Title Page
 Table of Contents
 List of Tables
 List of Figures
 Ionic reactions of CF2BrCFC1Br
 Free radical reactions of CF2BrCFC1Br...
 Free radical reactions of...
 Biographical data

Title: Some reactins of 1,2-Dibromochlorotrifluoroethane.
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00091325/00001
 Material Information
Title: Some reactins of 1,2-Dibromochlorotrifluoroethane.
Series Title: Some reactins of 1,2-Dibromochlorotrifluoroethane.
Physical Description: Book
Creator: Bentley, Floyd Edward,
 Record Information
Bibliographic ID: UF00091325
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000551274
oclc - 13310644

Table of Contents
    Title Page
        Page i
    Table of Contents
        Page ii
    List of Tables
        Page iii
    List of Figures
        Page iv
        Page v
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
    Ionic reactions of CF2BrCFC1Br
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
    Free radical reactions of CF2BrCFC1Br and other polyhaloalkanes
        Page 53
    Free radical reactions of CF2BrCFC1Br
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
    Biographical data
        Page 78
        Page 79
        Page 80
Full Text





June, 1957



ACKNOWLEDGMENTS . * * .* a. a * v

INTODUCTION . . . . . . 1


DISCUSSIONI .. . . . . 7


EXPERIl5nMTALI . . . . . . . . 30




DISCUSSION II .* . .. . . $53

lEPERIIEIITAL II . . . . . .. 65

SJitflARY a ........................ .. 73

LIST OF REFERENCES . . . . . . . . . 75

BIOGRAPHICAL DATA . . . . . . a. . . .a. 78


Lblo Title

1 The Products of Reaction Between CF2BrCFC1Br

and Basic Reagents * 0 a * * * * *

2 The Products of Reaction Between Vicinal

Dhalides of Fluorocarbona and I:ethanolic

Potassium Hydrod.de *. * * *

3 Products Obtained in Reaction of Halofluorocarbons

with Vethanolic Potassium Hydroxide in Presence

of Compounds Containing Carbonyl Group . .

4 Sumnary of Rate Constants for Reaction of

Dibarmides of Haloeluoroolefins with



. 17

. 21

Hydroxyl Ion *

5 Rate of Reaction of

Hydroxyl on ..

6 Rate of Reaction of

IHydroyl Ion *.

7 Rate of Reaction of

fydrxyl Ion .

8 The Products of the

of Bensoyl Peroxido

CF BrCCBr with

* 0 a 0 * *

CFrCFClBr with

CFBrCF2Br with


Thermal Decomposil



9999999999, r

9... 999999


at Reflux Temperature * * *56




Fig. 1. Second order plot: reaction of CP2BrCC.Br
with sodiut hydrodde in watar-dioxno (1:2) . *.. 49

Fig. 2. Second order plot: reaction of CFp2BrCFCBr
with sodium hydroxide in water-dioxane (1:2) 51


The author is under obligation to many people for their con-
tributions to the cccletion of this research problem. He gratefully
aa8 owLedges the assistance, guidane, e and advice of Dr. Paul Tarrant,
chairman of the author's Supervisory Ccamitteo and research adviser,
the members of the Supervisory Camitteo, the mnebcrs of tho faculty,
and the authors felloW students.
This research program was carried out under a fellowship financed

by the National Science Foundation whose support the author naO


The use of haloalkanes containing fluorine has become of increas-

ing importance in syntheses as a means of introducing fluorine into

an organic molecule. One field of study which has led to a convenient

method for preparing magr interesting and important fluorine-containing

compounds involves the peroxide induced addition of haloalkanes to

olefins. For example, the work of Ilyquist (31) and of Lovelace (32)

has demonstrated that such compounds as 1-chloro-l,2-dibromo-l,1,2-

trifluoroethane, CF2BrCFCIBr, and dibromodifluoromethane, CF2~r2, in

the presence of catalytic amounts of benzoyl pcracido react with such

olefins as ethene, fluoroethene, 1-fluoropopene, and 2-fluorobutene-2

to give one-to-one addition products. In mary instances, these products

nay be converted by chemical methods to important fluorine-containing

olefins and dienes.

These reactions have been of interest not only for their great

utility in syntheses but also for the information which they offer

concerning free radical processes. As facts accumulate, a number

of valuable conclusions undoubtedly will be drawn dealing with such

phenomena as bond energies, intermediate radical stability, ltbric

hindrance, and chain reactions. Thus a study of free radical

reactions nor se of many fluorine-containing halocarbons can pro-

vide auch valuable knowledge of a theoretical as well as practical


Iany applications of the postulates derived frcn c crinont

have been madc in predicting or e.mlaining the course of a particular

reaction. For maple, Haszeldine (17) has studied the mode of

addition of free radicals to olefinic cystcss where addition mray take

place in either of two ways. ThuS, when trifluoronothyl iodide, CF3I,

is added to CIF=CI2 either CF3CHFCH2I or CF3CHCfII may result

dcpcndin; on the direction of addition. Actually, only the latter

product is formed, an observation which Haszeldine attributes to the

greater stability of the intermediate radical *1CFCH2CF3 cnmnarcd ,rith

Furthcr, the inability of manyolefins with halocgn atoms sub-

stituted at the double bond to add CF2Br2 or CF2BrCFC1Br was explained

by Lilyquist (31) and by Lovelace (32) as the failure for step V in the

following sequence to occurs

c6HcouGli --- 2 C6H0O*
0 0 0

C6110 ------>- C0H5. 00 I2


C6115*+ CFz23r ------ C Br* *CFg2r


*CF23r + CH2=CC1CI3 -----------

C:2DrCHl200CC' + CF2Dr2 --" --

Tne inability of the intermcd iate radical sham in step V

above to extract a obroinc atom fra CE Br2 has been attributed to

a rcc.nancc soablilizaticn of the radical by the halogen atom on the

c=.; carbon. (32)

1:"n such considerations .will dictate whether a given reaction

..ill prcced as desired. Another circumstance urhich a y influence the

course of reaction is the presence of conpoting side reactions which

use up the initiation frcc radical as i iis forced or which occur in

preoercnce to the desired addition. For -~anple, it has bcon observed

(27) that compounds such as C.23rCCl23r and CH2BrCFClDr do not add to
olcrins in the presence of pcrc.--:idco. The molecule CIH BrC12Dl-

r iffcrs F-'r- thos 0 *:ich have been oshr.n to add to olcfins with ease

in that it contains h:-dro:cn atc.- but no fluorine ato m. Similarly,

alth-ouh CHi3 rCFClBZ contains one fluorine aton, it too ponsessos

hydr-zo-c atoms adjacent to the site of the expected free radical.

