temperature. For
comparison, the data for cyclohexaneacetyl peroxide are
also inc uded; the increase in rate with increasing
temperature is larger for the for er peroxide than for
the latter. The rate constants, ueasured to about 50%

TABLE 2

THE EFFECT OF TEMPERATURE UPON THE DECOMPOSITION OF CYCLO-
ALKANEACETYL PEROXIDES

 

RatesConstant Data From

 

Peroxide Temp.° x 10 , see."1 Table
Cyclopropaneacetyl 4%.0 5.50 14
44.0 5.07 15
4i-8 ll.i 1—
50.8 12.4 17
50.8 13.0 18
50.8 12.3 :9
56.7 25.8 20
56.? 25.7 23
nyzlohexaneacetyl un,4 0.308 35
64.3 1.14 57
I1-8 2.95 39

deecxaposition of tae peroxide and extrapolated to 25°,

werwa a factor of 103 smaller than those reported by Hart

fl

“an, and sl-ghtiy larger than those of Lau, bit the

has cycloprOpaneacetyl peroxide deco poses

C)
0
cf.
CT

80‘ " ‘, O F. 4-_.i _‘ "- w ~ ‘ ‘ q .
‘PPCCLiabl, faster ufldd cyclonexaneacetyl peroxide is

a. 0
LL.)

confirmed. The validity of these results is substantiated
by the reproducibility of the rate constant -n a nu ber
of runs from various batches of peroxide. To ensure that
systematic errors were not involved using the infrared
technique, titrimetric analysis was occasionally e ployed.

These results are listed in Table 3. The data are in good

TABLE 3

RATE CONSTANTS FOR THE DECOMPOSITION OF CYCLOPROPANEACETYL
PEROXIDE DETERMINED BY INFRARED AND TITRIMETRIC TECHNIQUES

 

 

Analytical Rate Constant Data From
Run Method Temp.° x 105, sec.'1 Table
l. Iodometric 44.5 5.07 27
2. IOdO$€tPIC 44.5 4.93 26
Infrared 44.5 5.03 26
3. Infrared 44.0 5.30 14

 

aggreenent with each other; this is especially evident
wkuen both methods were run concurrently. Because the
rerte constants were lower than previously claiued, it was
necuessary to recalculate the energetics of the reaction.
The: enthalpy of activation was 24.3 kcal./uole, or
appuaoximately 2 kcal./nole less than that reported for the
(H3181? cycloalkaneacetyl peroxides. The activation entropy
was 3.1 cal./deg. hole. The enthalpy of activation
obtaiJued from the meager decomposition data for cyclo-
hexarueaeetyl peroxide (26 kcal./nole) was in good

agreenuent with the 25.7 kcal./mole reported previously (1).

Jog 1C/Tempu

l9

 

 

 

 

 

~1"o].P-___——mT vhf-YW— “IT- .__._ - r h ' —e-..1...____r_._-_--. r".—' “T" 7” V l ' I _—T_-
‘1‘02— "
l
-4.3*' .1
i
\\“\\ f
-4.4r ‘ _\ _;
x“ )
)- ~r\ ‘
t
4.5— i
‘1’
‘406— —a
G)
i \q 1
-40 “\‘ "Ii
.\ f
\- 1
I
i
~4.8 i
.\‘
L \\ _.
‘4.9_ _.
r \x

~5.0P ©\. 5
i

-501k ‘ 1 J J L L 1 l i L L 1 1 a

3.05 3.10 3.15

(l/Temp” 0K) x 103

:F'igure l. Arrhenius Plot for the Decomposition of Cyclo-
IDPC’pé‘lhoeacetyl Peroxide.

.mopSpmpooEoB moospm>
.N ohswfim

pm seasoned impoommcmdosooflomo.go coapflwoosOomQ one now mm>mso spam soosonpmsam

.

.CHE awEHB
00H 03H ONH OOH ow

11114 W ) (#1 . - ... 4

CNN 00m ‘WQH

_ at _ (.1,

 

3

®

20

 

omm.o:
ooo.o+
ooz.o+
oom.o:

 

 

‘ . c L . h

.mcmse moan
.mcmna moHn
.mcmpB mean
.mzth meat

commoq

moH
moH
mofl
moH

«0N.wm
Howaom
«0&5:
«00.3:

<5moc2

 

 

I
u 11?

;-) SCI

('3
.30

Lb

IZBJ

2.
To determine the extent to which induced

decomposition affected the rate constants, the

decomposition was studied Ln the presence of iodine, a

radical scavenger. The results in Table 4 show that the

TABLE 4

EFFECT OF IODINE ON THE DECOMPOSITION RATE OF
CYCLOALKANEACETYL PEROXIDES

h;

 

 

0 Rate Data
Peroxide Temp. Normality Congtant_ Fran
Peroxide Iodine x 10, sec?’ Table
Cyclopropane- 56.5 0.09 .... 2.78 22
acetyl ‘
56.5 0.09 0.07 2.94 23
56.5 0.05 0.05 2.80 24
Cyclohexane- 64.3 0.06 .... 1.14 37
acetyl
64.; 0.06 0.18 1.23 38

addition of iodine does not decrease the rate, as would
be eXpected if there were induced decomposition; indeed,
a slight increase in the rate constant was observed.
Since a slight increase was also obtained with cyclo—
Ymexaneacetyl peroxide, tne possibility remains that
interactions of the carbonyl group with the iodine caused
an increase in the transmission at the wave-length used
(5.6274); iodine alone in carbon tetrachloride does not
absorb in that region. Additional evidence for the
absence of induced decomposition in the concentration
range used (0.02-0.09 N) is the invariability of the rate

constant with the initial peroxide concentration (see

22

Tatfle 5). At 0.18 N, the rate constant was appreciably

TABLE 5

TDHE EFFECT OF CONCENTRATION UPON THE RATE OF DECOMPOSITION
(3F CYCLOPROPANEACETYL PEROXIDE IN CARBON TETRACHLORIDE

Normality of Sefiate Constant Data From

 

TEnqu» Peroxide x 105, see."1 Table
49.8 ' 0.018 11.1 16
50.8 0.046 12.4 17
50.63 0.055 15.0 18
50-8 0.092 12.3 19
44.5 0.06 5.50 11;
44.5 0.186 (a) 29

‘

(£1). Initially very rapid, but approaches 5 x loussecf1

as concentration approacnes 0.09 N.

latrger, presumably because of induced decomposition. As
true reaction proceeded, however, the rate constant
{gruadually approacned those obtained at lower concentrations.
It; can be stated with certainty that the deco position, at
tdle concentrations used, is free of induced decenposition.
The decomposition of cyclopropaneacetyl peroxide then
is indeed faster than the higher nembers of the series,
anfii these rapid rates are a reflection of the lower
activation energy. If the reaction is free radical, and
not; ionic, there must be so_e particular stability
assOciated with the cyclOpropylcarbinyl radical. A dis—
tinCtion between these mechanisms could be made by

detkernnining whether the peroxide initiates the

23
g3olynerization of styrene. Two equal portions of styrene
were placed in separate flasks, one containing cyclo-
propaneacetyl peroxide; the flasks were heated to 50-700
and cooled occasionally to check the viscosity and clarity
(If the reaction nixtures. The styrene containing the
gueroxide was somewhat more viscous and Opaque throughout
tflie observation period. The ability of cyclOpropaneacetyl
EDeroxide to initiate polymerization, though sluggish,
sfinows that the decomposition proceeds at least in part
saith the production of radicals.

The possibility still remains that the decomposition

salso proceeds ionically, as in the case of phenylacetyl
10er0xide, which decomposes simultaneously by both
rnechanisms (24). Acid catalysis may have been the cause
of‘ earlier erratic results with cyclopropaneacetyl
'pearoxide; if any unreacted acid chloride remained in the
EDIWEparation of the peroxide and were incompletely removed
‘JEMDn hydrolysis, erratic results could have been obtained,
fTDI‘ peroxides capable of decomposing by ionic mechanisms
area sensitive to the presence of acid (18). Alternatively,
true reaction may have been autocatalyzed by the cyclo—
PI"DLJaneacetic acid formed during the decomposition. To
explore this possibility, the decomposition was carried
OLVC .in tne presence of several acids of varying acidity.

As is shown in Table 6, the weak acids acetic and

linethylacetic had little effect upon the rate of

o.
(D
O
9

:gposition of cyclopropaneacetyl peroxide. Although

 

.mummHSmoagu we coaumexnzoo asap :6 no popoaaxopaqm

 

24

 

no: osmomsoo mesh 03% .oapmnpo one: mpcaom .:Haooapnosocoa oosoflaom .Amv
@00.0
H: x:.: capoomopoflcoahe 00.0 :0
» :H.H ..... 00.0 :0 meosofioho
mdo.o
mm “av Ma oaoooooaoaeoaae Hmo.o m.:a
0m0.0
am “my mm oaooomoaoHcoaae Hoo.o m.:e
H00.0
on “av ow-mo oaoooooaoHeoaaB amo.o m.aa
m:0.0
an #0.: oaooooaszooaaae 0:0.0 m.::
om m0.e aao.o .oaoooa mao.o m.::
ma eo.m ..... oo.o m.aa aaaoaaaaaao
magma aa.oom .mofi x paod mo moaxosom go 0.0:09 m
eonm mama pompmcoo mpmm huaamsmoz hufla ssoz

 

exwoommomv .mmonommm
qwamoemzaxaaoqoso o onaHmomzoomo mme 20m: mmHoq mo Homm m mun
Lt.

o mamas

25
cyclopropaneacetic acid was not studied, a 00mparable
effect would be likely. Catalysis by carboxylic acids
does not seem to be the cause of the erratic rates
reported. The stronger acid, trichloroacetic acid,
however, did accelerate the rate, the increase being
proportional to the initial concentration of acid. For
equivalent amounts of peroxide and trichloroacetic acid,
the increase was a factor of l2—l5. The rate effect was
considerably larger than that obtained with cyclohexane-
acetyl peroxide. Since the decomposition of benzoyl
peroxide is unaffected by all but the strong mineral acids
(18), the increase in rate for cyclopropaneacetyl peroxide
(could be the result of ring protonation. If ring
garotonation were the cause of rate enhancement, g3—methyl-

lf-butyrolactone would be the expected decomposition

gxroduct (equation 25). Absence of the lactone band in the

9 ‘ CH3 0 ‘ . EB
= H+ I ” , \_\_
[>01ch 2 ——— > (IIH-CHCHzCOOCCHz/ —->( [-0 + 002 + CH2<]
CH269 v. O (25)

ijjiirered is evidence that ring protonation is not the
081188 of the extremely rapid rates. The products are,
iuovuever, appreciably different from those obtained fr0u
the: spontaneous decomposition, especially in the formation
0f easters of trichloroacet;c acid. This may be due to an
ioriicz mechanis and will be discussed later.

Since the presence of acid is not the cause of the

31841 rates for the decomposition of cyclOprOpaneacetyl

(j \

2
peroxide, the possibility of catalysis by-dissolved base,
which would result from the incomplete removal of the
sodiuu carbonate or sodium peroxide, was examined in an
effort to reconcile these results with those previously
reported. Because of the polarization of the cyclopropyl-

carbinyl group, a reasonable mechanism can be drawn for

basic catalysis (equation 26). To determine the extent

9, 9 Bjorn-Q
BCHaCOOCCHaQ —-B—°>[>CH:.;--<:\QJO iczo
I,/

(26)
OCH2<)
F I
;>CH2/C\O_ + 002 —-> DCH2002CH2<1 + 002 + B:

595

of‘this, the decomposition was run after washing the
goeroxide solution in carbon tetrachloride with 10% aqueous
isodium carbonate and in the presence of pyridine. The
Inesults are listed in Table 7. For comparison cyclo-
heaxaneacetyl peroxide also was deCOmposed in the presence
of‘ pyridine.

The results snow that the presence of neither
iricxrganic nor org anzic base inexplicably altered the rate
comlstant. Washing of the peroxide sol t on with aqueous
Sociituxcarbonate before had no effect on the decomposition
rains, Although pyridine accelerated the rate, the increase
was comparable to that resulting frou the treatment of
CYCJJDhexaneacetyl peroxide with pyridine. Tertiary amines
are cnown to react with acyl peroxides (37) as is

illnustrated in equation 27. The use of pyridine has also

2/
TABLE Y

THE EFFECT OF BASE UPON THE DECOMPOSITION 0F
CYCLOALKANEACETYL PEROXIDES, (RCH2C02)2

 

 

“ Rate Data
Normality Constant_ From
R Temp? Peroxide Pyridine x 10? sec.1 Table
Cyclopropyl 44.5 0.06 .... 5.07 15
44.5 (L05 0.03 6.86 32
M5 0.06 (a) 5.30 28
Cyclohexyl 71 8 0.06 .... 2.93 39
71.8 0.06 0.056 4.8} 40

(a). After washing the peroxide solution with sodium
carbonate and drying.

been reported; although the rate of reaction is increased,

0.

q e>0 -
R3N: + [00C 2 —*R3NOB©+ ©|CO ——7 Products (27)

the nature of the products is unknown (38). It would be
expected that if cyclopropaneacetyl peroxide were
Enasceptible to basic catalysis, the effect would be
(zonsiderably larger than obtained. The relative ease of
Checomposition of this peroxide then cannot be attributed
tC) basic catalysis.

