rML M1.
1..
La. Rafi
. 1....
he
. . 1
.63. witu.
Eu . e
mt. .‘u
PC
a. u. .w
3.... ‘ .
n J u
2,» F ..
. .
3 :“
fie- .” .1
1.: ....
3 ‘ A
.
k.

g»
a
b?-

o
*3.
_

H:'
v :..

bag-:- 1*?

II

 

 

 

 

{H5515 ( LIBRARY

Michigan Stale
University

 

r1... .
«.st

m A. U

 

 

 

ABSTRACT

PRODUCTS OF THE REACTIONS OF CERTAIN 2, 6- DISUBSTITUTED
ARYL GHLOROFORMATES WITH SILVER NITRATE

by Trudie S§.'Barreras

This investigation primarily involved an examination of
the reaction products obtained when several 2,6-disubstituted
aryl chloroformates were allowed to react with silver nitrate
under various reaction conditions. I

The products obtained from the reaction or 2,6-dimethyl-
phenylchloroformate with silver nitrate were 2,6-dimethyl-4-'
nitrophenol and 3,5,3',5'-tetramethyldiphenoquinone-(4,4').
From the 2,6—diisopropylphenylchloroformate, the analogous
2,6-diisopropyl-4-nitrophenol and 3.5,3'.5'-tetraisopr0py1-
diphenoquinonef(4,4') were obtained plus an additional product,
a yellow crystalline substance which is.not yet conclusively
identified, although considerable physical and spectral data.
were obtained in an effort to elucidate its structure. 'The
2,3,5,6—tetramethylphenylchloroformats yielded the expected
4-nitrophenol plus 2,3,5,6-tetramethquuinone, but no coupling
products were observed. . '

Variation of reaction conditions, including change of
solvent from acetonitrile to nitromethane, seemed to have little
effect upon the appearance of the various products, though the
relative proportions of each were somewhat influenced. At
temperatures below 0°, very little reaction was observed to
occur, and no products could be isolated. variations of temp-

erature in general affected the quantities of the 4-nitrophenol

Trudie S. Barreras

obtained most significantly.
In addition to this productnstudy an effort was made to

determine conditions under which a semi-stable intermediate
might be formed and isolated. Also, a possible means of

trapping Nog'was investigated briefly.

PRODUCTS OF THE REACTIONS OF CERTAIN 2,6-DISUBSTITUTED?
ARII. CHIDROFCRMATES'ZWITH SILVER NITRATE

By.
Trudie 8. Barreras

A TI-IEBIB’ ‘_

2 submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of
MASTER OF SCIENCE

Department of Chemistry

1965

ACKNOWLEDGEMENTS

The author wishes to express her sincere appreciation,tol
Dr. Robert ngachuetz for his guidance in this investigation.

Grateful acknowledgement is also extended to Dr. Matthew
J. Zabik fer his extreme generosity with experience and equip-..
ment, to Drs. William H. Reusch and Gerasimos J. Karabatsos
for their kind assistance in interpreting spectra, and to Dr.
Ralph Lu Guile and his research group for their assistance in
determining molecular weight.

Finally, deep gratitude is expressed to Raymond J.
Barreras, without whose patient encouragement as both.husband

and colleague this effort could never have been.concluded. .

11

t Tumormmmm

INTWDUOTION C O O . C O O C O C '. O. O I O O O O 0 O O 0
RESULTS AND DISCUSSION . . . . . . . .-. . . . . . . .
I. Preparation of Aryl Chloroformates . . . . . .

11. Efforts to Determine Conditions Under Which a
Possible Intermediate Might Be Formed. . .'. .

III. Isolation and Investigation of Products. . . .

IV. variation of Reaction Conditions to Determine
the Effects on PrOdUOt Fbrmatlon 0'. o o o 0.0

V. Trapping Experiments Attempting to Determine

Whether Ionic Species were Involved. . .‘. .

EXPERIMENTAL o ..-...-,.. o o o o o o 6 o o o e e

I.
II.
III.
IV.
v1
VI.

VII.

VIII.

SUMMARY.

Reagents o o o o o o o .‘. o o o o o 0
Preparation of Phenylchloroformates. .

Preliminary Experiments. . . . '. . . .

Investigation of By-Products . . . . . . . .

Products from the Reaction of 2,6-Diisopro-

pylphenylchloroformate . . . . . . . . . . .

Identification of Products from the Reaction

of 2,6-Diisopropylphenylchloroformate. . . .

Products from the Rearrangement of 2,3,5,6-

Tetramethylphenylchloroformate.. . . . . . .

Trapping Emperiment. .-. . . . . . . . . . .

LITERATURE CITED . . . . . . . . . . . . . . . . . .

~ 111

19
20
2O
2O
22

24 '

27

30

31
35
36
37

na-

SPECTRA
1.

2.

3.
4.

5.

6.

LIST OF SPECTRA

N.M.R. Spectrum'of 2,6-Diisopropy14
4‘111t1‘0Ph9H01oooooocouscous.o

N.M.R§ Spectrum of 3 5.3’.5'-Tetraisopropyl-
diphenoquinone-(4,4'S '

N.M.R. Spectrum of Uhknown‘Yellow Material. .
Infrared Spectrum of unknown Yellow Material.

Infrared Spectrum of 2,6éDiisopropy1-
PhenlethI‘Ofomate o o o o o o a .o _o o o o 0

Infrared Spectrum of 2.3.5.6-Tetramethyl-
quinone (Duroquinone)_. . . . . . . . . . . .

iv

Page

11
13

'03”

LIST OF TABLES'

TABLE - Page if
I. Results of Preliminary Experiments ... . . . . . 25

II. Reactions of 2,6-Diisopropylphenylchloroformats
withSilveertrate............... 29

III. Reactions of 2, 3, 5, 6-Tetramethylphenylchloro-
formats with.Silver Nitrate. . . . . . . . . . . 34,

INTRODUCTION

Little had been reported previously in the literature cone
cerning the reactions of aryl chloroformates with silver nitrate.
An initial extensive investigation.of these reactions was re-
cently undertakenfiby Matthew J. Zabik of these laboratories,
as the research for his doctoral thesis. The present work,
reported here, represents further research on the chemistry of.
aryl chloroformates in an attempt to elucidate in.more detail
the nature and mechanisms of the unusual'rearrangement reac%
tions of these compounds.