Cor.potin\; side reactions illustrated in oqaations VI through XI below

ra-y accoun-t for the failure to obtain simple one-to-one addition

products .

C B1 I --- >-l 2 Q JCOO. VI

c6115CO. --- ----- *C6115 C02 VI
C H5f + CIy C... ..- VnI

C653A r + *CC1SIi3 r
*CC 0C11 r ------ CI1lCC12 Br. 3Z
cI I cOO. + CI3rCC2r x--- z

5 05 C00l + *C0DrCC12r
06150c001 .CGlmrCCl2Br

C6H* + *B r CH----- %lPr XI

There was also seao speculation on the fate of ary froo radi-
cals produced from the haloalkanes in the absence of arcn olefin. For
thseo reasons, it seenod of interest to study the reactions of
several lalogon-containin- alkanes with perodmdes in amounts greater
than the usual catalytic quantities.
The thermal decomposition of bensoyl peroxide in a wide variety
of solvents has been reported by many invostiCators. (44) In
general, a comply; nmituro of products is obtained whose nature depends
an the solvent used. Hartman, Sellers, and Turnbull (15) docom-
posed bonsoyl peroxido in a number of solvents and accounted quanti-
tatively for the carboxyl groups. They found that in aromatic
solvents carbon dioxide as the principal by-product with smaller
amounts of the carboxP l groups appearing as acid ad ester. On the
other hand, in aliphatic solvents a larger proportion of acid was
formed vith very insignificant amounts of ester.

Other workers (1,7,13,18) have investigated the fate of the

phaiyl radicals in various solvents. The isolation of bipheWyl
derivatives from aromatic solvents has established many important free

radical phernlation reactions. This work has led to new concepts
in free radical aroaitic substitution with regard to the directive

nature of substituents on the aromatic nucleus of both the solvent

and the radical.

The only polyhaloalkane which has been used as the solvent for

the decomposition of arcyl peroxides is carbon tetrachloride. One

such instance was reported by Boesekon and Gelisson (3) who obtained

terephthalic acid by bydrolysing the reaction product of bensoyl

peroxide and carbon tetrachloride. These investigators very reason-

ably proposed para-trichloromotiylbenzoic acid as the precursor to
their observed dibasic acid.

Again, Kharasch and Dannley (28) reported the decomposition
of ( -naphthoyl peroxide in carbon tetrachloride and found 1,4-naphtha-
lene dicarboxylic acid among the products after btdrolysis of the

reaction mixture. They proposed 4-trichloromcthyl-l-naphthoic acid
as a reasonable precursor.
Because of the ease of preparation of CF2BrCFCIBr and its

demonstrated reactivity in free radical processes (12,31), a large

part of the present investigation vao devoted to establishing the

structures of the products formed during the decomposition of benzoyl

peroxide and this halofluoroalkane.

In addition to the free radical reactions of halogenated alkanes,

another area of interest in this work is that of transformations

involving ionic reagents and nechaniass. Simons (39) has stated that
"fe attcnpts to use the saturated fluorocarbon chlorides or bramides
as chemical intermediates have been reported. Rcplacenent reactions

by nucleophilic reagents have not been reported and are probably
very difficult to do at best."
Although he was referring piZncipally to ccmpounds of the gen-

eral fonrula CFI (CG Oj wheXre S is C1 or Br, a similar statement
can be applied to compounds such as CS BrCFC1Br. The halogon atoris
are not displaced under noral conditions by nucloophilic attack on
the carbon because of the large negative inductive effect of the
fluorine atoms.
Grignard reagents have not been prepared from these compounds

although magnesium reacts slo;wy with certain perfluoroallcyl halides

at low temperatures. Thus the only important class of reactions
involving CIF2rCFCIlr has been the peroxido induced addition to
olefins as established by LilyqtUit (31) and by Gilman (12).
Consequently, it was of sawo interest to observe that CF rCFClBr
and many other vicinal dihalidos of fluorocarbons reacted readily
with concentrated alcoholic caustic solutions to precipitate large

quantities of metal halides. Thus a second objective of this work
has been to identify the products formed, examine the scope, and
establish a reasonable nechanis for the reaction between anionic
reagents and vicinal dlmalidos of fluorocarbons.



In the course of some investigations on the reactions of
CF2BrCFClBr, it was observed that on treatinG this compound with a
concentrated solution of alcoholic potassium hydraxide, a copious
precipitate of an inorganic salt was formed. This salt was deter-
mined to be potassium brcnide and no trace of chloride ion was found
in the reaction mixture. Such an observation was intcreotins in
that nucleophilic displacments of halido ion fruo molecules con-
taining several fluorine atoms have not been reported (39). For
such a reaction. to occur, a heterolytic sciscion of a carbon-halogen
bond rxut take place at the approach of the basic reagent:

L ci 0ZII
a--- CFPBrCFClX + Br

This is an example of the typical bimolecular, nucleolphilic
substitution reaction, S.2 (23). The failure to observe this type


of transformation among the fluorocarbon halides may be attributed
to the difficulty with which a carbon-halogen bond is broken to give
a halide ion. This in turn is probably a consequence of the strong
negative inductive effect of the fluorine atams, as well as that of
the other halogen atoms on both the oC and /-carbons. In support
of this suppoition, IIne 09,20) has shown that halogen atoms in
both the oc- and a- positions decrease S,~ reactivity.
neovortholess, the appearance of approdxiately two moles of
bromido ion for each mole of CF2BrCFCIBr after a rapid and vigorous
reaction clearly indicated either that the above considerations nuat
be modified or that some mechanism vao operative other than nucleo-
philic attack on a carbon atom.
A search for the organic reaction products revealed substantial
quantities of CIIOCF2CHFCl and CF2*CFC1 in proportions which depended
on perimelontal conditions. Ioen CFPrCFClBr was added to nethanolic
potacoium hydraxide, the other CHOCF2CHF1 was the principal product.
However, when the reverse addition was employed, the olefin CFj2CFC1
was formed in more significant amounts. Identification of the ether
was established by its physical constants as well as by conversion
on hydrolysis to the known ester, methyl chlorofluoroacotate,
ClIFCCOOCI (42). The nature of the olefin was deduced frao its
boiling point and by conversion to the original dibramide, CF2BrCClBr,
by the addition of bromines
The appearance of CF2*CFC in the products was a valuable aid
in explaining what -may seem to be at first glance an anomalous
Williamson type of reaction. The addition of alcohols to fluoro-

olofins in a basic medium has been well knoun for s m tine (14)
and the reaction has been c~ploycd by Warrant and Braon (41) as well
as by others to yield a variety of fluoroethers. A reasonable
nechanis has been suggested (21) maldng use of the electrophilic
character of the double bond in Ifuoroolefins:

ROII + KOff i1120+ +* RO- XI2I

ROCFCFCl 4 RO ---- X

'ITese facts would indicate, then, that the ether isolated in
the present study vas probably formed through an olefin precursor
so that the problem las resolved itself into accounting for the
formation of this unsaturated natorial.
Subsequent to these initial observations, a review of the lit-
erature revealed that this fom of reaction as not completely without
precedent, :ccDo and olt (33) reacted CF2ClCC13 with sodium phcnx-
ide and obtained a 34 yield of C6115OCF2'CIG.* Hoanovr, their expla-
nation wan fra the standpoint of an "anraalouo Williamson reaction"
in tich the normal product underwent a oubsoquent reduction:

CFC100CC1 + I1aOcI0061 ---- C0C110fC aGC] XIVI

C611FrFCC 13 + 2 1iaOCr - X7II
.C4IOCFaCIICk + CI5O11c0OI4,0a + *IaC1

None of the required oxidation products wero demonstrated.
Corley an coworkers (6) studied the Williamson reaction of

sodium methoxide with CF2CC1FC12 with the hope of obtaining

C1SOCF2CFC2. Instead, they found the "abnormal" product
CHO3CF2CIIFC and made the reasonable suggestion that a prior elim-

ination of a molecule of chlorine was involved to give an olefin which

then underwent an addition reaction with the solvent alcohol.
The difficulty of displacing a halide ion in the normal
1illiammson manner from a carbon bearing fluorine atoms has been

emphasized by Tarrant and Brown (41) in the case of certain lhdrogen-
containing compounds. These workers suggested that the formation of
C615OCF2CH2Cl from CiH5ON& and CF2010R2C1 involved a prior debydro-
halocenation to give CF21CU1Cl followed by addition of phenol to the

CF2ClC~ 1 H alj c l CF 2 C 1 Kl 20 XVIII



This course of reaction is in contrast to a previous explanation hiich
proposed a Uilliamson type of displacement of the chlorine on the
-CF2G1 group as the result of some type of "activation" by the -CH2C1
end of the molecule.
An initial item of evidence in establishing the course of reaction

was the observation that CF2rCFCIBr reacted with sodium dictfyl malo-
nate in a methanol solution to give CI0OCF2CHFCl, CF2CFCl, and a
crystalline solid as the products. A similar solid with identical


properties and infrared spectrum was isolated fro the reaction of

CPF2D and sodium diethyl nalonate and ism identified as tetraethyl

o 0

Cj O0-C C-OCIiH5

0 0

Pcrtinont to this observation are a number of studies on the
reactions of the tetrahialoamthanea. Dolaa and Groves (4) hoated
carbon tetrabr'oide with alcoholic sodium hydrxide and obtained
odiu.m biroide and sodium carbonate. On heating carbon tetrabirau do
with othyl alcohol they isolated brcioform and detected a strong
aldehydo odor. Tids reaction may be represented by the following

CBr4 + CjI5OH ----C23r +* Ei t C+Ofi XXZI

IIef (34) found that carbon tetrachloride and carbon tetrabramido
on treatment nith alcoholic potassium hydro~dde gave the same products
as chloroform and braoform, respectively. 1tth alcoholic sodium
othomide the products wore carbon monoxide and ethyl orthofornate.
IHnover, when a solution of potassium hydroxide in methanol vas used,
potassium formate was detected rather than the ortho ester.

Ingold and Po3ell (22) reported an attempt to prepare:

0 0

by the action of carbon tetrachloride on sodium iiothyl nalonate.
However, every attmept resulted in the formation of:

0 0
11 II
( o cn-(coc -

a capound ihich Conrad and Guthoeit (5) had previously prepared
fror chlorofom and sodium diethyl nalonate. These results implied
that carbon tetraclloride, in a basic medium, was reacting as if it
were caloroform although Ingold and Prowel were uncertain of the
source of the hydrogen atom in their product. One suZcestion by
those investigators was that it was provided by the oxidation of
sora other molecule such as the solvent alcohol.
To demonstrate further that the tetrahalamltthanos were incapable
of yielding normal condensation products with sodium diethyl malo-
nate, Igold and Powell found that the reaction of the latter sub-
stance with carbon tetrabramide gave tetraethyl othane-tetracarboxy-
late (22).
Ingold (24) has elaborated on the mechanian of all these
reactions and on the similarity of the behavior between the totra-
halamothanes and the halofonrm by postulating an intermediate
trihalaml thide ion C" as a result of nucleophilic attack by the
strongly basic anion. In the case of the haloforms this ion is
produced by an initial removal of a proton by the basic reagent while

for the tetrahalamethanes it is probable that the extraction of a
positive halogen ion is the initial step.
The formation of fluoroform by the action of alcoholic potae-
sima hydroodd on trifluoromethyl iodide, CF I, is almply an
extension of this mechanism in which proton uptake follows the for-
nation of the trifl2uormthyl anion, CF37 (2).
In the present case, the isolation of tctraethyl ethano-tetra-
carboqlato frac the reaction of sodium diothyl malonate on CF2Br2
and CF2BrCWCIBr leaves no doubt that a positive halogen as extracted
by the diethyl malonate anion. The diethyl br~amoalonate thus pro-
duced coupled with another molecule of sodium diothyl malonate to
give the observed products:

(C25C Na + C 1ClBr2a----
(GC^ )CHBr *+ [c Cir]+ *1
(CH40 )2CIHBr (C25OC)2OCH- Na--+ III
0 0
0 0
C25 CO0C25

0 0O
o 0

A beta elimination of halide ion froa the fluorocarbon anion cea-
plates the formation of olefin:

1r(CF2CFCl)7 C- CF Cl Bt Xr17

Brackets are employed in the equation above for it has not been
established at this point which branine is raoved as the positive
br aonian ion nor has the intimate meemani of the beta halide
elimination been demonstrated.
On the basis of this information, the following steps will
account for the observed reaction of CF BrCFC1Br with alcoholic
potassium hydroxide:

C0OII + KOH CI0 + 1+* I1O Xx
CiU0 CF2DrCFClDr ----. ---m XZ!I

Dr(CF2CFCl) ----- CFC1 Br- XMVII
CiuOCF2CFCr +* C f3I ----F


Other basic reagents react similarly to yield the products
shown in Table 9 WiLth diethylamino, a LIxture of amides is formed
by the hydrolysis of the intermediate amine. On the basis of boiling
point, refractive indc, .and the value of molar refraction, the chief
constituent of this mixture is probably the o( ot -difluoroacotamido.
'ihe infrared spectrum of this product show two absorption bands in











CI 00co1a

(c2uI5) n

(1,A~i I

Ct OCT^H5?a





Io Reaction

(cn5) jcin'2


1o Reaction



, ,,,, ,, ,, ,,...J i, ,
, i i i i i

~--.~-- -.--~- ----1.-- I~----"1----~~.- ~--~--~ -------

"I ~~""' ~~""~~

the carbonyl region, a very strong band at 5.95 microns which is also
given by an authentic sample of the chlorofluoroacetamide and a medium
strong band at 5.56 microns uhich is probably due to the difluoro-

acetjmido. Fried (11) has dwaonstrated that at elevated temperatures,
the difluoroamido is formed in greater proportions than the chloro-
fluoroamide by the addition of mines to chlorotrifluoroethene followed

by hydrolysis.
The use of codiun iodide in acetone yielded no olefin although
iodine was formed in substantial quantities. This reaction was not
investigated further. Likewise, sodium acetate in ethanol was unreactive
toward CF2DrCFC1Br which indicates that a strongly nucleophilic reagent
is required to bring about reaction.
1:ethyl ragnosium broaide gave a 665 yield of chlorotrifluoroethene
when reacted with CFBrCFC12i 3 lrile sodium diethyl malonate gave this
same olefin as a major product. The appearance of the ether
C~IOCFCHFiC1 in the products of the reaction between sodium diethyl
nalonato and CF23rCFClBr in a methanol solution confirms the analogy
between this reaction and that employing alcoholic potassium hydroxide.
arnor (46) found no evidence of addition of the diethyl malonate anion
to chlorotrifluoroethene, an observation which is valid in this instance
A number of other vicinal dihalides of fluorocarbons were treated
with alcoholic potassium hydroxide and the formation of the products
shown in Table 2 indicates that the reaction is a general one. The
yields of ethers is not high in most instances although the principal
difficulty may lie in handling losses. Nevertheless, a limitation of



Reactant Product Y Yield



CF CCf~ Cl
2 2










No 0r n
No Reaction

(a) Ethanolic potassiut bydrt ide used

this procedure is that a more satisfactory method of preparing the other
can usually be found in the base catalyzed addition of the appropriate
alcohol to the olefin. However, the hazardous nature of tetrafluoro-
ethcno thich has curtailed its shiWpent by freight nmy make the dibro-
nide an attractive route to the ethers, PfDCF2CH e
Dibroaodifluormnethane reacted with alcoholic potassium hydroaide
quite vigorously although on only one occasion was any organic sub-
stance isolated other than starting material. This single instance
gave a liquid which boiled at -o but wtose identity was not established.
It is possible that this material iras the impure reactant, CF2Dr2.
Hasscldino (16) has reported on the alkaline hydrolysis of a number of
the nany possible tetrahalanethanes and haloforms. As he obtained
carbonates froa the totrahalo derivatives and formates fro the haloforms,
there is no reason to believe that the present tetrahalo canpound,
CFS2, should behave in a different nanner.
It was oriinally thought tthatthe compound CFP2BrCIaCII2 would
provide an interesting system for a study of its reaction with alcohol-
ic caustic solution. If a positive halogen ion were cstracted, there
could be no beta elimination as a resonant system would be established.

MP- CF2rCCH- H------

rO3r + IP20^ CH=0-2 CF2uHcH2

Proton uptake would possibly follow to give either the 1,1-difluoro-
propone or the 3,3-difluorropopn depending on the relative stabilities
of these two olefins.