The cause of earlier high rates is not clear. The
Vfixrk reported in this thesis shows that the decomposition
5&3 catalyzed neither by carboxylic acids initially present
OI‘ formed during the react on nor by bases. They nay have

tNNBn due to other impurities, such as unhydrolyzed acid

28
chloride, which, on occasion, was present in peroxides
used in this work and gave rapid rates.

Earlier product studies showed that ester (855) and
Carbon dioxide (85%) were the major products; alkyl
halides resulthg from the reaction of an alkyl radical
witn the solvent were not found (1). The absence of alkyl
chloride indicates that the cyclopropylcarbinyl radical,
if formed, reacts in other ways than by abstraction of a
chloride atom frOu the carbon tetrachloride. Since the
decomposition rates reported here are substantially lower
than those reported by Wyman, it was ihportant to verify
the product composition, with emphasis on the ester
produced.

The major products of the decomposition of cyclo-
propaneacetyl peroxide in carbon tetrachloride were
carbOn dioxide and an ester, probably cyclopropylcarbinyl

cyclOpropaneacetate; in addition, there was a small amount

F5)

0 cycloprOpaneacetic acid. The average results from
several experiments are shown in Table 8. Carbon dioxide
ivas determined graviuetrically; the ester and the acid
vuare determined spectrOphotonetrically sing respectively
tflle 5.75fk and the 5'85f* bands in the infrased. Together,
tfliese accounted for 37.5% of the initial carbonyl group,
vdiile the acid and the ester contained 59.5% of the alkyl
EITJUp initially present in the peroxide. Approximately

10 Inl. of carbon tetrachloride sol~tion was distilled

dirwactly fro; the deCOmposed peroxide solution; l-butene,

23
TABLE 8
PRODUCTS 0F DECOMPOSITION OE CYCLOPROPANEACETYL PEROXIDE IN

[-—

CARBON TETRACHLORIDE AFTER THREE HOURS AT 7&0 (2.45 smile;
of peroxide in 50 ml. of carbon tetrachloride)

 

 

\
_§roduct mwoles. mmoles/mmole peroxide
Carbon dioxide 3.24 1.32
Ester (e - 489 l/mole-cm.) 7.37 .56
Acid (6 ' 53;? l/z..ole-cm.) .l'] .07
.1?

Alkyl chloride (8)

(a). It is not known whether this is actually an alkyl
chloride or a diaikyl.

if formed, would appear in this fraction. Because of the
low concentration and the absence of an appreciable

olefinic peak in the infrared, the presence of small amounts
of pentane, which was used in the preparation of the
peroxide, could not be excluded. The remainder of the
carbon tetrachloride solution was distilled under reduced
gir ssure to remove higher boiling materials; redistillation
of uOSt of the solvent (75 hi.) through a packed column
affforded fractions whose infrared spectra possessed
ixlsignificant carbon-hydrogen absorption peaks. he
SPHactrum of the remaining carbon tetrachloride had a
Struong carbon-hydrogen peak, which, after correction for

a Ernall amount of ester present, accounted for 8.5fi of the
initial peroxide; whether an alkyl chloride or a dialkyl
W355 present was not determined. The remainder of the

Cyclopropaneacetyl peroxide was probably tar, which, along

30
with hexachloroetnane, was undetermined.

The results verify the higher ester and lower carbon
dioxide yields than expected for the decomposition of acyl
Peroxides, although a considerable difference still was
obtained from the results of Hart and Wyman (85% vs. 56$
for ester and 85% vs. 66% for carbon dioxide). For
comparison, the yields of ester, carbon d;oxide and alkyl
chloride for peroxides derived fron other alicyclic acids

are shown in Table 9. Yields of 75-80% of alkyl chloride

TABLE 9 '

PRODUCTS FROM THE DECOMPOSITION OF CERTAIN ACYL PEROXIDES
IN CARBON TETRACHLORIDE

 

 

a Carbon a Alkyl
Peroxide . Temp? Ester Dioxide Chloride
Cyclopropaneacetyl b 78 0.56 1.32 0.17
Cyclopropaneformyl C 70 0°16 1°61 1'33
Cyclohexaneformyl C 70 0.23 1~57 1'53
Cyclohexaneacetyl C 70 0.30 1.77 1-32
Benzoyl C 70 0.26 1.72 1.27

—*

b This work .

{£3§. Moles produced per mole of peroxide deconposed.
c . Ref. 1.

‘

Nexus obtained for the other peroxides, whereas in the
decnomposition of cycloprOpaneacetyl peroxide, the alkyl
chlxoride, if formed at all, was a minor product.

The ester obtained is probably cyclopropylcarbinyl

CyffiLOprOpaneacetate, since the infrared spectrum was

31
similar to that of authentic cyclOpropylcarbinyl cyc_o-
propaneacetate and passage of the higher boiling materials
from the decomposition through a vapor-phase chrouatograph
showed the existence of almost entirely one product. The
production of ester is a geminate reaction and is probably

explained by equation 28.

. H g
‘>-CH2EO 2 —-) EDCHzCO ' —> I>CH20 CHz<l + 002

(28)

The immediate recombination of the radicals within the
solvent cage has been demonstrated by KharaggLand
co-workers (39), who obtained from optically active
cx-Lethylbutyryl peroxide an ester in which both the acid
and the alcohol portions retained their Optical activity.
as P.
(CH30H20H002 )2 -+ CHSCHzCHOCCHCHacHa (29)
CH3 CH3
ITIe larger quantity of ester produced frOm the deconposi-
tjxon of cycloprOpaneacetyl peroxide shows that the reaction
bertween the alkyl and the acyloxy radicals occurs rapidly,
bexfore reaction with the solvent can occur. The small
anuJunt of acid present could result from abstraction of a
hyxirogen atOm within the solvent cage (27) or from reaction
0f tile acyloxy radical with a small amount of water present
(no).
The question of acceleration of the rate by trichloro-

acetix: acid being due to attack at the ring rather than

32
carbonyl group demanded a study of the decomposition
products under these conditions. Since cyclopropane
derivatives are sensitive to ring opening reactions by
strong acids (18), the deconposition Right be expected to
proceed either by proton attack at the ring or the carbonyl

group. The products obtained from the decomposition

l: H2gOJ2 33 [Dcflz-EilOOQCHafl] —->
{>~CH2@OH + C02 + [>CH239

varied considerably from those of the spontaneous
decomposition; along with less carbon dioxide, the type
and quantity of ester were altered. The products are

listed in Table 10. The hechanism chosen must explain the

TABLE 10

IHKODUCTS OF DECOMPOSITION OF CYCLOPROPANEACETYL PEROXIDE

JEN CARBON TETRACHLORIDE IN THE PRESENCE OF TRICHLOROACETIC
AKZID, 76°, 3 HOURS (2.5 mmoles of peroxide and 5.00 mmoles
(If trichloroacetic acid in 40 m1. of carbon tetrachloride)

 

Mmoles per

 

PIIOdUCt Mmoles mmole peroxide
Carbon dioxide 3.06 1.22 (a)

BL ter (5-76/*2 E = 489 l/uole-cn.) 0.74 0.30

Esiuer (5-67/A: E = 525 l/mole—cm.) 1.44 0.58

scixi not determined

 

k

(a). .Average of two runs.

 

 

N-
_. -.....

33
decrease in the carbon dioxide and tne ester (5.76/A)
produced, as well as the appearance of a material
absorbing at 5.67/4.

Formation OT'P-Nethyl- X-butyrolactone was excluded,
since the action of concentrated sulfuric acid on cyclo-
prepaneacetic acid produced a substance which absorbed
at 5.63;L, the same absorption peak possessed by

X—butyrolactone. The presence of the 5.67;“~ band was
attributed to an ester of trichloroacetic acid, since
cyclopropylcarbinyl trichloroacetate possessed the same
peak. The reaction then must be described by equation
30, with the cycloprOpylcarbinyl cation resulting as a

trichloroacetate ester. The extent of rearrangement was

not determined. Such a mechanism would also explain a

decrease in the yield of carbon dioxide. The ester
aabsorbing at 5.76/L was probably a cyclopropaneacetate;
the nature of the alkyl group was not determined, but the
(xyclopropylcarbinyl derivative could result from the
ccnicurrent spontaneous reaction.{ The amount of cyclo—

prananeacetic acid is expected to be at least equivalent

to ”the trichloroacetate ester formed. Assuming this value,

90%; of the init;al carboxyl and 70% of the initial alkyl

radicals would be accounted for.

TILE DECOMPOSITION OF OTHER ACYL PEROXIDES.

__

VVith a change in solvents, a change occurs both in

the {Drwoduct cemposition and in the rates of decomposition

34
of acyl peroxides. For example, the amount of carbon
dioxide evolved from the decomposition of benzoyl peroxide
increases in the order olefins (:paraffins (Laromatics<:
carbon tetrachloride. Factors influencing this change
are the relative reactivity of the solvent with the
radical formed and the extent of induced decomposition,
which becomes important at concentrations used to determine
products (41).

The decomposition of trans—4-t-butylcyclohexane-
carboxylyl peroxide in tetrabromoethane produced a quan-
titative yield of trans-4-t—butylcyclohexyl trans-4—t-buty1-
cyclohexanecarboxylate (42), whereas in carbon tetra-
chloride only 47% of presumably the same ester was formed.
'The decrease in yield might be the result of higher
‘temperatures and induced decomposition rather than
ixucreased reactivity with the solvent. Since the radical
txransfer constants of tetrabromoethane was approximated to
'bez slightly larger than that of carbon tetrachloride at
tale same temperature, the effect would be attributed to
tflie difference in reaction temperature (78° for the
deucomposition in carbon tetrachloride vs. 50.70 for that
1r) tetrabromoethane (42)), with induced decomposition by
true trichloromethyl radical playing the greater role.
Heruze, the yield of ester would be smaller.

In the decomposition of cyclopropanecarboxylyl
Perxxxide in carbon tetrachloride, a highly reactive

Solwnent, cyclopropyl chloride was the major product (1)-

35

Decompositions performed in the presence of iodine showed
that most of the cyclopropyl radicals exist outside of the
solvent cage. By using a less reactive solvent, coupling
of the radicals could conceivably be forced, presenting a
new method for the synthesis of dicyclOprOpyl.
Cyclopropanecarboxylyl peroxide was decomposed in
chlorobenzene and in t-butylbenzene, solvents with low
radical transfer constants (#3). In the decomposition in
t-butylbenzene, acid (8%), ester (37%) and presumably
carbon dioxide appeared as the major products, with no
dicyc10propyl being detected. Similar results were
obtained in chlorobenzene as the solvent. In spite of the
low radical transfer constants, the solvents were too
reactive to permit the coupling of the alkyl radicals, the

:najor reaction probably involving attack at the benzene

ring.

56

C. The DeconpjsitionugfwtlButyl Peresters of Some
Alicyclic Acids.

 

 

 

 

In contrast to acyl peroxides, relatively little
work had beenodone on the decomposition of t-butyl
peresters, R-E-o-Ot-Bu, until recently. Their
decompositions parallel those of the corresponding acyl
peroxides but proceed more slowly.

The most extensive work on the mechanism of
decomposition was conducted by Bartlett and co—workers
(2, 5, fl). Considerable evidence was presented by
Bartlett and Hiatt (2) for the importance of carbon-carbon
stretching in the decomposition transition state. Their
results are listed in Table ll. With increasing stability
of the resulting radical, there is a decrease in the heat
of activation and the stability of the perester for a
given series. The peracetate and the perbenzoate are
particularly resistgnt to decomposition because they would
produce the poorly stabilized methyl and phenyl radicals
respectively. A concerted decomposition was further
demonstrated by Bartlett and Ruchhardt (3) in the
decomposition or substituted t-butyl phenylperacetates,
where the rate was decreased by electronegative groups on
the benzene ring and increased by electropositive ones.