A As the previous work, in these laboratories, had given
some indications that the reaction of an aryl chloroformate
with silver nitrate might initially produce a semi-stable in-
termediate, an attempt was first made to determine the experi-
mental conditions under which such an intermediate might be
formed and isolated. This was done with the anticipation of.
then'being able to conduct studies involving isotopic label-'
ling of reactants to elucidate in detail the mechanism of the
rearrangement reaction of aryl chloroformates. These efforts
were unsuccessful. Hewever, during some of the reactions,
the appearance of seemingly anomalous products-led to a de-
tailed investigation of properties and structures of these
materials which isthe central subject‘matter of the present?
thesis.

RESULTS AND DISCUSSION

5 0

The investigation of the reactions of aryl chloroformates
with silver nitrate initiated by‘nz'abik in these laboratories
and continued in the present study represents the only inves—.
tigation to date to attempt to elucidate the nature of these.
reactions. Afimajor phase of the experimental effort was thus
directed to determining what products actually resulted from
these reactions and the methods of purification.most applicable
.in each.case. . . .

The experiments described in this paper can be placed into
five general categories; preparation of the aryl chloroformates,
efforts to determine the reaction conditions under which a. _
possible intermediate might be isolated, isolation and inves-
tigation of the reaction products, variation of reaction cons
ditions to determine their effects on product formation, and
trapping experiments to determine whether ionic species-were

involved in these reactions.

I. {reparatfon_g§ Arzl galoroformates

Since the methods used for the preparation.of the various
aryl chloroformates used are described at length in the experi-
mental section of this thesis, and are essentially those de- .

‘ veloped by Zabik and described in his doctoral dissertation,
they will not be elaborated on further here. Nb unusual re-
sults were encountered in.these reactions, though a nuclear 7
magnetic resonance spectrum determined on the neat 2,6-diiso-

propylphenylchloroformate for the purpose of’comparison.with

2

U!

the products actually isolated indicated that.the distillatién.
procedure employed for purification may not have successfully
eliminated all residual phenol. This may be pertinent since

this particular chloroformate gave rise to an additional pro-
duct apparently not encountered in the analogous reactions of

other 2,6-dia1kyl phenylchloroformates investigated.

on. Ef____f____orts to Det___e_____rmine Conditions under M a Possible
Intermediate Might.§_ Fermed

Studies of the reactions of alkyl chloroformates with.‘
silver nitrate indicated that two distinct steps occurred,

a—o-g-scl + AgNO3 '—-9539-15—-. He -ONO;_. + M

 

C

R-mg-onoz > R-ON02 + 6020

These steps occurred essentially simultaneously at normalfl
temperatures, but the second step was considerably slowar
'than the first below room temperature (2). It was anticipated
that in the ease of the aryl chloroformates, presuming that
the reactions proceeded via an analogous intermediate, step:
one might be considerably faster than the second step at low
temperatures. If this were the case isotopic labelling studies
involving the 2,6-dialkyl phenylchloroformates could have been
undertaken to possibly elucidate the mechanism of this rear-
rangement reaction. Though no exhaustive survey of possible

” reaction temperatures was conducted, reactions allowed to

proceed for times varying from two to five hours at ~30° and

er

-15° involved a minimal precipitation of silver chloride, such
precipitation‘being the only currently available indication of
the extent to which.the first step had proceeded. It was theree
fore assumed that virtually none of the indicated intermediate,
ArCOONOQ, waefformed. In any event, it was apparent that
‘extremely long reaction times would be required under these
conditions, and it was felt that the technical difficulties of'
maintaining these low temperatures for such periods would.be
prohibitive. . '

One other possibility was considered, however; if an ini-
tial step involved the formation of some kind of complex between
the silver nitrate and the chloroformate, and inthis had been
formed extensively in the cold, then perhaps the nitrophenol
from this material would still be the predominant product even
though another chloroformate was added before the silver chloride
precipitated and the reaction proceeded to completion. With-
this idea in mind, at the conclusion of each reaction time a
quantity of a second aryl chloroformate was added to the reaction
.mixture and the reaction was forced to completion.by fairly rapid
warming. While no extensive product analysis was undertaken,
it was evident that a fairly complex mixture of products resulted
in.most of the reactions, and no predominance of the product
expected from the initially used chloroformate was encountered.
JBased on these observations efforts to determine experimental
conditions conducive to the extensive formation of intermediates‘

'were abandoned in order to examine in detail the nature of the

as.»

complex products actually formed in these reactions of aryl 5”
chloroformates' with silver nitrate. ' _

It should be noted that in'e subsequent series of experi- ‘
ments (Table II) the reaction.of 2,6-diisapropylphenylchlorofor-
mate conducted fer three days in the cold failed to result in
the formation of any isolatable products; in fact unreacted
silver nitrate was recovered from the reaction mixture at the
end of this period. Hence it is evident, especially in the
case of the 2,6-disubstituted phenylchloroformates, that attempts
to obtain any appreciable reaction at relatively low temperatures
would necessitate either some form of catalysis or else exceed-.

ingly long reaction periods.

III. Isolation and Investigation of Products
In the course of the preliminary experiments aimed at iso:

lating intermediates as discussed above, it was observed'in.one
case that an unexpected red crystalline material precipitated~
along with the silver chloride from a reaction involving 2,6- .
dimethylphenylchloroformate and silver nitrate. This material,
' while sparingly soluble in many common organic solvents, was
recrystallized fairly-satisfactorily fromIbenzene or ethanol.
This material was identified by quantitative analysis and com;
parison of the physical properties with the known 3.5.3'}5'-
tetramethyldiphenoquinone-(4,4'), (3).