However, it tas found that no low-boiling material ais formed
thich would be anticipated had a propane been the product. Neither vas
there any indication that an ether of the form CH3OCF2CH2H3 as pro-
duced. he inorganic salt which was foamed was determined to contain
O.+48 mole of potacoiun br aido -while the remainder could be only

potasoium fluoride, 0.281 mole. Those proportions suggest that the
primary reaction was hydrolysis to give acrylic acid or its salt and
possible subsequent polymerization. Tis would perhaps account for
the transparent gelatinous material which was observed during the reaction.
The mechanism which has been given for the reaction of alcoholic
potassium hydrooido and such vicinal dihalides as CF2BrCFC1r3r has
depicted an intermediate carbanion. Although the actual science or
lifetime of this carbanion has not yet been shown, it sooncd of interest
at this point to attempt a utilization of this species in synthesis.
For sample, if the formation of the carbanion were accaqplished in
the presence of an electropositive center as provided by a carbonyl
group it may be possible to effect transformations reminiscent of the
Grignard reaction:

+ CFBrFClBr --- s Dr(CF2CFC1) + ROBr XXXI
Br(CF2CFCl) + RX-C-R ___
Br(CF2Cl)-CRR'o- -proton donar


Table 3 outlines the reactions attanpted with this end in view
and the products thus obtained. In many instances it 3 as found that
reactions between the strongly basic nediam and the carbonyl compound
accounted for the only observed products. For omple, iAwhn dibroao-
difluoranthane and benzaldehydo vre used, an almost theoretical
yield of bonslc acid was obtained by the Canniszaro reaction. In
general for thin series of reactions, there Vas either no observable
transformation or the isolation of only chlorotrifluoroothene, ether,
QCFJICIIFC1, and starting material.
Since those experiments shmowd at best only that the preferred
course of reaction its elimination with formation of the olefin rather
than addition of the carbanion to an electropositive center, sane
additional data on the mechanic~ scaneed desirable.
It has previously been postulated t tt the formation of tetra-
othyl etI:arlo-tctracarbcylxate from Cr2TrCFC1Br and sodium diethyl
malonate involves the removal of a positive bromine ion fran the poly-
halocthano and that the same step undoubtedly occurs vhen the basic
reagent is the all=dde ion. The constitutional factors whichh would
aid such a cleavage uould be strong electron ritihdreial frac the carbon-
bra~ine bond:

I '
B r






Ialofluarocarbon Carbonyl Chief % Yiold
Compound Product








C1MY00C 2U5




ho product







(a) Trcated vith po wdered einc at 00


This situation is favorably provided in the fturocarbon campounds
because of the high olectronegativity of the fluorine atom. This factor
is also probably chiefly responsible for the difficulty of nucleophilic
displaccnent of halide ion fro the fluorocarbon halides. loIo ier, it
vould appear reasonable, on the other hand, that the stronger the
electron idthdraual fran the carbon-bramino bond, the noro facile would
be the extraction of the positive brcoinium ion. As a consequence, one
would xpcct the braoino on the carbon bearing the greater number of
fluorine atoms to be initially attacked. Beta elimination of a halide

ion would then coapleto the reaction. ThiS sequence is characteristic
of the typical bifolecular elimination reaction (2) as caplified
by the dchydrohalognation of the al~yl halides by alcoholic caustic


Ingold (25) has noted that a true Eg elination should exhibit
second order kinetics, a concerted or simultaneous dlinnation of HX ,
and a dependency on a strong base for the initial rcaoval of the proton.
An alternative nochanism would involve formation of the carbanion
follotrd by unimolccular loss of halide on, a process which Ingold
has labelled Tlc3 (26) or unimolecular elimination front the conjucate
boo :

E0 lC l2-n2 R-c ;. -CF2CP.2X M"O
fast as
-CP2Cr- --- be ---di CIn-CR -+ is f wI

A similar caquence which Wy be designated as I1cD is the following:

WO- + H-CqgC- -x +OH -CiCIcX XXVmz

-CP2CREX ---- CeCR X- XXX mIi

Docauso of the finite existence of the intermediate carbanion

and because of the reversibility of the first stage, a reaction carried

out in deutcratod solvent amst show sono deuterium in the unreacted

alkyl halide after partial conversion if the EIcB mechanism is operative.
IHowver, Skell and Hauser (40) showed that for a typical elimination

reaction, the production of styrene from 2-phenylethylbroaide and

othoxide ion, no deuterium was detected in the bramide after partial


In the present study, several alternative events may occur follow-

ing removal of the brononiun ion aitch would account for the observed

products. The first is a single stage, concerted mechanisms in ihich

beta olimnination of a halido ion is simultaneous with removal of the
positive braoine ion. Ingold dosignatos this type of olcfin forma-
tion as E2 (25):

&+ .--- 6-
RO ** .*Br. tCF FCFC*.Br ) --P XXXIX
ROBr + CF2CFC1 + Br-"

The second alternative involves an intermediate carbanion with

a finite existence uhich may then undergo beta elimination. This

EIcD mechanism has two forms both of mhich require the sane kinetic


RD +* CFBrCFC1Br 1 ROBr + --CFpFC1Br Inl

-CFfFCB- r a-- m:0- C r.CF2-CFC1 + Br

-CF2FClDr -- CF2+CFCl + Br"

The possibility of addition products Tdth carbonyl compounds -ms based
on the existence of a relatively long lived intermediate carbanion as
provided by the above mechanism.
A third possible mechanism is abstraction of a proton by the
intermediate carbanion followed by dohydrohalogenation in the presence
of excess base:

no- + CF2BrCFCIBr --- ROr -CF2CFCLBr X LI

*-CF2CFC1r proton donor l :N- XI

An xaqple of a carbanion with sufficiently long lifetime to
abstract a proton in such a manner is provided by the work of HIazeldine
(16) mho obtained heptafluoropropane' CFCPCF2H, in 70% yield on treat-
ing hoptafluoroiodopropane with alcoholic potassium hydroxide.
Presumably the mechanism involves formation of the carbanion:

CF3CF2CFT- Rs+ ---o CF3CF2CF2I r RO" X~ I
Another related example where the presence of proton donors influ-

ences the fate of the carbanion is the addition of alcohol to
fluoroolefins to give ethers:


ROCF2CCl- *+ Rso1- fvastI


Hormover, in the absence of proton donors as when chlorotrifluoroethene
is treated with sodiin methomde in anhydrous other (9), an unsat-
urated other is obtained:


(CHOCFPFC1l) ----- CI3CF*CFC1 + F' L

Furthermore, Tarrant and Warner (43) noted that the addition of
Grignard reagents to a number of halofluoroolofins Gave nmo unsaturated
ccmpounds but noIne of the crpectod addition product:

RiXg + CF2X, 2 (REF2CXsIgx) LI
.-- . RC- CXt 2 + 1ijF

The kinotics of the reactions between several haloethanes and a
basic reagent were studied and the results are sum arieed in Table 4.
It was found that a basic medium consisting of sodium hydroxide dis-
solved in a water-di sano (1:2) solvent gave a more measurable rate of
reaction than could be obtained in an alcoholic system and the data are
therefore presented for these conditions.