These results helped put the concept of a concerted
decomposition of acyl peroxides on a firmer basis.
However, a work relating the two decompositions has not

been reported. As was mentioned previously (p. 12), a

 

37
TABLE 1 l

DECOMPOSITION OF t-BUTYL PERESTERS, RC03t-Bu, IN
CHLOROBENZENE (2)

 

 

Ha lf-life AH* AS*

R Min., 60° kcal./mole cal./deg. mole
(Di-t-butyl peroxide) 107 37.8 l3.8 -
CH3 5 x 105 38 17
C6Hs 3 X 104 33-5 7.8
(Benzoyl peroxide) 6000 32.7 13 3
C6H5CH2 1700 28.7 3.9
013C 9T0 30.1 8.0
(CH3)3C 300 30.6 13
CaH5CH-CHCH2 100 23.5 -5 9
(CeH5)2CH 26 24.3 -l.0
CsH5(CH3)2C 12 26.1 5.8
(C6H5)2CH3C 6 2a.? 3.3
CaH5(CH2-CH)CH n 23.0 -1.1

 

ciifference of 2.5 kcal./mole for the heats of activation
anud a factor of 137 for the rates of decomposition of
czycloprOpanecarboxylyl and cyclohexanecarboxylyl peroxides
were obtained. These were attributed to a decrease in
stnability of the cyclOprOpyl radical relative to the
Cytzlohexyl radical and to some carbon-carbon stretching
Occnarring in the rate-determining step. Bartlett (n4)
had :reported that the decomposition of t—butyl cyclo-
hexanepercarboxylate proceeded at 100° in chlorobenzene a

factsor of 120 times faster than the t-butyl peracetate,

38

with an activation enthalpy of 31.} kcal./ ole and an
activation entropy of 8.6 e.u. It was then a matter of
interest to compare the relative sensitivity of the
peroxide and perester decompositions to the same change
in the structure of the alkyl group.

The t-butyl peresters of cyclopropanecarboxylic and
cyclohexanecarboxylic acids were prepared by the reaction
Of t-butyl hydroperoxide in a pyridine-ether solution with
the apprOpriaté acid chloride and were purified by passing
through a chromatographic column containing Florisil,
using pentane as the eluent. Removal of the pentane
afforded t-butyl cyclohexanepercarboxylate of 77% purity
and t-butyl cyclopropanepercarboxylate of 76% purity.
Portions of a solution of the peroxide in carbon tetra—
chloride were sealed in ampoules and placed in a constant
‘temperature bath. Individual samples were removed
19eriodically and frozen in dry-ice until use. Transmissions
(Jf the melted samples at 5.62/A were determined and the
Iuate constants and heats of activation determined in a
nuanner similar to those of the acyl peroxides (p. 87).
IT“; rate data and the energetics for the decomposition are
Shown in Tables 12 and 1}.

From the work of Bartlett and Hiatt and of Hart and
thnari, these results are to be expected. A comparison of
the IDeresters with the corresponding acyl peroxides in
carWNDH tetrachloride is shown in Table l}. The t—butyl

peresrters are definitely more stable than the

\Q

3
TABLE 12

THE DECOMPOSITION OF t-BUTYL CYCLOALKANEPERCARBOXYLATE IN
CARBON TETRACHLORIDE

 

 

 

o Cyclopgopage Data From 0 Cyclohgxans Data From
Temp. k x 10,sec. Table Temp. k x 10,sec. Table
110 6.86 48 90 2.35 42
110 6.68 49 90 2.33 43
120 20.? 50' 100 9.48 44
120 20.2 51 100 9.28 45
129 48.3 52 110 20.0 46
129 49.8 53 110 22.0 47
TABLE 13

COMPARISON OF THE DECOMPOSITION OF t-BUTYL PERESTERS AND
ACYL PEROXIDES

 

 

 

‘ Half-life A11" __ A53 , cal.

__ sec., 70° kcal.mole 1 degglmole

(Syclopropanecarboxylyl 1.41 x 105 28.3 ' -5.7
Peroxide (a)

(Zyclohexanecarboxylyl 1.03 x 103 26.1 -3.l
Peroxide (a)

tiesutyi Cyclopropane- 1.2 x 106(b) 30.7 2.1
percarboxylate

t-qsutyl Cyclohexane- 2.4 x 105(b) 28.7 -O.6
percarboxylate

(a; . Ref. 40.

b . Extrapolated from higher temperatures.

 

 

log k/TemP .‘

40

 

 

 

 

 

 

T
1’
'5-4[

2.61 2.63 2.65 2.67.3. 2.69 2.71 2.73 2.75

Figure 3 .
Cyc thexa ne

_ 1..-.--__1_. _ __.| 1 L 1 BL L i

(l/Temp, 0K) x 103

Arrhenius Plot for the Decomposition of t-Butyl
percarboxylate.

 

41

...Hog.O:mMocoHo%o

.mmpzpmsozsme ozoapm> pm oumflzxooemo
mapsmlp @O CCWQHGOLEOOQQ 05$ QO% MO>LSD mpwm %@UQOIQwHHh .# mfififlflh

. CH; a 0:5?

00

 

 

 

0mm 00m owa owfi Oi: Omfl OOH ow
‘1 3 d J . l a q 1fl 1 4 4 _ q 4
a
i /
am
e
i.
I.
j
7
V//.
. x - 0,
d AU
ooflfl .o
r OOOH .m
com .<
IILIIIDPI . 1. -- F1 ....1. -.s -... .-..e\-Ju.l Ll if _

 

11414 1 _

 

 

 

 

So;-) 301

('sueal

42

 

5.91 1

6.0L 3

6.11 , 1

l ._ .
--...r 1-

 

T
/
...---__.L_ --.. L--__

O\
\N
I

 

10g k/TemP.

O\

.1;
T

/

ON
01
I

 

 

 

 

 

 

H— 1 L L 1 1 I - Li L.-..___._J__.____J______J__

l_ L
2’50 2055 - 2060
(l/Temp., ox) x 103

Figure 5. Arrhenius Plot for the Decomposition of
t-Butyl Cyclopropanepercarboxyla te .

.mopzpmpoesme

 

 

 

mSoHpm> pm

 

 

 

 

mpmahxonsmopmamcmoosaofione Hzpzmnp mo coauHmOQEoomm 02p mom mm>n50 mpmm smpnoupmpam .m mpswfim
.CHE “mEHB
00m owfi oma Oda omH OOH ow 00 0: ON
jIIJIII-Jtll 4 111111 .-.4111- .1 -- 11-11-4111! 6 .4. 11.11.1- 11111411111 AU.- 1- s 1 . -- .41 . _ .1114
r /// -
1 . /.// .1
// ,
/,/ /. '
T K/ / 1
f // f... 1, 1
T w
z
1 m x
72
.4/ .1
h.-
T
T @.I
/...
T
r omma .o
oONH .m
- ooHH .¢
IIIIL|||1|L1.1 . 11.111; . _ p 111.».11 . r Ll » 111.1 .111 s L . L

 

om...

me.-

 

1111

4-

corresponding acvl peroxides. AU 70°, t-butyl cyclo-
propanepercarboxy ate would decompose at about one-eighth
the rate for cyclopropanecarboxylyl peroxide; the effect
is even greater for the cyclohexane derivative. Although
the enthalpy Of activation of t-butyl cyclopropaneper-
carboxylate is 2 kca1./m01e larger than that for t-butyl
cyclohexanepercarboxylate, the inversion of entrOpies
partially nullifies the expected differences in decomposi-
tion rates.

That the differences in the decomposition rates of
t-butyl peresters is not the same as for the correSponding
acyl peroxides is not extraordinary. Whereas benzoy:
peroxide and acetyl peroxide decompose at essentially the
same rate and, within experimental error, possess the same
energy of activation (45), t-butyl peracetate decomposes
approximately 16 times slower than t-Ontyl perbenzoate
(stle 10). Also, phenylacetyl peroxide deco poses 8
tines faster at 0° than benzoyl peroxide at 80°, but the
ciifference for the correSponding t-butyl peresters at 60°
ifs only a factor of four. Although the genera; effects
OPHerative in acyl peroxide deCOmpositions are present in
true decouposition of t-butyl peresters. their magnitudes
are not comparable .

The decrease of the heat of activation in going frog
a tirree- to a six-membered ring, in conjunction with the

Work of Hart and Wyman (1.), is evidence that there is some

Carbmni-carbon stretching in the transition state of the

45
decomposition of t—butyl peresters. The explanation of
I-strain they offer for the decomposition of acyl peroxides
is also applicable here. In the cyclohexane derivative,
the bond angles are 109°and the carbon atoms possess
tetrahedral configurations. As the carbon-carbon bond
stretches, the carbon involved becomes trigonal and
expands its bond angles to 120°, thereby introducing
strain into the ring. In the case of the cy010propane
derivative, however, the strain produced is of more
importance. Using the conventional bond angles of 109°
for saturated carbon atoms, there is an existing strain
of 49° in the molecule. Upon transformation into the
cyclopropyl radical, the bond angles cannot expand as they
do in cyclohexane and the internal strain would increase

to 60°, which would impede carbon-carbon stretching.

46

SUMMARY

1. A reliable procedure for the synthesis of cyclo-
prOpaneacetic acid on a moderately large scale has been
develOped. Cyclopropyl chloride gave a good yield of
cyclopropyllithium, using ether as the solvent; treatment
with ethylene oxide gave fi-cyclopropylethanol in 66713
yield. Cyclopropylnagnesium chloride can be prepared,
using benzyl bromide as an entraining agent and tetra-
hydrofuran as solvent; it also gave fi—cycloprOpylethanol
when treated with ethylene oxide, but the overall yield
was lower and the reaction less clean than with the
lithium reagent. B-circiopropyiethanoi can be oxidized by
chromium trioxide and sulfuric acid to cyclopropaneacetic
acid in 57% yield.

2. The decomposition of cyclopropaneacetyl peroxide
;proceeds by first-order kinetics, slower than previously
:reported (1), but still 18 times faster than other cyclo-
aalkaneacetyl peroxides. The reaction, at the concentrations
used, (0.02-0.09 N), is free of induced decomposition,
Wiflll initiate the polymerization of styrene and results in
thus production of large amounts of ester, with low yields
Of‘ carbon dioxide. Also produced in the reaction were a
Snmtll amount of acid and an unidentified material, either
an zalkyl chloride or a hydrocarbon from the coupling of
tWC> alkyl groups. Whereas the carboxyl group was quan-

titertively accounted for, only three-quarters of the

47
cyclopropylcarbinyl radical was detected; the remainder
escaped detection, possibly as tar.

5. The decomposition of cyclOpropaneacetyl peroxide
is insensitive to weak acids such as acetic or trimethyl-
acetic acid. In the presence of trichloroacetic acid,
however, there is an acceleration in the rate of decompo-
sition, the increase being proportional to the concentra-
tion of acid. A similar effect, though not so large, was
Obtained in the decomposition of cyclohexaneacetyl
peroxide. This action results in a decrease in the yield
of carbon dioxide and cyOIOpropylcarbinyl cyclopropane-
acetate, as well as in the formation of an ester of
trichloroacetic acid, the nature of the alkyl group not
being determined.

4.’ Pyridine also accelerates the decomposition of
cycloprOpaneacetyl peroxide. Cyclohexaneacetyl peroxide
behaved similarly; the effect, however, was not large.

5. t-Butyl cyclohexanepercarboxylate decomposed
tflxree times faster than t-butyl cyclopropanepercarboxylate
irl carbon tetrachloride at 110°. The enthalpies of
acrtivation for the reactions were 28.7 and 30.7 kca1./mole
I'EBSpectively.

6. An improved procedure was developed for the
preparation of cyclopropanecarboxaldehyde. Cyclopropyl-
carfiainol was catalytically dehydrogenated over a copper—

Zinc: catalyst at 300° to give the aldehyde in 81% yield.

EXPERIMENTAL
I. THE PREPARATION OF CYCLOPROPANEACETIC ACID

A. Displacement Reaction of Cyclopropylcarbinyl
Benzenesulfonate with Potassium Cyanide.

Cyclopropylcarbinyl benzenesulfonate was prepared
according to the procedure of Bergstrom and Siegel (46).
A mixture of 53 ml. of 2,4,6-collidine and 15.8 g.
(0.22 moles) of cyclopropylcarbinol was cooled to -5° and
58.7 g. (0.22 moles) of benzenesulfonyl chloride was added
at such a rate that the temperature was maintained between
0-5°. Methylene chloride (50 ml.) was added as solvent.
The temperature was allowed to rise to 12°, where it was
maintained for one hour. The 2,4,6-collidine was slowly
neutralized with 50 ml. of ice-cold 10 N sulfuric acid,
so that the temperature did not rise. The resulting layers
were separated and the water layer was extracted with
several portions of methylene chloride. The combined
extracts were then washed several thes with cold 2.5 N
Stflfuric acid.

After removing the methylene chloride, 40 ml. of
water and 50 g. of potassium cyanide were added to the
resxidue and the mixture was stirred for several days at
rocnn temperature. Sufficient water was then added to
diseuolve the inorganic salts and the mixture was extracted
With ether until- the extracts were colorless. The
eXtPalate were combined and dried over potassium carbonate.

After; the drying agent was filtered, the ether was removed

119

on a steam bath and the remaining liquid distilled under
reduced pressure. One fraction, boiling from 43-63°/20 mm.,
was collected, the remainder being tar and high boiling
materials. The infrared of the distillate indicated a
preponderance of an isonitrile, along with small amounts
of nitrile, alcohol and olefin.