An analogous quinone was expected-to result from the reaction
of the 2,6-diisopropylphenylchloroformats and was indeed isolated.

t
t 3

However, far from being insoluble in the reaction solvent and” ‘
coprecipitating with the silver. chloride, it proved extremer
soluble not only in acet'onitril'eflbut also in hexane and ethanol.
The product composition resulting from this reaction was fur- .
ther complicated by the presence 'of a third product, a bright
yellow crystalline solid which had solubility properties simi- .
lar to those exhibited by the quinone. Numerous efforts were
made to separate these materials 'by fractional recrystalliza-
tion, many of which resulted in intimatemixtures of red and
yellow crystals of various sizes. In some cases these crys-
talline products formed in crystals of sufficient size to be
partially separable mechanically. This, however, was hardly

a practical procedure. In several other cases the reaction
products took the form of a dark red'cdlored syrup which was

a mixture of the two substances plus other impurities. .Even-
tually a procedure for chromatographing the combined reaction
products on a silica gel column was developed which gave a '
relatively satisfactory separation, even though a certain amount
of material appeared to decompose on the column. Unfortunately
this technique was. developed after most of the work discussed
in the experimental section had been concluded 'and the products
had been roughly separated by crystallization procedures. For
this reason the actual percentage yields of the specific pro-
ducts obtained from any given reaction can only be approximated.
Further, since the structure of the yellow crystalline material
must still be considered as not yet fully confim'ed,: any yield

l
t
c

calculations must of necessity depend upon an assumed molecular)
weight. However, postulating this substance as a dimer of I
molecular weight approximately hDO, it is estimated to be
formed in at least a 6-10% yield in all.reactions conducted in
acetonitrilexat room temperature or above. ‘The 3.5.3'.5'-tetra—
isopropyldiphenoquinone-(4,4') was isolated in a smaller yield
under these experimental conditions; the maximum yield of this.
material obtained being about 5%. The maximum yield of the
expected product, 2,6-diisopropyl-4-nitrophenol, was 60%, ob-
tained from a reaction conducted at 55° fer a short.period of
time followed by immediate product isolation. It is interesting
to note, however, that the yield of the expected product obtained
seemed to vary the most significantly with varying reaction.cone
ditions, while the proportions of the other materials appeared
to remain fairly constant. . '

The expected product, 2,6-diisopropyl-4-nitrophenol wasr
identified by its phenolic properties and from the known.lite-I
rature value}of its melting point (4). The infrared and ‘
nuclear magnetic resonance spectra of this compound were as:
expected (Fig. 1). h

The 3,5,3',5'-tetraisopropyldiphenoquinone-(4,4') iso-
lated in the reactions described above was identified by
quantitative analysis and its analogous properties to the pre-
viously discussed 3,5,3',5'-tetramethyldiphenoquinone-(4,4‘).
Spectra were determined and its nuclear magnetic resonance

was exactly that anticipated for this structure (Fig. 2).

‘ C
t . .

Both of the products gave n.m.r. spectra typical of
symmetrically substituted aromatic compounds; They showed
sharp doublets representing the methyl protons on the iso-
propyl groups, a septet for the tertiary isopropyl proton,
and singlets for the aromatic (and phenolic) protons. Inn
view of this fact, it is considered quite significant that
the n.m.r. spectrum of the unknown yellow material differed
radically from this pattern (Fig. 3). The splitting in the.
region of the methyl protons, which is clearest and easiest}
to determine, indicates that there are definitely three none
equivalent types of isopropyl group present in the structure.
Further analysis of the spectrum and a determination of the
area ratios of each peak indicates that there are two iscpropyl
groups whichrare equivalent, and two which differ from these
and from eachother. In addition, on the basis of twenty-four '
methyl protons on the four isopropyl groups, the spectrum O
accounts for a total of nine others:‘ four isopropyl protons'
and five in the aromatic region. On the'basis of this evidence~
it is necessary to postulate a dimeric structure of some nature,
which is unsymmetrical with respect to both the isopropyl groups
and the aromatic hydrogens.v

The results of elemental analysis are not as conclusive as
had been hoped for,‘but they strongly support a proposed di-
meric structure, with a minimum Csfl ratio of 24:1. The mole-
cular weight data is rather inconclusive, but since the come

pound appears to be unstable above its melting point, all

.1 a . HocmzuouuuCIeuaxaounonaaouo.N do sauuumam mucmcommm uuumcoos humans: .« shaman

 

 

 

. . Shaun ~c..m 2.6 . 1M . 5.3.. . ..o.n.~
Lu“ L" u of o 1I1+ I . I . 0
9 .

 

 

 

 

10

A. e .evum.cocqoaocmnnwu

ua>aouaonuouumpn.m. .n. m. n no Eauuumnm mocmcommm o«.umcmma nomaoaz .N shaman

mm. m NM

I
*

 

 

m . . mm.c. . .. .. . menu
4

flees

.
b
d

 

 

 

11

Hedgehog so-m> ceocxc: do eauuomnm couscous:

unamcmos unmauaz

.n muaowu

 

a...

 

 

 

 

 

 

 

don 2.6 . . «To in. ..oo.n.$..~ 2...”
a u T . n n L. a

i

analytical and instrumental methods involving the use of ale-pv-

vated temperatures would be expected to give questionable re;

sults. The osmometric method, which is expected to give a ten

percent error, strongly indicates a molecular weight of 400 or‘l

above, supporting the dimeric structure.

A structure which reasonably satisfies the above evidence,

and particularly the n.m.r. data, is,

- Molecular formula
0 ' ' 024333N04.

 

It is unfortunate that the unusual and complex nature of this

structure makes it difficult to find models for the purpose of

comparison of the infrared and ultraviolet spectra. There is
an unusual absorption at 1805 cm"1 in the carbonyl region of
the LR. spectrum of the material (Fig. 1+), a'definitive inter-
pretation of which could aid significantly in a clarification
of the structure. This absorption is so close to the typical
absorption.of the chloroformate itself at 1795 cm“ (Fig. 5)
that repeated elemental analyses were made to determine that:
halogen was definitely absent.' In any event, it is suggested
that if such a carbonyl exists, stearic and polarity effects .
of the two groups in the ,9-position might indeed yield unusual
spectra. ‘

In the absence of specific chemical evidence, which was
not sought as part of the present investigation, other struc-
tural possibilities cannot be completely eliminated. Some of

12 ,

Hedgehog sedan» caocxc: do sauuowmm omuuuueH .e muamuu

. U . .
.oom coop Damp Fla coca Door acme Loewe
n. a P n n n .