The experimental data gave the best straight line fit when expressed

in a second order kinetic form, first order with respect to both hydrcryl

ion and dihalide. Although there is some curvature resulting from those

points taken early in the reaction, a reasonably good linear plot is
obtained* (See Figure 1., p. 49 and Figure 2., p. 51)
It has already been indicated that a necessary feature of the
reaction between these dibromides and anionic reagents is the removal
of a positive halogen ion resulting from an electron withdrawing prop-
erty of the fluorine atom. From the order ,of magnitude of the rate
constants in Table 4, CF2BrCCl2Br > CP2BrCFClBr >) CF23rCF9Br, it may
be concluded that the ease with which beta elimination ,of halide ion
occur s i also a rate determining factor.
These data are of limited value from the standpoint of providing
an un=abiguous reaction mechanism. The four alternative .mechanisms
which have been given in equations XXXIX, XL XLI, and ILII-XIU all
exhibit second order kinetics. Furthermore, three of the four will
result in the same order of reactivity within the series studied.
however, the second form of the E0IB mechanism illustrated in equation
YLI depends only on the ease of removal of the bromonium ion and
should consequently give an order of reactivity opposite to that actu-

ally observed. This scheme may therefore be discarded as a mechanism
for the reaction.
In the case of dehalogenation using the Grignard reagent where pro-
ton donors are absent, equations XLII-XLIV will be non-operative and
the alternative mechanism s my be narrowed to the E2, equation XIX and

the first form of the ElcB scheme, equation XL. A cyclic intermediate

may also be involved:


Br Br


In order to decide between a concerted mechanism and one requir-

ing an intermediate carbanion, a specialized system would be necessary.

For instance, assuming that the necessary pure isomers could be ob-

tained as starting materials and that the products could be identified

accurately, meso-CFClBrCFClBr should give pure trans-CFC1'CFCl if a

concerted mechanism is operative and a mixture of the cis- and trans-

olefins if an intermediate carbanion is involved.


T m 30.20 A 0.01
Dibrcaide iraidoe [o]0 k2 (moles / / sec.)

CF2rcCM gr 0.039 M 0.1125 11 2,2 x 10-2

CF2BrCFClBr 0.0327 M 0.1175 X 7.8 x 10~

CF2CF2B 009 0.09 M 0.0944 4 Unroactivo

iam in vater-dioxane (1:2) solvent system


All teaporatures recorded in this dissertation are uncorrected

and are given in degrees Centigrade. Distillations were carried

out for the root part using a 20 cnm coumnn packed with protruded
packing. Preosures were measured with a Zfinerli gauge which could

be road to + 0.2 Zm. Refractive indices were determined with an

Abbe rcfractcQcetor at the temperatures indicated* Densities were

easurcd in pyoiaeters calibrated against water at 25 and corrected

to water at 4o* IHolar refractions were calculated by the Lorens-
Lorents equation.

Infrared spectra were obtained on a Perin--Elmer model 21
doublo-bcan Infrared Spectrophotnaeter*

Carbon4-hydrogen analyses were performed by the Clark Macro-
analytical Laboratory, Urbana, Illinois and by Drs. G. WOeiler and

F. B. Strauss of Cxford, England.


A. Reaction of CFPBrCG1rPwith Basic Reagents

1. lith Alcoholic Potassium Hydroxide
A solution of 100 g. of 85, potassium hydroxide (1.5 moles) in
200 ml. of methanol was added dropdise to 138 g. of 1,l,2-trifluaro-
2-chtlro-12-dibramoethane, CF2BrCFC1Brs (0.5 mole) contained in a
500 ml. thrco-necked flask equipped dwth reflux condenser, addition
funnel, and a mechanical stirrer. The reflux condenser was backed
by a Dry ice-acetone cold trap.
After the alcoholic potassium hydroxide had been added the solu-
tion ims stirred for one hour. The mixture was allowed to cool and
the vhite precipitated solid was filtered and dried, One hundred and
sixteen grams of an inorganic salt was obtained thich gave a qualita-
tive test for broaide ion but not for chloride and was determined to
be 99.51 potassium bromide by titration (Volhard).
The filtrate was distilled through a 30 cm. column -ith protruded
packing to give the following:
Fraction b.p.0 weight 5 d25

1 27-32 2.0 g.
2 55-60 40.5 1.3325 1.21

3 62-67 112*5 g. 1.3285 0.781
4 49/88 ram 27.1 g. 1.3338 0.965

Fraction 2 was washed twice with cold water, dried over calcium
chloride, and redistilcd to give 21 g. of a pleasant smolling liquid
with the following physical properties: b.p., 70-71; n1 1.3321;
d4 1.3618; IM found, 22.4. Reported for CH0CCF2CFC1t b.p., 70.60;

n 1.3340; Dd, 1.3636; % called 22.5.
Thi material was further proved to be the other CI30CF2CHFC
by hydrolysis with concentrated sulfuric acid t t the known ester, methyl
chlorofluoroacetate (42). This hydrolysis product was found to have
a boiling point of 214.5-316o; nf 1*3920. Reported for
CHFCCOOC3: b.p., 1160; 5 ,1.3903. Furthon or e, infrared spectra
of a center cut from fraction 2 and of an authentic sample of
CHI3OCFCHFC1 were identical.
In the cold trap was found 21 g. of a colorless liquid which had
a boiling range of -2Pto -220 and added bromine readily to give a
liquid vith a boiling point of 920 at atmospheric pressure. Reported
for CF2-CFC1: b.p., -28o. Reported for CFjDrCFCl8r: b.p., 93.50*