Refluxing this mixture with aqueous potassiun
hydroxide, followed by acidification, afforded a small

amount of crude acid, which was not identified (Figure 7).

B. Attempted Condensation of Cyclopropanecarboxa1dehyde
33th Rhodanine.

l. The Preparation of a COpper-Zinc Catalyst.

The method used was essentially that of Fenske and
Hart (47). Cupric nitrate trihydrate (1044 g., 3.54 moles)
-and zinc nitrate hexahydrate (136 g., 0.56 moles) were
dissolved in one liter of distilled water. The solution
iuas heated to 70°, filtered to remove insoluble particles
£n1d treated at 70° with concentrated ammonium hydroxide.
Ikefore an excess of ammonium hydroxide was added, the
prwecipitated hydroxides were filtered. The addition was
Ccnitinued until precipitation no longer occurred and the
mijcture of hydroxides was filtered. The hydroxides were
Wasfiied with 1.5 liters of water and dried in an oven at
125° for 6 hours. The cake was ground to a powder, which
was (iried at 125° for an additional 24 hours.

.An.eighteen-inch glass column equipped with a

5O

.owflcmzo aduwmmpom cam opmcogfidmmcmNCmm
Hmsannmoflhdopmoaoho mo posoosm coapomom song oocflmpno cso< mo ESonmdm ommeQCH .s madmflm

mcoaoaa CH epwcmfim>m3

i: 9 ms. 2 S a w s m m a m
._ . . J

_ A it
i; i,
a _

 

 

‘1
and
‘

 

 

 

 

 

 

 

 

 

 

 

___/

 

 

 

 

 

 

 

 

 

 

 

 

51
thermocouple and wound with Nichrome wire was packed with
the powdered hydroxides, using a glass wool support. The
column was heated to 325°, and air, previously dried by
phosphorus pentoxide, was passed through until evolution
of oxides of nitrogen ceased. The column was then allowed
to cool to rOOm temperature.

The packed column was flushed with hydrogen while
being gradually heated to 200°. At this temperature,
water was evolved and the temperature rose abruptly to
250°, where it remained until the evolution of water
ceased. The column was then allowed to cool and the flow

of hydrogen stOpped.

2. Dehydrog enation of Cyclopropylcarbinol.

To the packed column was attached a dropping funnel
and a condenser. The column was heated to 300° and 60 g
(0.83 moles) of cyclopropylcarbinol (46) were added drop-
wise. The reaction was followed by measuring the amount
of hydrogen liberated by means of a flowmeter. The yield
of aldehyde (b.p. 100°/744 mm.; n55 1.4265) was 33.1 8.,

81% conversion based on 42 g. of unrecovered alcohol.

3. Attempted Condensation of Cyclopropanecarboxaldehyde
with Rhodanine.

' s 0
H20? /0 1) NaOH H H
H + - *a CH C-COH
[::>E 8—CZH4 [::>CH— —?HQNH 2)‘H® [::> 2
/
st

s (II)

 

(I)

52

O
i II A O l KOH
M DCHzg-COH‘ iii—a DCHzCN 2) H+ {>CH2C02H
HON

(III)

To a solution of 10 g. (0.075 moles) of rhodanine
and 5.3 g. (0.076 moles) of cyclopropanecarboxaldehyde in
25 g. of glacial acetic acid was added 20 g. of freshly
fused sodium acetate. The mixture was refluxed for 35
minutes with occasional stirring. Upon addition of the
mixture to 150 ml. of water, a reddish-black gum,
presumably I, separated. Attempts to crystallize the gum
failed. It was then dissolved in 200 ml. of 2 N sodium
hydroxide and the solution was refluxed for twenty minutes
filtered and allowed to cool. Upon rapid neutralization
with 2 N hydrochloric acid, 7 g. of a rust colored powder,
presumably II, was obtained. Attempts to crystallize this
also failed.

The powder (7 g.) was added to 5.0 g. of hydroxylamine

in ethanol, prepared by mixing 10.4 g. of hydroxylamine

lhydrochloride and 3.45 g. of sodium in 100 ml. of ethanol

euud filtering the sodium chloride formed. The mixture
lwas heated until hydrogen sulfide ceased being evolved;
‘the alcohol was then removed by evaporation under reduced
Ixressure. The residue was dissolved in 100 ml. of 2 N
scxiium hydroxide. After filtration, the solution was
cooled and acidified with 1.5 N hydrochloric acid to a

Congo Red endpoint, yielding 4.5 g. of a yellow

53
precipitate, presumably III.

The oximino-acid (III) was refluxed with 23 g. of
acetic anhydride for twenty minutes and steam-distilled,
250 ml. of distillate being collected. A carbonaceous
residue remained in the reaction flask. The distillate
was neutralized with sodium carbonate and extracted with
three 35 ml. portions of ether, which were combined and
dried over magnesium sulfate. The drying agent was
filtered and the ether evaporated. A small amount of
liquid, whose infrared spectrum contained no nitrile band,

was obtained.

C. The Preparation of$§-Cyclopropylethanol.
1. Using Cyclopropylmagnesium Chloride.

In a two—liter three-necked round-bottOued flask
equipped with stirren dropping funnel and dry-ice condenser
was placed 49.6 g. (2.04 g, at.) of magnesium turnings.
To this was added slowly, with stirring, a mixture of
125 g. (1.63 moles) of cyclopropyl chloride (48) and 50 g.
(0.29 moles) of freshly distilled benzyl bromide in 376 g.
<3f dry tetrahydrofuran. After the addition was complete,
'the mixture was refluxed for 18 hours. Treatment of the
Iaixture with standard acid and back-titration showed that
31.46 moles (76%) of Grignard reagent was present.

To the ice-cold mixture was added dropwise a solution
(If 64.4 g. (1.46 moles) of ethylene oxide in 300 g. of dry

tetrahydrofuran. After the addition was complete, the

54

mixture was refluxed for two hours. It was then allowed
to cool to room temperature, filtered through glass wool
to remove unreacted magnesium, poured onto ice and
acidified with 10% sulfuric acid. After the layers were
separated, the aqueous layer was saturated with sodium
chloride and extracted with three lOO-ml. portions of
ether. The ethereal extracts were combined with the
tetrahydrofuran layer and the solvents were removed by
distillation through a Vigreux column. To the remaining
solution was added 250 ml. of 4 N sodium hydroxide and the
mixture was refluxed for four hours. iAfter cooling, the
layers were separated and the aqueous layer was extracted
with three 50-ml. portions of ether. The extracts were
added to the organic layer and dried over magnesium
sulfate. The ether was removed and the residue was
distilled through a packed column to yield 59 g. (42%) of

fi-cyclopropylethanol, b.p. 73-750/50 mm.; n§° 1.4328.

2. Using CycloprOpyllithium.

The method used for the preparation of cyCIOpropyl-
lithium is that developed by Hart and Holzschuh (9). A
typical preparation is described.

To a flame-dried one-liter three-necked round-
bottomed flask continuously flushed with argon and
equipped with a high-speed stirrer and a reflux condenser
was added 7.3 g. (1.06 g. at.) of. lump lithium and 250 ml.

of nxineral oil, which had been dried by heating over

55

sodium. The flask was heated by seans of a Fisher burner
until the lithium melted, at which time the mixture was
vigorously stirred. when the lithium was thoroughly
dispersed, the heating was discontinued, but the rapid
stirring was maintained for a snort period to prevent the
lithium sand from fusing. After the mixture cooled to
room temperature, host of the mineral oil was removed by
the addition of anhydrous ether in portions, followed by
the application of suction, until 500 ml. of ether had
been used.

To the lithium sand was added 250 ml. of anhydrous
ether. A thermometer and a dropping funnel were attached
to the flask, which was cooled to 2°. A mixture of 46.3 g.
(0.60 moles) of cycloprOpyl chloride in 150 m1. of
anhydrous ether was added slowly so that the temperature
of the reaction mixture never exceeded 10°. nfter the
addition was complete (about three hours), the mixture was
stirred for an additional hour. The cyclOpropyllithium
was then ready for use.

To the ethereal suspension of cycloprOpyllithium was
added a dry-ice-cold mixture of 53.2 g. (1.21 moles) of
ethylene oxide in 300 m1. of anhydrous ether at such a
rate that the temperature of the mixture never exceeded
10°. After the addition was complete, the stirring was
continued for an additional hour at 2°.

The mixture was poured into water and acidified with

ice-«301d lop sulfuric acid. The layers were separated and

56

the aqueous portion was saturated with sodium chloride.
The water layer was then extracted with three 75-n1.
portions of ether; the extracts were combined with the
other layer and dried with anhydrous potassium carbonate.
After filtration, the ether was removed by distillation
through a Vigreux column. Distillation of the r sidue
under reduced pressure afforded 30 g. (66%) off -cyclo-
propylethanol. The infrared spectrum is shown in Figure

8.

D. Oxidation of fig-Cyclopropylethanol.

 

l. with Potassium Permanganate.

In a two-liter three—necked round-bottomed flask
equipped with stirrer, dropping funnel and condenser was
placed 27 g. (0.31 1:03.88) of 8-cyclopropylethanol and a
solution of 7.0 g. of sodium carbonate in 70 ml. of water.
To this was added dropwise, over a period of 2.5 hours, a
solution of 66 g. (0.42 moles) of potassium permanganate
in 1800 ml. of water. After the addition was complete,
the flask was cooled to ice temperature and then stirred
at room temperature for a day.

The precipitate of uanganese dioxide was then
filtered. The filtrate, which was still purple, was
heated to boiling, at which time the color disappeared.
The solution was refiltered and the filtrate was
evaporated to 150 ml. The solution was cooled and washed

witfil 50 ml. of ether. It was then covered with 100 m3. of

57

 

msopOH; ca npmcmdm>m3
i 9. NH 2 S m w a m m is n
fiIllllltiaxilnllllJllirla- lll..|!-il..-.luJ.¥rilr-r---._.l-! .- . (it. 11111 1:4: _ ill—sill- (1.- (I.
_
i i
i_. i. _
m i i
_
i i i
p i . i
i u _
, P
r k / . R \i m
_
m
i _ _
i 6 j _\ i
i 1 i
vi wiltl- Ll: ,-r/.i---_-‘_....ir-. - i - _ _ x , .--KxLlL/V,/.

 

.ocwnofisomppoe 202.80 5- Hocmspmfmaosaofi98..fl mo 15.50on 695295” .m. mhsmfim

 

 

 

 

 

 

 

 

 

58

ether and acidified with 6 N sulfuric acid to the Congo
Red endpoint. The layers were separated and the aqueous
portion was extracted several times with ether. The
combined ether extracts were dried over magnesium sulfate.

after filtration, the ether was removed by distil—
lation. Distillation of the remaining liquid afforded
11.5 g. of acidic material, b. p. 92—930/15 hm.;
nil 1.4375. The index of refraction indicated that it

was a mixture of cycloprOpanecarboxylic and cyCIOprOpane-

acetic acids.

-. o _ i l .
CycloprOpanecarboxyiic acid n: 1.4990 b.p. 8:0/45mm.(7)

- . . , 2 , -
CyClOprOpaneacetic aCio nDl 3.4340 b.p. 850/s5mu.(()
2. Oxidation Using Chromium Trioxide-Sulfuric Acid.

To a one-l ter three—necked round-bottomed flask
equipped with reflux condenser, dropping funnel and
magnetic stirrer and set in a water bath was added 39.5 g.
(0.46 moles) of '8 -cyclopr0py1ethanol and 92.5 ml. of
acetone. To this well-stirred solution was added slowly
a solution of 61.5 g. (0.615 moles) of chromium trioxide
and 95.8 g. (1.04 moles) of sulfuric acid in 405 :1. of
water. The addition required approximately thirteen hours;
the temperature of the bath was never a lowed to exceed
30°. Water (100 ml.) was added and the mixture stirred
for an additional hour; the aqueous layer was still acid

to Congo Red. The layers were separated and the water

iayenr extracted with three lOO-ml. portions of ether.

59

The ether extracts were added to the organic layer and
the solvents were removed by distillation on a steam
bath. To the residue was added 300 m1. of 4 N sodium
hydroxide and the whole was refluxed for 5 hours. After
the mixture cooled, it was extracted with three 50-n1.
portions of ether. The ether extracts were combined,
dried over potassium carbonate, and filtered. The ether
was removed by distillation. The residue yielded 6 g. of
unreacted alcohol.

The alkaline solution was acidified to Congo Red and
was extracted with four 75-ml. portions of ether. The
extracts were combined, dried over magnesium sulfate,
filtered and the ether removed by distillation. The
residue afforded 22.3 g. (57%, based on unrecovered
alcohol) of cyclopropaneacetic acid. The infrared
Spectrum is shown in Figure 9.

b.p. 910/15 mm. n25 1.4321 Neut. Eq. 100.1

D
Lit. Values ( 7' ) b.p. 890/15 mm.; 1135 1.4320 Neut. Eq. 100.1.