 

13

a. . .w J
.o

l . . vow

 

 

 

 

 

 

 

no

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

14'

 

 

ouoeuoaoueanoaxcm:naxmounomwwono.N.uo.e:uuumam monogram

oops,

 

 

3

 

 

 

new.

 

P-

 

 

 

Eu

I‘
.

new.

 

 

 

 

 

090%

.m shaman

camp

2

 

 

IDQ
pm

you

 

‘t 15 ’-

. , .
these structures appear to be more satisfactory if only partsf"
of the data are considered. rHowever, the structure proposed
above seems at present to be the one best supported by the .
totality of the evidence and especially.by the fairly clearly '
interpreted ngm.r. spectrum; .'

The formation of these unusual compounds in the course of
the reactions of the 2,6-dialkyl phenylchloroformates with
silver nitrate leads naturally to speculation concerning the
mechanism by which they are produced. Although.no experiments
were undertaken in the course of the present investigation to
establish the possibility, it is felt that a homolytic cleavage.
of the bonds in.the hypothesized nitrate ester intermediate,”
leading to a phenoxide free radical, could be considered,.

0- a-o-N'o2 o- '

, etc.

 

These radicals are known to be stable and have been.demonstrated
to couple giving the type of quinone encountered (5). Although?
in his investigations zabik failed to find specific evidence of‘
free-radical formation, the major efforts of his study were
concerned with rearrangements to the ortho positions. In these
cases the probability of a concerted intramolecular rearrangement
is greatly enhanced, while the stability of a possihle phenoxy
radical is decreased. One further indication favoring a possible
free radical mechanism is the consistent appearance of oxides

of nitrogen being evolved during the course of these reactions.

f

Ifafree radical mechanism is assumed to explain'the ‘
appearance of the quinones, it can probably likewise be usedd
in 'a modified-.‘way to discuss the" formation of the third product
from the 2,6-diisopropylphenylchloroformate reaction, since ' ‘
this also seems to involve a dimeric structure, suggestive of
coupling of some sort..

The appearance of the yellow crystalline unknown in the
reaction of the 2,6-diisoprcpylphenylchloroformats led immedi-
, ately to speculation as to why a similar'prcduct had not been
encountered in the case of the previously investigated 2,6-di-
methylphenylchloroformate. An immediate possibility seemed to
be that the extremely stable crystalline structure of the.
3,5,3',5!~tetramethyldiphenoquinone-(4,4'), which rendered it
so insoluble-in the reaction mixture, might energetically favor
this by-product to such.an extent that any analogous reaction.‘
could not compete. The apparent way to overcome this diffi-.
culty was to try to find another compound which closely re-
sembled the 2;6-dimethylphenylchloroformate in.properties,
but which could not effectively form a diphenoquinone.‘ The:
obvious choice was durenylchloroformate.(2,3,5,6ftetramethyl-
phenylchloroformate)..This compoundeas therefore synthesized.
and allowed to undergo reaction with silver nitrate under the
usual conditions. The only product which was obtained in addi-

ticn to the expected nitrodurenol was duroquincne, ‘

 

a ‘7 ’.

1- , .
which may likewise be rationalized in terms of a free radical, ‘
formation and subsequent oxidation..Iwaever, this does not '
offer the desired experimental evidence in rationalizing the-
nature of the unknown material obtained from the 2,6-diisopror
pylphenylchloroformate rearrangement. ‘It should be pointed out,
however, that the possibility of the presence of a third product
was not absolutely eliminated either in the case of the 2,6-did
'methylphenyl- or the 2,3,5,6-tetramethylphenylchloroformats' .
rearrangements. In both situations far lessrthan 100%’cf the
reactant was accounted for in the products, and due to the
crystallization procedures generally undertaken for purification,
it is possible that a highly soluble third substance might
easily have'been lost in product isolation, especially if it
was present in extremely small quantities. .

A fruitful area for future investigation of these reactions
would be careful, detailed product analysis to determine the-
exact amount of each substance obtained under specific reaction
conditions. in addition it would be informative to prepare. I
various para-substituted 2,6-dia1kyl phenylchloroformates and
'allcw these to react with silver nitrate. To date this has
only been attempted with the 2,4,6-trimethylphenylchloroformats,
and though the major product from this reaction was identified
as 3,5,3',5'~tetramethylstilbenequinone-(4,4‘), an extremely
complex product mixture was obtained (1). However, since .
interesting coupling products have been reported from a pro-

cedure which.generates phenoxide radicals from various

18 I

an"
.

2,4,6-trisubstituted phenols (5). it would.be interesting tojh
investigate the extent to which the chloroformate reactions I
might give analogous products. EL

1?; Variation of Reaction Conditions to Detem mine them mfegts

on.Prodggt Formation .

It should be noted that at no place in the present study‘
were genuinely ”standard" reaction conditions established, nor.
was an attempt made to systematically change one variable, such.
as time or temperature, in order to determine the exact effect
of an individual change of this sort. However, examination of
Tables II and III demonstrates that in reactions conducted at.
temperatures of 25° or above, and allowed to proceedfor times
varying upwards from three hours, the same products were always
obtained, though in varying quantities. This has already been
dealt with in the discussion of yields in Section III above.

Additional variables examined briefly were purity and
nature of solvent. Though these results are not conclusive '
kdue to the difficulties of product separation already mentioned;
examination of entries (1) and (4) in Table II indicates that,
other conditions being analogous, using anhydrous, distilled
acetonitrile in place or that taken directly from the supplier's
container without any purification increases the preportions of
the by-products somewhat with respect to the nitrophenolo‘

Changing the solvent from acetonitrile to nitromethane does
not seem to prevent formation of any of the products, though the
quantities of each.are reversed; the diphenoquinone is obtained

. 19 ' ,
t . .
in.greater than 10% yield from the 2,6-diisopropylphenylchlono-l
formats reaction, while the unknown yellow material is onlyflf I
present to the extent of about 3% and the nitrophenol, while
still the major product, was obtained in only a 23% yield.

This reactionzwas catalyzed-by.the addition of several drops
.'or pyridine, further modifying the conditions.