2. Tith Ethyl Ia nesium Bramide
The Grignard reagent was prepared by the reaction of 55 g# of
ethyl bromide (0.5 role) with 15 c. of reagent grade magnesium chips
in 300 ml.* of anhydrous ethyl ether. To the Grignard solution, 133 g.
of CF2BrCFClBr (0.5 mole) was added dropwise and a lowi boiling liquid
appeared in the Dry ice-acetone cold trap. Fractionation of this
liquid gave 38.2 g. of a liquid with a boiling point of -26.50 and
Ahich added bronine readily. Reported for CF2GCFC1: b.p., -2e o

3. Idth Sodium Diethyl Ilalonato
To 200 ml. of absolute ethanol was added 11.5 g. of freshly cut
sodinu at such a rate as to maintain reflux. Diothyl malonate

(78 g, 0.417 mole), redistilled at reduced pressure, was added to
the sodiiun nethoxido solution with stirring.
One hundred and thirty eight grams of -CF2BrCFC13r (0.5 mole)
was added dropwise to the refludng alcoholic solution of soditu
diethyl nalonate. A brown solid precipitated and heating was con-
tinued for two hours. At the end of this time, the solution was
cooled and filtered to yield l48 g. of solid and 210 ml. of broun
liquid. In the cold trap was found 12.2 go of a low boiling liquid
which added brranne readily to yield 20 g. of material boiling at
92P, 1.4258. Reported for CFDrzCFC13Brt b.p., 93.50; 5 1.4Ml .
iho filtrate was distilled to yield an alcoholic aseotrope and a
dark colored still residue. By washing the alcoholic portion in cold
water, an insoluble organic liquid was obtained which was dried over
calcium chloride and fractionated to give the following:
Fraction b.p.o wight n20 d

1 70-75 8.0 g. 1.340 1.3650
2 75-90 1.5 8.
3 90-94 54,0 g. 1.4245

On the basis of the physical constants, fraction I appears to
be the ether, Cs0OCi CHFC1, rwhle fraction 3 is unracted CFpBrCFCltr.
The dark residue was poured onto crushed ice and a quantity of

crystals formed. The crystals wre filtered and lashed several tin e
with cold water. 1o qualitative tests for halogen src obtained upon
fusing a recystallized sample with sodium. An infrared spectrum
exhibited absorption bands at 6.70 microns and at 8.50 microns, char-
acteristic of an ester. The solid had the following properties:
n*p., 73-740; saponification equivalent, 77.3.
Anal. Calcd. for CI 20: C, 52.8; H, 6.92. Found: 52.27;
H, 6.37.
For ttraetathyl ethane-tetracarbxylates m.p., 760; saponification
equivalent, 79.5.

4. ti th Sodium Iodido in Acetone
A rniture of 411.3 g. of CF2irCFCIBr (0.15 mole), 22.5 eg of
sodium iodide (0.15 mole), and 100 ml. of acetone was placed in a
lack and re ~fe ed, th stirring for 48 hours. The solution
beceae dine colored but no cei boiling atrial was collected in the
cold trap* After several days, a quantity of crystals of iodine was
noted but the rixture tas not studied further.

5. tith Sodiaun Acetate
A mixture of 61.5 g. of sodim acetato (0.75 molo), 138 E. of
CFDprCF1Dr (0.5 mole), and 120 al. of ethanol was reflucd for 4
hours dwth stirring. At the end of this time, the mixture was poured
into ice water and the insoluble organic layer was removed, washed
with cold water, and dried over calcium chloride. COne hundred and
five gra s of liquid was thus obtained whichh was distilled through a
30. cm. coltan with protruded packing to give a single fraction,

b.p., 93-940; n25 1.4268. Reported for CF2BrCFC1Br: b.p., 93.50;
25, 1.4278.

6. With DietyWlamine
One hundred and thirty eight grams of CF2rCFC1Br (0.5 mole)
was added dropwise to 80 g. of diethylamine (1.1 moles) at reflux
temperature in a 200 ml. three-necked flask. The mixture darkened in
color and solidified partially on cooling. A small portion of the
solid was removed and washed with ether mntil all color was absent from
the crystals. Recrystallization from ethanol gave a white crystalline
solid which formed a precipitate with silver nitrate solution and had
a melting point of 212 Reported for dietlhlamine hydrobronide:
m.p., 233.50
The remainder of the reaction mixture was steam distilled. The
organic layer was separated, washed, and dried over calcium chloride.
Distillation gave 50.0 g. of unreacted CF2BrCFClr and 5.5 g. of a
pleasant smelling liquid with the following properties: b.p.,

50-55 / 5 man; 5 1.443; d 1.121. Upon fusing with sodium, a
qualitative test for both nitrogen and halogen was obtained. An
infrared spectrum showed a strong band at 5.95 microns and a medium
strong band at 5.56 microns.
For (CI 5)2NCOCHEFC: MIR calcd.,38.50. Found, 37.44
For (C2H5)jCOCHF2: 1%M calcd., 33.43. Found, 33.67
U,Ti-diethl-l-lfluoro-l-chloroacetamide was prepared by reacting

36 g. of diethylamine (0.5 mole), 29 g. of chlorotrifluoroethene
(0.25 mole), 5 g. of Borax, and 50 ml. of ether in a small autoclave
for 8 hours at 30. Distillation gave a liquid with the follo-inE prop-

erties: b.p., 70/ 9 mr; n4; 1.4547; d5 1.18.
Called. for (Cj5)ICOCHFCl1: t1% 38.50. Found: 1 M, 30.3.

An inrared apectrm showed a single band in the carbonyl region at
5.96 nicrons. he physical properties and infrared spectrum of this

authentic saz;ae of the chlorofluoroacotamide indicate that the material

obtained by hydrco3ysing the reaction product of diethylamino and
CF2DrCFClJ3r is probably a mixture of chlorofluoroacetamide and difluoro-

7. lath Triothylzaine
To 10L g. of triethylamwine (1.0 nole) at reflux temperature s
added 69 g. of CF PrCIDCr (0.25 mole). At the end of oix hours, the

rdzture vas allcIrod to cool and the small amount of solid matter which
had formed was filtered. 'Thi solid ms rocrystallized by diesolving
it in mcthanol and reprecipitating by the addition of ether. Thie Cried
natcrial gave a nelting point of 241.5 Eoported for triethylniino
hydrobruar2 c: sinters at 2+20h nclts at 20o
hic reaction riture .as poured into cold water and tU lo=er
layer was separated and iaslhed coveral ties rit.Lh dilute Ihyroclilric
acid to raMov residual amino. Jie resiltin; organic liquid fms
dried and distilled to give only unreacted CF?32CFC1Dr.