II. THE PREPARATION OF ACYL PEROXIDES.

A. greparation of the Acid Chlorides.

The acid chlorides were prepared from the
corresponding acids either by reaction with thionyl
chloride or by exchange with benzoyl chloride.

1. Preparation of Cyclopropanecarboxylyl Chloride

To a loo-m1. round-bottomed flask equipped with a

distillation head was added 10.0 g. (0.12 moles) of

.ooflhoazomhpoe cophmo CH Uwo: oapmommCmdosgoHohu mo :zhpomdm omhmpMCH .m waswam

mcopoaz :H newsmam>m3

em Ma we HH OH m -w

 

 

lull—(dill! 1.4lallcllil4 21.3- .- .a .-- - d IJI

-,...._.—~._..— ,7

 

 

 

 

 

 

m.

:3:-

 

 

 

lo

 

f
l

 

 

.--—-——w *

 

 

‘ _.»fl_- 9——
\

 

 

 

 

61
cynzlopranecarboxylic acid (49) and 33 g. (0.24 moles) of
innizoyl chloride. The mixture was heated and the distillate
was collected until the temperature of the distillation
Jhead.began to drOp. Redistillation yielded 7.0 g. (57%)
of cyclopropanecarboxylyl chloride, b.p. 113-50. Lit.
'Value (40) 112-50.
2. Preparation of CycloprOpaneacetyl Chloride

(a) By Exchange with Benzoyl Chloride

 

The reaction was similar to the one above. Cyclo-
propaneacetic acid (10 g., 0.10 moles) and benzoyl chloride
(29 g., 0.21 moles) afforded 6.0 g. (51%) of the acid
chloride, b.p. 132-1340. Lit. Value (40) b.p. 152-1550.

(D) gy Reaction with Thionyl Chloride

A mixture of 10 g. (0.1 moles) of cyclopropaneacetic
acid, 13.1 g. (0.11 moles) of thionyl chloride and 25 m1.
of chloroform was refluxed for 4 hours. After cooling,
the chloroform and the excess thionyl chloride were
removed by distillation through a small Vigreux column.
The fraction boiling at 132-33° was collected and amounted
to 7.5 g. (63%). The infrared spectrum is shown in
Figure 10.

3. Preparation of Cyclohexanecarboxylyl Chloride

A 50—m1. round-bottomed flask was charged with 6.5 g.
(0.05 moles) of cyclohexanecarboxylic acid and 6.4 g.
(0.054 moles) of thionyl chloride. After the initial
reaction subsided, the mixture was refluxed on a steam

bath for an hour. The mixture was then distilled. Upon

62

.oUHQOazomhumB conhmo CH mofisoflzo Hmpmommcmaopmofloho mo Esspowdm pmamp%CH

eschew: CH newcmfim>m3

.... a m .1. e m
a A} a ll! .

—— 23'
TH
A
——r—‘|
(\J

H

r—u‘

H

O

H

.ofl themes

 

q 1fi ql

...-—
.'.-J-

 

 

 

_’-

 

 

 

 

”I
- -..—o".

 

—.—-‘—q ’ —' -.~'--

 

 

 

 

 

 

~
‘h-—
v

 

 

 

 

4:?

 

 

 

 

65

redistillation, the fraction boiang from 77-800/20mm.
was collected. This amounted to 5.7 g. (0.059 moles) or
78%lof the theoretical amount.
4. Preparation of Cyclohexaneacetyl Chloride
Cyclohexaneacetic acid (10 g., 0.07 moles) and
thionyl chloride (9.5 g., 0.08 moles) were treated as
described above, except that after the reaction was
complete a partial vacuum of 2 mm. was applied for one
hour to enSure the complete removal of the excess thionyl
chloride. Distillation of the residue afforded 9.5 g.

(0.059 moles, 84%) of the acid chloride, b.p. 114-60/15 mm.

5. Preparation of Trans-4-t-Butylcyclohexanecarboxylyl
Chloride

The procedure was identical to that of cyclohexane—
acetyl chloride. Three grams (0.016 moles) of the
correSponding acid (42) and 1.9 ml. (0.024 moles) of

thionyl chloride yielded 2.75 g. (0.013 moles, 84%) of the
acid chloride, b.p. 850/1 mm.

B. Preparation of Acyl Peroxides

The acyl peroxides were prepared by treating the acid
chloride in ether with an equivalent amount of sodium
peroxide. Details of the preparation of cyclopropaneacetyl

peroxide are given as an example.

on

In a BOO-ml. three-necked round-bottomed flask, ice-
cooled and equipped with a magnetic stirrer, reflux
condenser, dr0pping funnel and thermometer there was
placed 2.0 g. (0.026 moles) of sodium peroxide and 40 ml.
of anhydrous ether. To this was added 5.8 g. (0.049 moles)
of cyclOpropaneacetyl chloride and the reaction was
initiated by the addition of several dr0ps of water, with
additional dr0ps added later to keep the reaction going.
The reaction was assumed to be complete when the yellow
color of the sodium peroxide had disappeared and the
addition of water no longer caused the temperature to rise.

Cold water was added to dissolve the salt. The
mixture was stirred for several minutes and the layers were
separated. If the aqueous layer was not basic to litmus,
the ether layer was washed with 10% sodium carbonate
solution. The ether layer was then dried over calcium
chloride. '

The mixture was filtered and the ether removed on a
Rinco evaporator. The remaining liquid was dissolved in
20 m1. of pentane and crystallized at dry—ice temperature
in an atmosphere of carbon dioxide. Tne peroxide would
remain crystalline if stored at -20°, but melted when
warmed to room temperature. Titration of the peroxide
according to the method of Sllbert and Swern ()6) indicated
a peroxide content of 85%. The infrared spectra in

carbon tetrachloride and carbon disulfide are shown in

.Flgures 11 and 12. On one or two occasions, unreacted acid
chloride appeared as an impurity and was removed by
washing with SOdiUm carbonate solution; this was necessary
for the acid chloride accelerated the decomposition rate
of the peroxide. I

The other peroxides used were prepared in a si_ilar
manner. Cyclohexaneacetyl peroxide, also a liquid, was
crystallized in an identical fashion and possessed a
peroxide content of 88%. CycloprOpaneformyl and
trans-4-t—butylcyclohexaneformyl peroxides, which were
solids at room temperature, were recrystallized from

n-hexane and possessed purities of 99+».

III. THE DECOMPOSITION OF ACYL PEROXIDES
A. Kinetics of Decomposition

The decompositions were performed in the apparatus
illustrated in Figure 13. The peroxides were used in the
form of a standard solution in carbon tetrachloride. h
20-m1. portion of the solution was placed in the reaction
flask; the flask was then set in a constant-temperature
bath and allowed to attain thermal equilibrium. The first
sample was taken to be zero-time; samples were withdrawn
periodically, placed into vials and frozen in dry—ice
until analyzed.

For analysis, the samples were allowed to thaw and
the peroxide content was determined either spectrOphoto-

metrically by following the disappearance of the 5.62fix

.‘

.owflpoHsommpoe cognac ca ocfixopo; Hzpoommcwuoscoaoho mo Espoomcw poLQAQCH .«H magmab

~ r

mcomoaz CH zpmcofio>m3

 

 

 

 

 

 

 

 

 

 

 

 

3 NH 2 o H m. m s w m a m
i _ 1 ... « I1 1 a _ _ i
w
_
.
.
i
.
_
.
is ..
—| ‘1..- hr r P b _ II. b F _

 

 

 

67

1|n|lll|

‘

w

:H

 

.004Hazuafl

1 I P\ rl

9.. S

lull. II J I ‘II.I""I It‘d.

conpmo Ca

mwaxopom nzpmommcmaon

 

aofiomo mo azspommm copQLQCH

acomoiz Cw Spmcmfim>m3

m w

----..-|..-..|II.I.II £4

 

 

 

 

x.

.mfi ohsmwb

 

«\O

m
il

 

. '1 .l I

 

 

—.__._ ---... _

 

 

n
l a

rig:

‘fit .- _.

i”.--

-J__i

 

 

 

 

.oQSthmmgmu Umpfimmc pm Umcflmpcflms Spam :
.Hmmmm> coflpommm n
.cmmoppH: oumndpmm on weapoanomppmp

connmo wcflcfiwpcoo cam pmmcmvcoo spa: commando xmmfim s

.mpanmapm u
.mpflpoano ESHOHmo n
.cmmoppac hue on .UHom Oshawasm n

_

.mcssomgoO samfism m>osmp 0p .pmpma CH mpmpmom cmma ommeSpmm u

.cmmmxo m>osmn on ucoHpSHom m_mmmoflm n

 

 

 

mamano >m

Ct.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘I

.mcsm coapamomeoowm pom pom: mapmpwmz< mg» no Emmwmam

¢|umz

.mfl madman

 

69

band, or by iodometric titration, using the method of
Silbert and Swern (56).

The iodometric titrations were performed in the
following manner. A l-ml. portion of the peroxide solution
was dissolved in 5 ml. of glacial acetic acid containing
0.000S% ferric chloride hexahydrate, FeCls-6H20. The
solution was degassed with carbon dioxide, 0.5 ml. of
water saturated with sodium iodide was added and the
mixture was placed in the dark for twenty minutes. After
the addition of 5 ml. of water, the mixture was titrated
with standard thiosulfate, with 2 m1. of starch being
added shortly before the end-point was reached.
Standardization of the thiosulfate solution using this
method with benzoyl peroxide produced results identical,
within experimental error, to those obtained using

potassium iodate.

B. groducts of Decomposition of Peroxides

 

l. Trans-h-t-butylcyclohexanecarboxylyl Peroxide in
Carbon Tetrachloride

A solution of 0.041 g. (0.11 maoles) of the peroxide
and #.5 ml. of purified carbon tetrachloride (35) in a
50-ml. flask was refluxed in a nitrogen atmosphere for
five hours. After the mixture cooled, the ester content
was determined by measuring the intensity of the 5.75,“
band and translating this into concentration using a
standardization curve prepared from pure trans-h-t—butyl—

Cyclohexyl trans-4~t-butylcyclohexanecarboxylate; this

70
showed that 0.05} mnoles of ester, or 48.8% of theoretical,
was obtained.
2. Cyclopropanecarboxylyl Peroxide in t-Butylbenzene

Cyc10propanecarboxylyl peroxide was decomposed in
t-butylbenzene with the hope of obtaining dicycloprOpyl,
[>>—<:]. A BOO-ml. three-necked flask equipped with
reflux condenser, dropping funnel and nitrogen inlet tube
and maintained at 78° was charged with 50 ml. of t-butyl-
benzene. To this was added over a period of five hours
a solution of 5 g. (0.03 moles) of the peroxide in 90 ml.
of t-butylbenzene. The mixture was heated for an additional
seven hours. The infrared spectrum of the reaction mixture
indicated that the peroxide was not entirely decomposed
and that the principal products were ester and acid. The
solution was refluxed for two hours longer and the infrared
spectrum of the first few milliliters of distillate
indicated that no dicycloprOpyl was formed.

The reaction mixture was shaken with 150 ml. of l N
sodium hydroxide for fifteen minutes. The alkaline solution
was acidified and extracted with ether. The extract was
dried with magnesium sulfate and filtered and the ether
evaporated. The residue was titrated with standard base
and contained 0.4 g. of acid, calculated as cyclOprOpane-
carboxylic acid. Distillation of the dried t-butylbenzene
solution did not separate the ester from the solvent. The

yield of ester was estimated from the infrared Spectrum to

Y1
be l.# g., or 37% of theoretical, calculated as cyclo-
propanecarboxylate.

Decomposition of the peroxide in refluxing
chlorobenzene yielded similar results, with no dicyclo-

propyl being detected.

3. Decomposition of CyclOprOpaneacetyl Peroxide in
Carbon Tetrachloride

The apparatus used for determining the products of
the decomposition was essentially that used for kinetics
runs (Figure 13), except that two dry-ice traps and a tube
containing Ascarite and Anhydrone for the determination of
carbon dioxide were placed, in that order, between the
reaction vessel V and the drying tube D'. A known
amount of a standard solution of peroxide in carbon
tetrachloride was refluxed for three hours in a stream of
purified dry nitrogen (50). The flow of nitrogen was
continued for a half-hour while the reaction mixture .
cooled. The carbon dioxide evolved was determined by the
increase in weight of the Ascarite tube, while the ester
and acid were determined spectrOphotometrically, using the
5.75ii and the 5.8?u.bands respectively. The results are
listed in Table 8, page 29. i

To obtain more information regarding the nature of
the ester, most of the carbon tetrachloride was distilled
through a packed column. The remaining liquid was washed
with 10% sodium carbonate solution and dried over calcium

sulfate. The filtered liquid was distilled under reduced

,{. 2

pressure and, upon redistillation, yielded a material whose
infrared Spectrum was somewhat similar to that of cyclo-
prOpylcarbinyl cyclOprOpaneacetate. The mixture consisted
of almost entirely one product, as was indicated by the
vapor-phase chronatograph; purification of the compound
using this method was impossible because of extensive
decomposition occurring in the column. A spectrum of the
high boiling material appears -n Figure 14.