Finally, it is observed that reactions conducted in the
dark except for mixing, occasional momentary checks on.progress,
and the time necessary for removal of solvent, showed no sig-
nificant variation.of products. , '

In summary, the major influence on the reaction seems to
be temperature; in general, in the case of the 2,6-diisopropy1-
.phenylchloroformate reaction,-the higher the reaction tempera-
ture the greater the total product, mainly representing the
expected 2,6-diisopropyl-4-nitropheno1. Within the limits of .
experimental error, changes of temperature and other variables

seem to have relatively little effect of the appearance or

quantities of the by-products.

v. Trapping magenta Attempting Lg Determine Whether Ionic
Species were Involved

These efforts, which are discussed in detail in the experi-
mental section of this thesis, gave inconclusive results which»
in view of more extensive work along the same line by zabik, I
did not seem to offer a particularly fruitful area for further

investigation.

1 ' EXPERIMENTAL

5’ ‘
‘

.'
‘I

1. Reagents
All reagents were used as obtained from the supplier withe

pout additional purification, with the exception of the aceto- .-
nitrile designated "anhydrous." ,This was distilled from calcium
hydride through.a one-foot column packed with glass helices,

and the fraction.boiling in the range 80. 5-81.5° was collected t

and stored over calcium hydride until use.

II. Preparation of Phenylchloroformates

The chloroformates used in this study were prepared using
the general procedure described by Zabik (1).

In the case of 2,6-dimethylphenylchloroformats, 50.0 g.w
(0. 41 mole) of 2 ,6-dimethy1phenol was dissolved in 225 ml. of
benzene and placed in a 3-necked round—bottomed flask fitted
with an isopropyl alcbhol-dry ice condenser, an equilibrating °
dropping funnel, and a gas inlet tube. All connecting tubing
was glass or tygon. tA set of traps, one containing a solution
of sodium hydroxide in methanol, were connected to the top of °'
the condenser. Approximately 55 ml. of phosgene, previously;
liquified by passing the gas from a cylinder through a spiral
condenser immersed in an isopropyl alcohol-dry.ice slush, was.
added rapidly through the gas inlet tube, which was then re-
placed by a glass stopper. Then 54.7 g. (0.45 mole) N,Nrdi-
methylaniline was added dropwise to the stirred dialkyl phenol
solution via the dropping funnel, while maintaining the reaction

mixture at its reflux temperature. Fbllowing each addition of

.20

21

nP‘

a quantity of the base, pronounced color changes in the reaction
mixture were observed; in this_case from light yellow to deep
blue-green. The rate of base addition-was regulated so that

. the color nearly faded to pale yellow‘between each addition.
Immediately following the final addition of the aniline the
reaction.mixture crystallized to a bright-yellow solid. The
reaction mixture was hydrolyzed by cautiously adding 10 ml. of'
distilled water dropwise.' During the hydrolysis gas was
vigorously evolved, heating of the mixture was observed, and
the solid dissolved to yield a yellowborange colored solution.
This was initially washed with dilute hydrochloric acid, then
with dilute sodium hydroxide, and finally with distilled water.
The benzene solvent was dried over anhydrous magnesuim sulfate,
removed by distillation, and the pure product was isolated by
distillation under vacuum. ‘A yield of 45.99 g. (0.25 mole, .
60.5% theoretical) of material boiling at 73-760/3 mm. was
collected. niterature value (6), b.p. 71-720/5.5-6.o mm.

In a similar preparation of 2,6-diisopropylphenylchlorc-
formate, an immediate solidification of the reaction mixture on
addingrthe last or the N,N-dimethy1aniline failed to occur and
the reaction.mixture'was stirred until refluxing had ceased
and then it was hydrolyzed, washed and dried as described above.
Following removal of the benzene solvent the crude product was
distilled to obtain 68.1 g. (0.28 mole, 69% theoretical) of a
material boiling at 108-112°/3 mm.

panethylphenylchloroformate was similarly prepared and

22

sor-

obtained in a 69% yield. Since these chloroformates had‘been' '
previously synthesized and reported (1), they were used without
further analysis or purification.- ' .

In the preparation of 2,3,5,6-tetramethylphenylchloroformate,
34 g. (0.22 mole) of the phenol was employed with proportionate
amounts of the other reagents. The reaction mixture took on a ‘
»reddish€brown coloration during the addition of the N ,N—dimethyl-
aniline, turning yellow and solidifying following the final.
addition of organic base as in the case of the 2,6-dimethyle
phenol. Following washing of the reaction mixture, drying and
removal of solvent, the product was found to be a solid. Cone °
siderable experimental difficulty in recrystallization of the
crude product resulted in a rather poor yield, but a minimum of
15.0 g. of fairly pure material was obtained on recrystalli-
zation from benzene. This product, after two additional re-
crystallizations from benzene, yielded a white flaky material
melting at 54762°. '

Anal. Calc'd. for Gi1H130201: c; 62.11%: H, 6.14%;_cl, 17.23%

Founds. c, 62.10%; a, 6.22%; 01, 17.25% ' '

III. Preliminazy Experiments .
In an initial series of experiments involving the reaction

of 2,6-dimethylphenylchloroformate with silver nitrate an
attempt was made to determine a set of experimental conditions
under which the hypothesized intermediate I would form with an
appreciable precipitation of silver chloride, but without the

23 . ’_

or)”

rearrangement of the intermediate with the accompanying evolution
of carbon dioxide and formation of the final products, ‘
0 , 0
O-G-Ol . 0330N 0-Cr0N02

 

u . i I
I A ; product(s) + 0021'

 

These first reactions were conducted at low temperatures and with
the addition of various other reagents. .In general the results
indicated that the rate of formation of the intermediate at'
these temperatures was too slow to be of practical experimental
value for mechanistic studies. ;

In.a typical experiment, 3.4 g. (0.019 mole) of 2,6ndimethyl-
phenylchloroformate (approximately a two-fold excess) in 75 ml.
of acetonitrile were cooled to -30° in a‘bromobenzene-dry ice .
bath and mixed. The reaction was allowed to proceed with.stir—
ring at this temperature for two hours, but no appreciable 1
clouding of the solution or precipitation of silver chloride
was observed. A.1.7 g. (0.011 mole) quantity of phenylchloro-
formate in 25 ml. acetonitrile was added to the cold reaction
mixture and the reaction warmed to room temperature. An imme~
diate cloudiness appeared in the mixture and a few'bubbles of
gas escaped. Further heating brought a heavy precipitation of
silver chloride, and the reaction mixture turned yellow, then

brown in color. After removal of the silver chloride by fil-
tration and the solvent by distillation, the product was

24 r

0""

distilled under vacuum to obtain a mixture of liQuid fractions.
‘boiling ca. 70-80°/ 4 mm as well as some solid 2 ,6-dimethyl-
4-nitrophenol, m.p. 165~169° after recrystallization from a

4:1 ethanol-water mixture. Literature value (7), m.p. 169-1700.
No further separation or identification of products was ate
tempted. Table I gives aschematic representation of the re-.

sults of a number of these types of experiment.

IV. Investigation.p; gx-Ezoducts
The appearance of the red crystalline solid in preliminary

experiment four (Table I) led to an effort to determine whether
this unexpected product was obtained when the reaction was
allowed to proceed in the absence of any interfering reagents

as well as in their presence. Two parallel reactions were cone
ducted. In the first, 4. 63 g. (0.025 mole) of 2 ,6-dimethyl-
phenylchloroformate in 50 ml. acetonitrile was mixed at room
temperature with an equivalent molar amount (4.25 g.) of silver
nitrate in 50 ml. acetonitrile, and the reaction was allowed to.
proceed with stirring for 72 hours. At this point the solution
was light yellow in color and some dark red crystalline material
‘hsd precipitated along with the silver chloride.‘ The red solid
after recrystallization from benzene gave 0.07 g. of a material
melting with decomposition in the range 205-2250. The remainder
of the reaction mixture was heated to remove the reaction sol-
vent. However, since it was observed that N204 fumes were

being evolved, heating was discontinued and the remainder of .

Table';: Results 9: greliminarx_§xperiment§

25 /

Reagents in'brackets were mixed at the initiation of the
reaction and allowed to react at the given temperature '
for the recorded time; the observations refer to the reac-
tion mixture prior to the addition of the other designated
reagents. ,

Reagents: I, 2, 6- -dimethylphenylchloroformate: II, phenyl-
} chloroformate; III, phenol. All in ca. 50 ml.
a! acetonitrile solvent.

Product: IV, 2,6-dimethy1-4-nitrophenol .

Reagents Temp. °§T Reaction Time hrs. Products
.1. 3.4 g. I i -30 4 2: Obs: No IV + liq.
' olouding_ mixture,
1.7 g. AgNO, b. p. 70- 80°/
4 mm.
1.7 g. II
2. 1.5 5. II ' ~30. 5; Obs: no pre- : Liq., b.p.
' cipitation 700/5 mm.
1.7 g. AgNOB . oily resi-
107 8. IY ‘ . 30116. .
30 107 80 I “15 1"; Obs: Slight (1., bop
discoloration 63-68°/3.5 mm.
1.7 g. AgNOB A and clouding +IV
1.5 g. II I
4. 1.7 g. I Rm. tem .' 2.5; Obs: preci- IV + red
' (ca. 25 ' pitation crystals +
1.7 g. AgN03 _ tarry resi-
ca. 40 - 4; filtered, stored due
1.0 g. III ,' ' ,

26 /

9-4-

\
9

the solvent was removed under vacuum on a rotary evaporatoraj
The residue was crystallized from.a 4:1 ethanol-water mixture,
and an initial yield of 0.57 g.‘(0.002 mole, 15% yield) of
2,6-dimethyl-4-nitrophenol melting at 164-1670 was obtained.

.Simultaneously with the above reaction a second reaction
containing the same proportions of chloroformate and silver
nitrate, but with an added 2.00 g. (0.021 mole) of phenol, in
a slightly more dilute solution (150 ml. acetonitrile) was
allowed to proceed. The latter reaction.mixture turned brown
in color and after 72 hours of reaction a somewhat larger pro-
portion of the red crystalline material (0.10 g.) had precipi- ,
tated from the solution. Following removal of the reaction
solvent by rotary evaporation an oily brown residue was ob-
tained. This was dissolved in ethanol'and decolorized with,
Nerite. However, the filtrate remained dark and no solid could
be induced to recrystallize from it. .

The red crystals were submitted for analysis for carbon,
hydrogen and nitrogen.

Anal. found:: a, 79.36%; H,.6.80%; N, less than 0.2%. '-
These analytical results yielded, by calculation, the empirical
formula Ogflgo, and suggested the possibility of the dimeric
structure,

3.5.3',5'-tetramethyldi-
phenoquinone-(4,4').

 

The proposed structure was verified by referrence to Beilstein (3),

27 /

‘which contained a description of this compound as a sparingly”
soluble material which crystallized from solution in the form
of'dark red colored needles, melting with decomposition.in
the range 207e217o. Since the solubility in common solvents
‘was low, thexonly other information.obtained on.this compound

'was an infrared spectrum.

a

'V. Products from the Reaction of 2,6-Diisopropylphegylchloro-

formats

In a similar manner investigation was initiated to deter-
lnine whether analogous results would occur with other 2,6-dialkyl
jphenylchloroformates. 2,6-Diisopropylphenylchloroformate was-
examined using 0.025 mole quantities of the reagents (6.19 g.

of 2,6-diisopropylphenylchloroformate and 4.25 grof silver
nitrate) each in 50 ml. of acetonitrile. These reagent solu-
tions were mixed and stirred for six hours at 25° and then set'
aside at room temperature for an additional 48 hours. .At the
end of this time no red crystalline material was observed to
separate from the reaction.mixture. However, the reaction
solution had taken on a dark red coloration. The silver chloride
was removed by filtration and the reaction solvent removed under
reduced pressure in a rotary evaporator. The residue was dis-
solved in'boiling hexane. The hexane solution was allowed to _\
cool and filtered to recover the crystalline product. This
material was recrystallized a second time from hexane after
decolorization with Nerite to obtain 1.5 g. of a pale yellow
colored pure material in the form of needles, m.p. 114-116°.
Literature value (4) m.p. 2,6-diisopropyle4-nitrophenol, 112°.