For the determination of alkyl chloride, 96 ml. of a
0.076 N peroxide solution in carbon tetrachloride was
refluxed for three hours. The infrared Spectrum of the
first ten milliliters obtained on distillation through a
packed column possessed a small carbon-hydrogen absorption
peak. A partial vacuum was applied to the remainder of
the solution and everything volatile at room temperature
and 5 mm. of pressure was collected in a dry-ice trap.
Most of carbon tetrachloride was fractionated; each of the
fractions possessed a slight carbon-hydrogen absorption
peak in the infrared. The remaining Liquid (10 ml.) was
distilled, not allowing the temperature of the oil bath
to exceed 100°. The temperature of-the distillate
gradually rose from Y6-80°, after which it dropped slowly.
The infrared Spectrum showed that, in addition to small
amounts of ester and acid, a substance, either an alkyl
halide or a hydrocarbon, was present. The amount,

calculated as alkyl chloride, was estimated from the

HI.

coapflmoQSoomQ mSP

:H
il

m--_‘

.—

_ h“-.— ,—

P.———__-——.--.c.—-_~ “‘0“ ""

eon

c

.ooapomzomppoe cophmo ca summons; Hmpoomcrmmongofioho no

u UmchpQO Hmanoum: mCaHHom swam m2» mo .Sppomam poLmAQCH .ja panama

mCOLOHE c4 Spmcmfim>m3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- 4 q 4 Jr 4 . . a q
:
...
j _
w
. .
W
..
u
_
i
_
i121}! L
llfpll Ir lllll F lJIlL-l|ll _

 

 

 

 

74
carbon-hydrogen absorption. Absence of an olefinic peak

indicated that the compound was saturated.

4. Decomposition of CyclOpropaneacetyl Peroxide in the
Presence of Trichloroacetic Acid

The apparatus used for the decomposition is that
described above. A mixture of carbon tetrachloride
solutions of the peroxide and of trichloroacetic acid was
refluxed for three hours and the carbon dioxide evolved
determined by the increase in weight of the Ascarite tube.
The solution was cooled, washed with 10% sodium carbonate
solution and dried; the filtered solution was then diluted
to 50.0 ml. The ester produced was determined by infrared;
two ester bands were present, one at 5.6Tit and another at
5.76/L. The data are listed in Table 10, page 32.

5. Initiation of the Polymerization of Styrene by
Cyclopropaneacetyl Peroxide

Two flasks were each charged with twenty grams of
freshly distilled styrene; to one was added one gram of
cyclopropaneacetyl peroxide. The flasks were placed in a
water bath maintained at 40-600. The extent of
polymerization was noted by cooling the flasks at intervals
and comparing the viscosities and Opaqueness of the
solutions; the flasks were then returned to the water bath.
Throughout the four hour observation period, the styrene
containing cycloprOpaneacetyl peroxide was both more
viscous and opaque. At the end of four hours, this

solution required twice the time as the uncatalyzed

tr 5
styrene to drain approximately the same length.

IV. PREPARATION OF t-BUTYL PERESTERS OF SOME ALICYCLIC
ACIDS

A. Purification of t—Butyl Hydroperoxide

t-Butyl hydrOperoxide (Lucidol Corp.) was dried by
refluxing at 40 mm. and removing the water as it was
collected in a distillation head. The hydrOperoxide was
then distilled and the material boiling at 51-520/40 mm.
was collected; n? 1.3995. Lit. value his 1.3986 (51).
B. Preparation of the t-Butyl Peresters

The peresters were prepared by treating the
appropriate acid chloride (prepared as described above)
with t-butyl hydrOperoxide in the presence of pyridine.
This is based on the method of Bartlett and Hiatt (2).
The preparation of t-butyl cyclohexanepercarboxylate is
described.

In a lOO-ml. round-bottomed flask was placed 25 ml.
of dry pyridine, 25 ml. of anhydrous ethyl ether and 10 g.
(0.11 moles) of t-butyl hydroperoxide. The solution was
stirred with a magnetic stirrer and was cooled to -6 to
~80 by means of an ice-salt mixture. Cyclohexanecarboxylyl
Chloride (10.1 g., 0.97 moles) was added at such a rate
‘that the temperature did not rise above 5°. After the
achdition was complete, the reaction mixture was placed in
tflie freezer compartment of a refrigerator (-17°) for

Exthy hours. During this period, a white precipitate,

76

presumably pyridine hydrochloride, had separated. The
mixture was washed with small portions of iced lop
sulfuric acid to remove the pyridine, then with 10» sodium
carbonate to remove any acid present and finally with
water. The ether layer was then dried over magnesium
sullate. The solution was filtered and the ether removed
by a Jet of dry air. The residue-was passed through a
column of Florisil and eluted with pentane. The first
fifty milliliters, after removal of the pentane by
evaporation, gave a residue of 6.5 g. Its infrared
spectrum (carbon tetrachloride) possessed no extraneous
acid or hydroxyl peaks and is shown in Figure 15.
Titration of the perester using a method based on a note
of Simon (52) indicated a peroxide content of 77%.

The t-butyl perester of cyclOpropanecarboxylic acid
was prepared in an analogous manner. Its infrared spectrum
is shown in Figure 16. The purity of the perester was
76%, as indicated by titration.

C. Method of Titration of the Peresters

 

Aliquots of carbon tetrachloride containing a known
amount of perester were added to flasks containing a
solution of 10 ml. of acetic anhydride, 10 ml. of glacial
acetic acid, 30 m1. of absolute alcohol, 6 g. of potassium
iodide and 20 ml. of water. The mixture was degassed with
carbon dioxide and was allowed to stand for two hours with

occasional shaking. Water (150 ml.) was added to the

.ocflnoHcomapme conpmo CH mpmfimxonmmopmgmcmxogoHoho Hzpzmnp mo esppooom possum:H
mconOHz CH zpmcmflm>m3

:a ma mu Ha 0H m m s o m

4!

 

I i--- 2-1! a
7 _ ._ _ _ _ q a J—

- ~— -..-‘-

m-.o— n-n-‘fi

 

 

 

 

 

 

 

 

 

.mfl magmas

_——‘-.—_¥

 

 

 

 

 

.mofipofigomppoe :onpmo CH opmfihxonpmopoomnmaopLOHoho thsmup mo guppomom omamLmCH .wfl magnem

~

 

mcopoaz 2H Spmcoam>m3
ms, .2 3 m m i. m m

u . . I l)).lu§fll , . . . - . . q . . I -

3 a l

”:1“
M

 

._.__.'._.- {*1

 

 

 

 

 

 

- 4.—V§ 7“...” V-_~-_..

 

 

 

 

 

 

 

 

 

 

 

 

-\

 

 

n'

T

\(‘5

J

solution, which was then titrated with standard thio-
sulfate. gtarch was used as the indicator, although the

color was brown instead of purple.

V. THE DECOMPOSITION 0F PERESTERS

The perester was dissolved in sufficient carbon
tetrachloride to make an approximately 0.06 N solution.
Portions of this solution were sealed in ampoules, placed
in an oil bath maintained at the desired tenperature and
allowed ten minutes to attain thermal equilibrium. The
first sample was taken as zero-time; samples were taken
at intervals and frozen in dry—ice until analyzed. They
were thawed and the peroxide content determined spectro-
photometrically. The kinetics were determined by following
the rate of disappearance of the 5.62;k band, which is
characteristics of peroxide carbonyls. The rate constants
were obtained from the transmissions by means of the

equation listed on page 87.

VI. MISCELLANEOUS EXPERIMENTS

A. The Preparation of CycloprOpylcarbinyl Cyclopropane-
acetate

0 I ’ o
. , . - l
\>CH2|001 + [>CH20H ”names [>CH2EOCH2<)

A l25-ml. Erlenmeyer flask was charged with 20 ml.

 

of dry pyridine, 2.6 g. (0.022 moles) of cyclOpropaneacetyl

chloride and 1.6 g. (0.022 moles) of cycloprOpylcarbinol.

80
The flask was stoppered and the mixture was allowed to
stand at room temperature for two hours, after which it
was diluted with 50-ml. of ether. The ether solution was
washed with two 50-ml. portions of water, then with 25 ml.
of 10% sodiun carbonate and again with 50—ml. of water.
After drying over wagnesium sulfate and filtering, the
ether was removed under reduced pressure. The remaining
liquid was distilled through a small Vigreux column and
2.1 g. (62%) of the ester was obtained, b.p. 65—660/2 mm.;
n§5 1.4470.
Lit. values (40) b.p. 63-650/2 mm.; His 1.4475
The infrared Spectrum of the ester is shown in Figure 17.

B. The Preparation of Cyc10propylcarbinyl Trichloroacetate

 

A mixture of 4 g. (0.024 moles) of trichloroacetyl
chloride, prepared by the action of thionyl chloride on
trichloroacetic acid, and 1.8 g. (0.025 moles) of cyclo-
propylcarbinol in twenty milliliters of dry pyridine was
allowed to stand, with occasional shaking, for three houni
Fifty milliliters of ether were added and the mixture was
washed successively with 50 ml. of water, 25 ml. of cold
10% sulfuric acid, 25 ml. of 10% sodium carbonate solution
and finally again with 50 ml. of water. After the ether
solution was dried and filtered, the solvent was removed
under reduced pressure. Distillation of the remaining
liquid through a small Vigreux column afforded 2.8 g.

(0.014 moles) of material boiling from 81-850/5 mu.;

.oUHwOHzomnpoE
coonmo cH mumucommcmgosmoHoho HthQsmOHzoopaoHoho mo gospoomm UmummmcH .wH mwowHa

mconon cH npwcmHm>m3

0H m m i m m. is m
a I. W

l _ 1 Juill _
i
.
_

fill m i- «in.

1“!
‘M—1

--—... ._.> . ._—- _M_—

\
_ _ ”...-“ww-

 

 

 

)Ilbiul. . Lll) 1 1h .‘l‘lLti! I . . . g —

 

......——...—....__._ . _

82

nD 1.4690. The infrared Spectrum (Figure 18) showed a

carbonyl absorption at 5.67/4.

C. Reaction of Cyclopropaneacetic Acid with Sulfuric
Acid

A mixture of l g. of cyclOpropaneacetic acid and
15 m1. of 20% sulfuric acid was refluxed for thirty
minutes and, upon cooling, extracted with 25 ml. of carbon
tetrachloride. After the extract was dried over calcium
sulfate and filtered, an infrared spectrum was taken. In
addition to an acid carbonyl peak (5.8?AL), a band at
5.63}x,was present. This is the range of 7’-lactones and

the peak is ascribed to the presence of /6>-methyl-Y1butyro-

lactone.

-—,_.._.-Auw~_

.opHnoHEQmppoB sopsmo

 

l‘
qf“
N

CH

mumpmooosoHchsE Hchosmo

% \J.\u I \...
Fer.Pl.PH.LA.,

J
\n > 4. .. a). .1
.-.Lro (O. r ' u.

moosoHE CH cpwcmHmsmE

m

.
I KAI ' .' Il1‘.‘i‘

...o

‘_

 

w

..H. - -

 

 

‘~

-\.’3

 

W

..._.._..-—

 

 

 

cocm oopspACH

O \J.
jfl
ix

 

10.

11.

12.

13.

15.
16.

84

LITERATURE CITED

H. Hart and D. P. Wyman, J. Am. Chem. Soc., 81, 4891
(1959).

P. D. Bartlett and R. R. Hiatt, J. Am. Chem. Soc.,
@2,13y8 (135L/)o

P. D. Bartlett and D. M. Simons, J. Am. Chem. Soc.,
is. 1755 (1960).

P. D. Bartlett and c. Rfichhardt, J. Am. Chem. Soc.,
@2. 1756 (1360)-

J. D. Roberts and V. C. Chambers, J. Am. Chem. Soc.,
‘12, 3176 (1951).

I. Smith and S. MacKenzie, J. Org. Chem., 15, 74
(1950)-

. H.Turnbu11 and E. S. Wallis, J. Org. Chem., 21,
663 (1956)

t"

. D. Rosenburg, A. J. Gibbons. Jr. and H. E. Rausden,
. Am. Chem. Soc. , 79 213/ (1957)

. Hart and A. Holzschuh, unpublished results.

. c. L. Wilson, J. Chem. Soc., 39, (1946)

S
J
H
K. Bowden, I. M. Heilbron, E. R. H. Jones and
B
C. Walling, "Free Radicals in Solution”, John Wiley
and Sons, Inc., New York, 1957, pp. 474-503.

A. V. Tobolsky and R. B. Mesrobian, ”Organic Peroxides”
Interscience Publishers, Inc., New York, 1954, pp.
72-87

J. Hine, “Physical Organic Chemistry", McGraw-Hill
Book Co., Inc., New York, 1956, pp. 412-419.

E. S. Gould, ”Mechanisms and Structure in Organic
Chemistry”, Henry Hoit and Comp., New York, 1959,

pp. 714-720.