28 x

*P'

The hexane filtrate from the initial crystallization was’ ‘
dark red in color. Removal of the hexane from this solution:
left a dark red, semi-solid residue. ,The residue was treated
with 95% ethanol, in which it partially. dissolved, leaving a
red crystalline solid. The latter was removed by filtration
and after further purification yielded material melting at
215-22o°. ‘

Concentration of the ethanol solution yielded a quantity
of bright yellow colored crystalline material as large rhombic
crystals melting at 145-1500. A second recrystallization of
this material from ethanol raised its melting point to 152-1559.

The reaction of 2,6—diisopropylphenylchloroformate with.)
silver nitrate described above was repeated under varioue
experimental conditions in an effort te determine the effects
these variations had on the products formed. Some difficulty '
was encountered in separating and analyzing these‘products,.
as both'by-products, red and yellow, seemed to be very similar
in their solubilities. Initial efforts to separate these
substances by fractional crystallization were only partially
successful. Table IIIsummarizes the results of these experi-
ments. All reactione described in Table II used 6.19 g. (0.025
mole) of 2,6-diisopropy1phenylchloroformate, and 4.25 g. (0.0254
mole) of silver nitrate. The products are designated I; the
expected product, 2,6-diisopropyl-4-nitrophenol3 II, the red-
'crystalline solid, identified as 5,5,3',5'-tetraisopropyldi-
phenoquinone: and II, the yellow material, m.p. 152-1550.

29

cor
0

Table II: Reactions of 2.6- Diisopropylphenylchloroformate with
Silver nitrate

 

Solvent Temp. &”Rxn. Time... _ Products

1. 100 m1. 25° 4' I: 1.5 g.
CH3CN 6 hrs. stirring- II: 0.07 g.

5 ' 48 hrs. standing ’ III: O.d0 g.

2. 160 ml. -1°mix1ng -

CH30N w/stirring, . -----------
72 hrs.

5. 100 ml. 0° 5 hrs. w/ ’ unreacted AgNO3,
CH3CN stirring; 67 hrs. - ; 0.7 g; No orga-
(anhyd. ) stored in cold . . nic solid

4. 100 ml. - 25° w/stirring' I: 0.95 g.

. (anhyd.) standing r 7' , III: 0.52 g.

5. 2 100 ml. 55° w/stirring :I: 3.35 8.
CH3CN 3 hrs. II: 0.05 g.'
(anhyd.) - Immediate work-up III: 0339 6.

6. 100 ml. 4 45° w/stirring: I: 2.58 g.
CH30N 48 hrs. in dark II: 0.10 g.

72 hrs. Stored III: 0030 80‘
in cold - .

70 100 ml. 350 W/Btirring ' I: 1028 60'

CH3N02 28 hrs. 36 hrs. II: 0.51 g..

1 stored in cold III: 0.15 3.

Comments

under experiment: 2. Products syrupy, difficult to
' recrystallize
3,4 Further precipitation of AgCl
during removal of solvent under
vacuum '
5. Further yield of red and yellow
products, but co-crystallized
7. Silver nitrate sparingly soluble in
nitromethane; 3 drops pyridine added
as a catalyst. Immediate pptn. 0f
AgGl. Darkening of 801' n, sudden
1 evolution of oxides of nitrogen
after 5 hrs. rxn. time.o Dark red:
crystals formed on AgCl during
storage

[Md .-

im' - “.41 -

 

 

30 , ,

ur

VI: Identification of Products from the Reaction of 2,6- Diigo-
pr opylphenylchloroformate _

Since the expected 2,6-diisopropyl-4-nitrophenol had been
previously prepared and reported by zabik in these laboratories,
no further identification of this material other than its
melting pointiwas employed, theugh n.m.r. and I.R. spectra
were determined for purposes of comparison.

The red crystalline solid melting at 215-2200 was postulated
to be the 3,5,3',5'~tetraisopropyldiphenoquinone-(4,4'). .A
pure sample of this material was submitted for quantitative'
analysis.

Anal. Calc'd for (124113202: 0, 81.90%; H. 9.14%

Found: 0, 80.96%; H, 9.14%.
The I.R. and n.m.r. spectra were determined to further confirm
the structure. '
Efforts were made to identify the unknown yellow product
melting at 152-155° using various analytical data and instrur
mental methods. ,A summary of these data follows.

Yellow Compound from 2,6-Diisopropylphenylchloroformate Reaction

1. Melting point 152-155o (Three recrystallizations from 95%
ethanol); some decomposition above melting point, melt be—
coming dark red. Total decomposition by 250°.

2. Elemental analysis: No chlorine present.

3. Quantitative Analysis: (Found values average of five

 

 

determinations)
Found Ratio x/N Cald. %, 024333N04
c: 73. 02% 24.3 72.15%
H: 8.18% - 32.5 8. 33%
N: 3. 50% 1.0 3 57%

31 ’

'3’"

Proposed Empirical Formula: C24H33N04
Calculated Mblecular Weight: 399.51
4. Mblecular'Weight Determinations
Rast (commercial): 221
Osmometer: 456 ' ,
mass Spectograph: No mass peaks above 194
5. Infrared”Bpectrum (Fig. 4).

6. Ultraviolet Spectrum: Low intensity peak, 413 my.
High.inteneity peak, 283 mu.

7. Nuclear Magnetic Resonance Spectrum in 0014 (Fig. 3)
A. Considering methyl peaks as three doublets, centered at:
t: 80 879 8083, 8077
(Individually,‘z’= 8.90, 8. 88, 8.82, 8.79, 8.75, 8.70)
B. Septet (7) centered at ZT= 6. 94 H

C: In Aromatic: Quartets (7) centered at i: 3.00, 2.74
Singlet t: 2087 .

Ratio A:B:C‘= 24:4:5

VII. Products from the Rearrang_ment of2 2,§,§,6- Tetramethy -
phenylchloroformate

In order to investigate the possible formation of other;
products whenlthe possibility of para coupling was greatly
reduced, the 2,3,5,6-tetramethylphenylchloroformate was pref
pared and allowed to react with silver nitrate. In three sepa-
rate experiments, conducted under various experimental conditions,
two products were_obtained. One was the expected 2, 3, 5, 6-tetra-
methyl-4-nitrophenol, identified.by comparison of its properties
determined in this work with those reported in the literature.