A. V. Tobolsky and R. B. Mesrobian, loc. cit., p. 107.

W. E. Cass, J. Am. Chem. Soc., 68, 1976 (1946).

17.

18.

19.
20.
21.

22.

23.

24.

25.
26-

27.

28.

29.

30.

31.

32.

33.
34.

35-

36.

85

K. Nozaki and P. D. Bartlett, J. Am. Chem. Soc.,
68, 1686 (1946).

. E. Leffler, J. Am. Chem. Soc., IE; 67 (1350).
12.

. Szwarc, Chem. Réfis., 41, 75 (1950)-

. Walling, loc. cit., p. 478.

J

M

C

G. S. Hammond and L. M. Soffer, J. Am. Chem. Soc.,
72, 4711 (1950)-

G. S. Hammond, J. T. Rudesill, and F. J. Modic,
J. Am. Chem. Soc., 12, 3929 (1951).

D. B. Denney and G. Feig, J. Am. Chem. Soc., _33
5322 (1959).

P. D. Bartlett and J. E. Leffler, J. Am. Chem. Soc.,
12. 3030 (1950).

C. Walling, loc. cit., p. 493.

A. Rembaum and M. Szwarc, J. Chem. Phys., 22, 909 (1)55)-

r
bx
\ J

D. F. DeTar and C. Weis, J. Am. Chem. Soc., _83 42
(1956).

J. D. Roberts and V. C. Chambers, J. Am. Chem. Soc.,
12. 5034 (1951).

R00. 1, ,-
A. Streitwieser, Chem. R#ws., 563 5;] (1950).

\i)

R. H. Mazur, W. N. White, D. A. Semenow, C. C. Lee,
M. S. Silver and J. D. Roberts, J. Am. Chem. Soc.,
Q}. 4390 (1959).

C. G. Overberger and A. Lebowits, J. Am. Chem. Soc.,
Zéy 2722 (1954).

J. D. Roberts and R. H. Mazur, J. Am. Chem. Soc.,
,12, 2509 (1951).

Private Communication to Professor Hart.
H. Hart and H. H. Lau, unpublished results.

L. F. Fieser, ”Experiments in Organic Chemistry”,
2nd. Edit., D. C. Heath Comp., Boston, 1941, p. 565.

L. S. Silbert and D. Swern, J. Am. Chem. Soc., 813
2364 (1959).

37-

39.

40.

212.
43.
an.
45.
46.

47.

48.

49-
50.
51.
52.
53-

f“ j
C‘

D. B. Denney and D. Z. Denney, J. Am. Chem. Soc.,
8 , 13:9 (1360). ~

L. Horner and H. Junkermann, Ann., 591, 55 (1355).

h
M. S. Kharagh. J. Kuderna and W. Nudenberg, J. Org.
Chem., 19, 1283 1954)-

D. P. Wyman, Doctoral Thesis, Michigan State Univ.,
1957-

C. Walling, loc. cit., p. 476.
H. H. Lau and H. Hart, J. Am. Chem. Soc., B}, 4837 (1;59).
C. Walling, loc. cit., p. 152.
P. D. Bartlett, Organic Symposium, Seattle, June, 1957.
C. Walling, ioc. cit., p. 491.

C. G. Bergstrom and S. Siegel, J. Am. Chem. Soc.,
11. 145 (1,152).

M. R. Fenske and H. Hart; unpublished results.

J. D. Roberts and P. H. Dirstine, J. Am. Chem. Soc.,
§ZJ 1281 (.945).

”Organic Syntheses”, 24, 36 (1944).

.L. Fieser, loc. cit., p. 595.

P. D. Bartlett and R. R. Hiatt, 100. cit., p. 1403.
Reference 2, footnote 6.
A. R. Frost and R. c. Pearson, ”Kinetics,and

Mechanisms“, John Wiley and Sons, Inc., New York,
1953, p- 96-

APPENDIX

87
Derivation of Equation Used in Calculating Rate Constants.

The equation used to follow the decomposition spectrophoto-
metrically was obtained by the following manipulations. From the

equation for a first-order reaction, one has
_ -kt
log 0 - m + log co , (1)

where co is the initial concentration of peroxide and c the concen-
tration at time t. From Beer’s law,

= £319.31 .12 = 2-303 _1_
c k’l’ 1°“; I W103 Tr ’ (2)

where I and L, are the intensities, k’ the absorption coefficient, 1’

the length of the cell and Tr the transmission. Combining (1) and

(2) , one obtains

2.303 1 __ -kt
log(-E,-1-r— log Tit) - m + log Co . (3)
Further,
1 -kt 2.303
log logfi=m+log co -log—k—,-1—,-—. (4)
Then

log(-log Tr) = 5.231333. + log co - log 21:31.93. . (5)

88

Combining all of the constants,
log(-log Tr) = 425’— + K (6)
2.303 '

The rate constants were obtained by plotting log (-log Tr)
versus t and multiplying the slope of the line, which was obtained
by the method of least squares, by 2.303.

The enthalpy of activation, AH“, was calculated using a form
of the Eyring equation (53)

k __ -AH*
l°g 7f ’ 2.303 RT + Z ’ (7)

where R is the gas constant, T the temperature in degrees Kelvin

and Z a constant.

The entropy of activation, 158*, was calculated directly from

the Eyrlng Equation (53)

e , (3)

ka AS*/R - AH*/RT
k = T e

where h is Planck’s constant, 6.62 x 10"27 erg. sec., and kb is

16

Boltzmann’s constant, 1.37 x 10- erg. deg-1.

89

 

 

 

 

 

 

mmm. smm .oH
:Hm. mmm .0
:mm. NON .w
:om. 05H .w
mmm. Hma .o
Ham. omH .m
sea. no .:
mmfi. so .m
mHH. mm .N
omo. o .H
.mcwpe .cHE «mafia maggmm

 

 

 

 

 

 

Abomm mIOH um HN. M. Fo.m n x

z $0.0 .ocoo

00.3: .QEmB

 

 

ma mqm<e

wan. :wm .ofi
0mm. oom .m
mfim. mmfi .w
m . RH é
amp. w mos .m
«9. .3 .m
0:. m. .3 A
mom. h on .m
omo. W om. .m
«so. m o .H
.wsmhe .CHE amean. mHQEmm
a-.omm m-oH x on. w on.m u x
2 00.0 .ocoo 00.x: .Qsme
:H mamas

mmHmOAmU¢m9m9 zomm<o 2H mmeommm ANBMO<MZ<mommoqowo mo ZOHBHmomEOOmQ

 

 

>mm. o:H .m
mom. omfi .w
mmm. mos .s
wad. om .w
nw:.. ms .m
mmm. om .:
o - Rm. mm .m
c, :Hm. mm .m
2mm. OH .H
.mcmne .CHE «mafia omQEmm

 

 

 

al.0mm vIOH K #0. # :N.H u x
2 0:0.0 .ocoo ow.om .ane

 

 

NH mqm<9

 

 

 

 

 

 

 

 

 

as». mmfl .o
mss. moH .m
owe. mm .:
Ham. mm .m
hmm. o: .m
moo. mm .H

.mCmnB .CHa .oEHe oHQEwm

an .omm wnoH K mo. % HH.H n x

z wflo.o .ocoo om.m: .QEmB

ma mum<e

mQHmCQmo¢mEME zommdo 2H MQHNOmmm Awemo<mz<mOMmOAo>O mo ZOHBHmomzoomQ

A it?! 1:11

91

 

 

 

 

 

 

 

 

 

 

mom. omH .w
Dim. ONH .N
Non. OOH .0
mmm. ow .m
mom. om .:
omH. m: .n
NNH. on .N
moo. ma .H
.mcmae .CHE .mEHB mHQEmm
mhmmwaécwos. Mamas?

O

 

 

ma mqm<e

awn. J oafi .m
U _

mam. m mmfi ‘ .w

mm:. M mofl .e

ems. W om .o

mm:. om .m

mmm. om .:

mam. m: .m

mmm. mm .m

wflm. cm .H
.mthB .CHE nofifle mHQEmm
HI .omm vIOH X mo. # OM.H u x
z mmo.o .ocoo om.om .aame

ma mqm<a

mnHmoqmo¢mBm9 zomm<o 2H mQHXOmMm Qwemo<mz¢mommoqowo mo ZOHBHmOmEOomm

’nl'II'.II.v

92

 

 

 

 

 

ems. ms .Oa
0mm. mm .m
mam. Hm .0
«mm. m: .w
mam. em .0
0mm. mm .m
05H. 0m H .:
mmfi. ms h .m
mofi. a W .m
mmo. o .H
.wcmpB .CHE .mgae mHQEmm

 

 

0mm. w: .w
00m. 0: .0
ozm. mm .m
mmfi. am .:
omfl. 0H .m
NOH. N .N
mwo. 0 .H
.mcmhe .CHE «mafia mHQEmm

 

 

 

 

Hu.omm ¢-oH x :m. « em.m u x
z mo.o .ocoo

h.@m .QEmB

2 00.0

.0000

a: .omm vIOH x ma. % mm.m u x
ow.©m .QEmB

 

 

Hm mqmde

ON mAm<B

mQHmOAmo¢mBm9 zomm<o 2H mQonmmm AMEmo¢mz¢mommoquU mo ZOHBHmomEOUMQ

 

 

 

 

 

 

 

 

 

 

 

 

 

Hum. 0m .0

mom. a: .s 00m. m: .m

mmm. :m .0 0mm. mm .0

:NN. wm .m HON. mm .m

«NH. 0H .: wwfl. Hm .2

Baa. NH .m mmfi. :H . .m

n, NHH. m m Nmo. b .N

a mac. 0 .H mac. 0 .H
.mCmpB .cHE .mEHB mHQEmm .mcmpe .cag .oEHB mHQEmm
Hnomm auofl x «a. « :m.m u x a-.omm e-oH x no. « m~.m u x

2 No.0 mcHOOH .0500
2 00.0 mcaxopmm .ocoo 2 00.0 oofixonmm .ocoo
.m.om .aEme _ .m.mm .QEme
mm mamda mm mqm<8

mmHmoqm0¢mHmB zomm<0 2H MQHNOmmm Qw9m0¢mz<m0mmoqowo m0 ZOHEHmOmzoomm

 

 

 

mm.m m: .0

0:.m 00 .0

05.0 mm .:

00.0 0H .0

m. ma.m m .m

_y o:.: o .H
edc.nmpfia .cHE .mefie magsmm

 

 

 

.-.omm .-oH x :H. « mm.m u x
0HH0.0 u Amommmmzvz
zaflmoahumsampae boonHom
m . 0m . 9.209

O

H

mm mamas

 

 

000. 00 .0
ms:. 0: .0
02:. >0 .w
00:. H0 .0
000. 0m .0
0mm. 0H .3
:0m. NH .0
03m. 0 .m
mom. 0 .H
.mcmpe .CHE .mEHB mHQEmm

 

 

 

H-.omm ¢-oH x mm. « om.m . x

 

2 00.0 unanOH .ocoo
2 000.0 mcaxopmm .oCOO
om.®m .0509

:m mam<e

MQHmOAm0<MEME 20mm<0 ZH mQHxOmmm AMEMUdmz<m0mmOAONO m0 ZOHBHmomEOUmQ

lill!.ll{ .Il'll il. l I

 

95

 

 

 

so.m omfi m
0H.N 00 .:
o:.m m0 an
Fm.m 00 .m
aw.m o .H

«u.omm maofi x am. « 00.0 n x
oscaczomp cmnmpmc« 020 mean:

 

ll

 

 

.Ha .ampfie .cHg .msfle mmazmm

 

 

 

 

 

 

an.oom muoa x mm. a $0.0 u x
mHHo.o u Anemmmmzvz
zfifimoaaposflupfls coonHom
00.:: n0809

00.0 H00 .:
mH.H 00m .0
MN.H ONH .N
mfl.m 00 .H

. HE « .Hwnw HR. . CHM...“ a GEHB QHQEMW
HI . 00w

mIOH x NH. 00.: I x
mflfio.o a Macmmmmzvz

hafimoappmSMAuHB coonHom

00.:: .0809

 

 

Fm mqmde

0m mqm<9

mmHmOAm0<m9me zomm<0 2H mmeommm qwemo<mz<mommoqowo m0 ZOHBHmOmEOOMQ

96

 

Hu.omw

muofi N 0 UoSQmQLQQm

pampmcoo mama 0:0 .2 00.0
0000000000 coaumppcmocoo 000 m

 

 

 

 

00.m 000 .0
0H.m 0:: .n
NH.¢ N00 .0
30.0 0:m .0
00.0 00H .:
:0.0 :NH .0
wn.HH 00 .m
mfl.0a o .H
.H2 .pmpfia .cHE .mefie madamm

 

 

 

z mHHo.o

u AMONmNmZVz

hammOthmEthHB Umzoaflom

z ©wH.0 .oCoo

00.:: .anB

can. omH .0
0:0. am .:
0H0. 00 .0
00m. 00 .m
Hmm. o .H
.mcmpe .CHE .0505 ofiqemm

 

 

 

 

0N mqm¢a

a-.omm m-oH x mfi. «.mo.m a x
.mpmconhmo ESaUOw 00H
msoosvm spas 003mm; coausflom
00.3: .0309

mm mqm<B

mQHmoqmo¢mEmH 20mm<0 2H mQonmmm AMBMO<mz<mommOAUMO 00 ZOHEHmOmzoomm

‘ E It. ..Yytlfi. .