The material recrystallized twice from ethanol melted at 120-1249.
Literature value (8), m.p. 123-12#°. This compound was:verified

by carbon, hydrogen and nitrogen analysis.)

32
Anal} Calc'd. for C10H13N03: C, 61.50%; H, 6.68%; N, 7.22%
Found: a, 61.71%; H, 6.83%; 'N, 7.35%.

The other product isolated was a bright yellow colored .
solid which crystallized in the form of needles from ethanol or
hexane. Thexmaterial sublimed'above 90° and melted at 110-1140.
Quantitative elemental analysis gave the following values,

C, 72.88%; H, 7.40%; N, less than 0.3%. Calculation from these

analytical results yielded the empirical formula 05H50. A.

molecular formula 0,0H1202, corresponding to the structure .

ago

was proposed for the structure of the yellow product. This
compound, duroquinone, reported in literature as melting at
111° (9) and having a characteristic infrared absorption at'
1640 ch‘ (10)} (Fig. 6), was in fact verified as the one
obtained in the reactions of durenylchloroformate with silver
nitrate. _ '

Table III summarizes the several reactions of 2?3’5’6i
tetramethylphenylchloroformate and silver nitrate under varying
experimental conditions. I

O

 

.,\ll|lll
\ul.» '1 I133: I‘ll.

 

 

 

 

33

Amcocqaaounov mcocwnamxsumemuuohnm.m.n.N.ao.e:u»umum umumuecn .m manned

EU

 

 

 

 

mom . one. soup a- . one. can. can. some a.
b n n b p w p .

.uowm

 

AVOQ

 

 

 

a“

 

ism

 

1

34 7 I

Table III: Reactions 2; 2.3.5.6-Tetramethylphenylchloroformate

Compound I, p-NO2-durenol, Compound II, Duroquinone

Experiments'1 and 2 used 0.025 mole quantities of the
chloroformate (5.33 g.) and silver nitrate (4.25 g.)
Experiment 3 used half this amount of each reagent.

Reaction Conditions .' Products
1. Solvent, 100 ml. CH CN. I and II

Note:

2.

3.

25° w/stirring 48 s.
Stored in cold 24 hrs.

Solvent 100 ml. CH CN. ' _ Mainly II
45 w/stirring 5 hgs. , . .
Stored in cold ca. 60 days.

SOlvent 60 m1. CHBCNO I I: 0027 60
55° w/stirring in the dark
24 hrs. Stored in cold 4 days. II: 0.11 g.

In experiments 1 and 2 above, numerous separate portions
of product were treated with different solvents, sublimed,
extracted with base, and so on, in an attempt to find a
good separation procedure. Thus no accurate account could
be made of the actual quantities of product obtained.
However, it is estimated that in (1) the proportions were
approximately equal, while in 2 the quinone was defi-
nitely the major product. In 3 , the residue dbtained
by removal of the acetonitrile was dissolved in hexane,
and compound I recrystallized from solution. The hexane
was removed and the residue was treated with dilute
sodium hydroxide. That portion which did not dissolve
in'base was removed and recrystallized from ethanol.

35 ,

VIII. Trapping Experiment

One additional experiment was undertaken in an effort to
determine whether any free N02 was formed in the reaction of
aryl chloroformates with silver nitrate, This involved the
reaction of prmethylphenylchlonoformate with silver nitrate
in the presence of p-methylanisole. Two parallel reactions,
each involving 0.025 mole quantities of the chloroformate and
silver nitrate in 100 ml. of acetonitrile were initiated with
stirring. To one of these was added 3.05 g. (0.025 mole) of
p-methylanisole, and the reactions were allowed to proceed for
four hours. The reaction mixtures were then chromatographed
in parallel on silica gel slides. Although no cenclusive re-
sults were obtained, it appeared that both reactions had.given
rise to a fairly complex series of products, which were never-
theless similar enough that no specific assignment of any given
spot as due to a nitro-substituted p-methylanisole could.be .

made‘.

? (SUMMARY , .

1. The diphenoquinone by-products from the reactions of
2,6-dimethyl- and 2,6-diisopropylphenylchloroformate with silver
nitrate were isolated and characterized. I

2. An additional by-product from the 2,6-diisopropylphenyl-
chloroformatc reaction was isolhted and studied, though conclu-
sive identification of this material was not accomplished.

3. The previously undescribed 2,3,5,6-tetramethy1phenyl-.
chloroformate was prepared and characterized. The products of
the reaction of this material with silver nitrate were identi-
fied as the 2,3,5,6-tetramethyl-4-nitrophenol and the 2,3,5,6-
tetramethquuinone. .

4. The effects of variation of reaction conditions-on.the
proportions of the various products obtained were considered.

It was noted that temperature effects seemed to be the most:
significant, and that the quantities of the p-nitrophenbl
product obtained seemed to be the most susceptible to such
temperature changes.

5. Means of isolating a possible intermediate, as well as

a method for trapping any N03 which might have been produced,
were briefly investigated. .

3.6

1.
2.
3.
4.

5.
6.

7.

8.

9.

0".

LITERATURE OI TED

Matthew J. Zabik, Ph. D. Thesis, Michigan State University.
8. A. Mortimer, J. org. Chem’.,.gI,.1876' (1962).

'Beilstein, .1. 743

T6 J. Barnes and w. J. Hickenbottom, J. Chem. Soc., 126 ,
2 15.

Go Do 0001! and No Do Gilmour, Jo Org. Chem}, E5, 1429 (1960).

J. H. Barnes, M. V. A. Chapman, P. H. McCrea, P. G. Marshall
and P. A. Walsh, J. Pharm. and Pharmacol. 13, 39 (1961);
Chem. Abstr., 55, 15383h (1961).

Beilstein 6, 486.
Co E. III-sham and Go Go Hampson,‘J. Chem. $000 1222, 9810
L. I. Smith, Org. Syn.) Coll. Vol. II, 254.

10..R..J.‘.Bates, J. Cymerman-Craig, M. Moyle and R. J. Young,

J. Chem. 8000, 12:; , 3880

37

 

‘l ' . an: - . . 1
-' 1