 

 

 

 

 

 

00:. 000 .0
«0:. 0mm .0
mo:. :0H .w
«00. 00H .0
«mm. wma .0
~00. NOH .:
00m. 00 .0
Hmm. on .0
How. 0 .H
.wcmme .CHE .mEHB oaasmm

 

 

Nan. Ham .0
N::. m sea .e
N00. 0:H .0
0mm. 0mm .0
000. 00 .3
$00. 00 .0
H00. 00 .0
00m. 0 .H
.mcmne .CHs .msfia maagmm

 

 

 

 

aa.omm mIOH x :H. fl #0.: u x
2 0:0.0 0Ho¢ capoomahsposflue .ocoo

an.omm mnofi x 0H. a 00.: u x
z :H0.0 cfioa capmo¢ .ocoo

mcflxopmm .ocoo

mUHXonmm .0200 z N:0.0

2 030.0
om.:: .0208

00.3: .0509

 

 

H0 mqmde 00 mqm¢9

mmHm04m0<mEm9 zomm<0 2H mQonmmm qwemo¢m2¢mommoqowo m0 ZOHEHmOmEOUMQ

 

98

 

 

 

H0.o H0 .0
00.0 00 .:
om.H 0: .0
00.0 0H .m
00.0 0 .H
.Hsanmpaa .CHE .mEHB mHQEmm

 

 

 

 

al.0mm vIOH K miw I x

0HHo.o u Amomommmzvz

haamoahmeHhuHB UwonHom
z H00.0 UHo¢ oaumomopoHSOHQB

z H00.0

moHKOQmm

.ocoo

o0.:: .0509

 

 

 

 

00 mqm<9

H00. me .0
0:0. 00H .0
N:0. 00H .0
mm:. mmfi . .0
00:. :oH .:
:H:. 00 .0
N00. 0: .m
00m. 0 .H
.wcmpe .cHE .mEfiB mHQEwm
a-.omm m-oH x «H. a 00.0 a x

2 00.0 maficfimhm .0000
2 00.0 ocfixoamm .ocoo
o0.::.QEmB

m0 mqm<9

mmHmOAmo«mHme zomM¢0 2H mmeommm qwemo<mz<mommoqowo mo ZOHEHmOmEOOmQ

 

 

 

 

 

1:
00.: 00 0
00.0 00 .:
:H.0 H: .0
0:.0 om m m
Hw.m o .H

.H0 .000He .005 .08:9 magamm

 

 

00.0 0 H0 0
00.0 0: .:
0:.0 on .0
00.0 0: .m
00.0 o .H
.He .pmpaa .002 .mefis «HQEmm

 

 

 

 

Ht.omm ¢IOH K M.H u x

.0HH0.0 . Amomommmzvz
zHHmOanmeflnpfie umsoflflom

z 0Ho.o 0:0< capmomonoHsOHpa
z H00.0 00HK0900 .ocoo
om.:: .0508

Hloowm QIOH N m n X

0HHo.o n Amomommmzvz
zfiamofippmsflgpfla cmzoflflom

2 000.0 cao< 0:0000000000009
2 H00.0 mUHKOhmm .OCOO
om.:: .0506

 

 

00 00009

:0 mqmda

mQHmOAm0<mEMB zomm<o 2H mmeommm Awamo<mz<mommoqowo mo ZOHBHmomEOUMQ

Ivlrf. P: Ill LII I t

lOO

 

 

 

 

00H. 0N0 .0
00H. 000 .0
00a. 0:0 .0
m::. 00: .0
:mH. 000 .0
:HH. Ohm .:
NOH. 00H .0
000. 00 .N
H00. 0 .H
.mcmne .CHE .meHB mausmm

 

 

00:. 00 .:
0H0. 0.00 .0
0:0. 0.0: .0
0:0. :0 .H
.mcmpe .mp5 .msHB 0H0E00

 

 

 

 

2 00.0

41.000 QIOH K mm. fl :H.H u x
00onmmm

.0000

0.:0 .0509

7.8.. .72 0. mo. ..m wen .. x

2 00.0

moaxonmm .0000

o:.:: .0509

 

 

N0 mumde

00 mum<9

mmHmoqmo<mBmB zommdo ZH mQHNOMMO ANBMO<mz<xmmoqowo 00 ZOHBHmomsoomQ

‘! it! i!

101

 

 

 

 

 

 

 

 

 

 

 

 

 

mom. 000 .0
000. m 000 .0 mam. 000 .0
00:. w 000 .0 00m. 000 .0
00:. m 00: .0 0:0. 000 0
won. W 03M _ .m NNN. om: .m
000. , 00m .: Hma. omm .:
Hmm. 00H .m 00H. 0am .0
00a. om .m 00H. om .m
NMH. O .H MHH. O .H
.mcmge .caE «mafia mHQEmm .mcmne .cHE .mEHB mHQEmm
H|.omm muOH x HH. % m0.m u x an.ommZmefiox :Owcwomm.womow
2 00.0 mUHXOnmm .ocoo 2 00.0 moaxopmm .ocoo
ow.H0 .0809 00.:0 .asme
mm mum<e . 00 04009

mQHmOAmO¢mBme zomm<o 2H WQonmmm qwemo<m2¢xmmoqowo m0 ZOHBHmomEOomn

 

102

 

 

 

 

00.0 :00 .0
00.H 00: .0
00.H 000 .0
0m.m 000 .:
00.0 Ham .0
00.0 omfl .m
«0.: 0 .H
.05 .pmpfla .cfie .0200 0005mm

 

 

 

 

au.0mm mIoH x 0H. fl 0:.: I x
0H00.0 u Amomommmzvz
0HHmOHupmsfippfle cmzofiaom

2 000.0 0000 000000000000009
2 00.0 mcfixonmm .0000
00.:0 .0809

munuuuuu
000. 000 .0H
:00. 000 .0
:00. 00: .0
000. 00: .0
:0:. 000 .0
«0:. 000 .0
0:0. 000 .:
000. 00a .0
Hma. 00 .m
00H. 0 .H

.00009 .008 ”05.3. mHQEmm

 

 

 

 

H: mqm<9

Hl.0mm mlOH un HN. ...—... mm.# H x
2 000.0 00000000 .0000

2 00.0 mnfixonmm .0000
o®.H0 .0609

1|"

o: mam<9

MQHmOQmo<m9m9 zomm<o 2H mQHXOmMm 499mo<mz<xmmoqowo mo ZOH9HmOmSOomn

 

 

 

 

nmd. OH .N
non. m .0
3. m6 _. .m
mam. m H .:
.9 mnm. m.m .n
m HON. N.N N
mmfl. o .H
.mcmne .mp2 .mEHB mflqemm

 

 

 

an.omm muoH K HH. # nm.m u x
om.mw .QEmB

 

m : mama

 

mmm.
8n.
mom.
mmm.
8m.
mwfi.
S...
ozfi.
m2.

(“Md-mKOb-q)

 

.mthB

 

.mps .meHB

 

mfimsmm

 

.nl.omm mIOH K mH. a... mM.N I x
om.mm .QEmB

N: mflm<8

"

mmHmOAmo<mEmB zomm<o 2H m9<qwxomm<ommmm24xmmoqowo AMBDmnp mo ZOHBHmomzoomn

 

lO#

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

moo. 05m .mH
mww. . cam .HH
mum. W owfl .oH was. o:m .oH
0mm. omfi .a HH:. omH .m
Hmm. o:H .m nmm. omfl .w
on. omH w n:m. oafl .§
omz. 00H .m 0mm. omfi .w
mflz. ow -.m mwm. 00H .m
wmn. om .2 mam. om .:
mmm. o: _ .n owfl. om .m
~:m. om w .m Hmfl. o: .m
mad. 0 .H moo. om .H
.owLH .CHE .mEHB mamzmm .mcmmB .cflg .mEHB wHQEmm
Hu.omm muofl x we. a mm.m n x a-.omm m-oH x m». « w:.m u z
. oOOH .QEwE 0OOH .QEmB
m: mqm¢e . :: mqm<a

MQHmOAmo<mBmB zomm¢o 2H mE<AMNOmm<ommmm2<xmmoqowo HNBDmsp mo ZOHBHmomEOUMQ

105

 

 

 

 

 

 

 

 

 

 

 

 

 

m:m. moH .0.
0mm. om .m
mum. om .m mwm. moH .u
gas. ow .5 wom. “ om .N
ova. oo .0 mm:. .mw .m
wwm. on .m mfl:. om .m
wmm. 0: .: mwm. m: .:
00m. om .m mom. on .m
mwa. om .m mwa. ma .N
QMH. OH .H Nwo. O .H
.mCmpe .CHE .wEHB mHQEmm .mCmnB .CHE .mEHB maasmm
an.omm ¢|OH x :m. a omwm u x al.0mm ¢aoH x an. % oo.m u x
. oOHH .QEmB oOHH .QEmB
N: mnmde m: mamde

mmHmOAmo<mBmB zomm<o 2H mB<QNXOmm<omMmmzdxmmoqowo qwfibmnp mo ZOHBHmOmEOomQ

-06

 

 

 

 

 

 

 

 

 

 

 

 

mma. oma .HH
00:. omH .oH
can. OHH .m 6:. OHN .m
0zn. OHH .w Nmi. Owfi .w
0mm. OCH .5 nm:. 00H .5
wom. 00 .0 mmm. o:H .0
dam. om .0 0mm. ONH .m
me. 00 .: mom. om .:
:mm. om .m «mm. 00 .m
mam. on .N NNN. on .N
00H. 0 _ .H wwa. o .H
.mcmme .CHE .maae wagzmm .mcmpe .CaE .mEHB mHaEmm
.-.... m-ofl x mm. « mo.o . x .-.omm m-od x mm. . ow.o u x
ooHH .asme ooHH .Qsme
m: mum<8 m: mqmde

mQHmoqmo<mEme zomm<o 2H m9<waomm¢OMMmmz<mommoqowo AMBDmup mo ZOHBHwomzoomm

 

x ‘gluln. I . ,
It». “I y I... I. IV 4 .
F ill. ‘LI E. a I ‘I

 

 

107

 

 

mmm. m» .mH
mfim. .0 .HH
#5:. 00 .oH
0H3. mm .9
max. Hm .w
mmm. on .w
m3m. mm . .0
non. .0m .m
owm. mm .:
#:m. Ma .n
mmm. NH .m
oma. o .H
.owpB .CHE .wEaB mamamm

 

 

 

a-.omm .-ofl x.:~. « mo.m u x
coma .meme

 

Hm mqm<e

 

 

m:m. :0 .0
mam. NF .w
0mm. mw .0
«0:. 00 .m
awn. w: .:
mzm. 0m .m
wmm. NH .m
mom. 0 .H
.mCmpB .CHE .mEHB mHQEmm

 

 

 

.....m ..OH x :m. « No.m . x
coma .Qsme

 

om mqm<9

mQHmOAmo<mBmB zomm¢o zH MB¢ANXOmm<Ommmmz<mommoqONO Awebmlp mo ZOHBHmOmEOOMQ

\ .I I. .1111: Inlr

n

08

..

 

 

 

 

 

 

 

 

 

 

 

 

 

00:. m: .N

wan. NH .0

mg:. mm .m mmm. N: W .m

mfld. om .: mam. H: m .:

00%. MH .M 0mm. 0 .M

.mHm. 0H .m m0m. m .m

Hmm. m .H mam. o ..H
.mCmpB .CHE .mEHB maqsmm .mcmpe .CHE .mEHB mfimsmm
an.omm vIOH K OH. fl mm.: H x Hu.omm :nofl x Om. fl mw.: u x
0mm: .QEmB coma .QEmB

mm mnm¢9 Nm mam¢9

mQHmOAm04MBMB zomm<o ZH m9<qwxomm<Ommmm2¢mommoqowo qwfibmlu mo ZOHBHmOmEOomQ

InfilalIllllullll r). $lv‘z lt..ll...lr’|.t .‘\ I1 .I-Iul..l|. .F Ill!

CHEMISTRY -me‘m w

CHEMISTRY LIBRARY

 

 

 

 

   

M'CIHIIWINHIW WWIiiWHVIIEIIWIWIILHWI'ITS
3 1293 03046 4840