D3. reaction of Other Vicinal Dihalidos with Alcoholic Potassium

1 1,1,2--Triflumo-l,2dichloro-2-iodocthano, CF2C1CFCUI
To a solution of 66 g. of 85G potassimu hydrm:de (1.0 mole) in
200 ri. of nothanol was added 339.5 g. of CF2ClCFCI (0.5 nole)
Sdropidso with stirring. The white solid which precipitated was
removed by filtration at the end of the reaction and the filtrate ams
distilled until the alooholic aseotropo no longer gave an insoluble
liquid portion upon dilution with water.
The distillate was poured into cold water and the lower layer
was rraovod, twahod several times idth cold water, dried over calcium
chloride, and rediatillod. In addition to a small quantity of fore-
run, there was obtained 3JE4 g. of liquid with the follourng properties:
b*p., 700; j 1.3369; 5 1,3625.
Calcd. for C1 0OCF2CIFC1: I~D, 22.5. Found: 22.4.
Th reported constants for CII30CF2CIICl are: b.p., 70.60; n25 1.3320;
d2, 1.3366.

2 1,2-Dichloro-tetrafluoroothane, CF2C1CF2CI
CFrC1CF2CI was treated with alcoholic potassiua hydrodde under
various conditions with no successful isolation of the desired other,
CH3C02CI=F2, In glassaro there was no evidence of reaction and the
CF2C1CFCI1 could be recovered unchanged. In an autoclave at tempera-
turea up to 1500, scn inorganic salt was formed uhich gavo a quali-
tativo halide ion test twth silver nitrate solution but no ether was
isolated in any instance.

3. 1,2-Dibrmano-tctrat.ueroothane, CFjPrCB?'21r
One hndrd and thirty g s of CFP FrP (045 nolo) was added
dropdise to a rofll ng solution of 65 g. of potasium hydro:ide
(.0 mole) and 175 nL. of methanol. A xiito salt precipitated and
was rmaoved by filtration after the reaction was coaploted.
Distillation of the filtrate gave 37 g. of a colorless liquid
boilirg at 3-40; 1.32, Reported for CII00CF2SIIF2 b.p., 36.5;
ra0, 1.3.

14. l,2-Mibrc2o-2,2-dichloro-1,l-ditluoroethane, CF2BrCC010 r
C2PrCC012r nas prepared in a 97 yield by passing gaseous

CPF2*CC2 through brarine under illuc-nation by an ultraviolet slmp.
'li product was purified by distillation, b.p. 385.P.
One handrod Mgras of CF2BrCCl2Ur (0.342 nble) tos dissolved in
a small voluno of retihnol and placed in a threo-necked flask. Sixty-
six 9~=rs of C5% potassiun hdraccde (1.0 rolo) was dissolved in
200 ml, of ethanol and added dropuise to the contents of the flask.
A considerable amount of heat was evolved and a white precipitate
formed rtich uas raoved by filtration at the cnd of the reaction.
Sixty-fivo grams of potassium braido twa thus obtained after dryin
in an oven.
The filtrato was distilled and the alcoholic portion was poured
into cold wator. The organic layer was separated and wsahod with
Eator to give ZS g. of crude product uhich was dried over calcium
chloride and redistillod. FAghteen grams of a liquid wms collected
with a boiling point of 105 ad 5 1.394.

Reported constant for C150CF2CHCH : b.p., 1050; 1. 361.3

5. 2-Taritifuoranothyl-, 5-dicasioro-4,5, 5-trl uoropenteno.2,

ibs olefin as prepared by the dohydroiodination of
CF2jc1:ccIIcCI(C)Gu wi-th alcoholic potassium hydroade.
To 70 g. of the above olefin (0.27 nole) in a threo-necked flask
wtas added 53 g. of potassium hydraddo (0.81 mole) dissolved in
150 ml. of nothanol. The mixture was heated to reflux and a amal
quantity of potassium chloride precipitated. On distillation, no
organic materials vare isolated other than nothanol and unreacted
starting material.

6. l,l,3,3,3-Pentafluoro-1,2-dihbrao-2-ch3loropropano CF3CClDrCP2r
In a 200 ml. round-bottao flak equipped with a rcflw: condenser,
nochanical stirrcr, and an additional funnl was placed 45 g. of
potaosium hydrmdde (0.7 mole) dissolved in 150 ml. of ethanol. The
CF3CClrGcFbDr (70 g., 0.21 aolo) ian added dropuaeo to the woll stirred
caustic solution. The reaction vas highly osothemic and cooling tao
provided by an ico-bath. A lihito salt eas procipltated t:iodiatoly
and, after addition was complotod, the solution was rofluxed for one
The solid ias filtered, the filtrate poured into cold water, and
the insoluble layer separated. hs organic liquid wma washed thoroughly
tdth cold zater, dried over calcium chlorido, and distilled through a
20 acm column vith protruded packing. the folo ~ ing fractions 1ore

obtained rC a 30



g. charge:





7.5 1.3C77

5.7 g. 1.3765

5.5 g-

Fraction 4:

Physical cons

Anal. Calcd. for CF3CIl1COOCIIy: Cl, 17.1; I3RD, 31.81
Found: Cl, 16.3; I'%, 32.17

rtants reported for CFHCIlCOOi:C btp.,1280 n,1.3635
3c~I~ 2 ~L b.p.,12S0: 1~,*3~

C.- fLoactions of Other Ccaplounda tdth Basic Roagents
1. Iibrancodiluorcaiethane with Alcoholic Potassium fyodrca de
A solution of 100 g. of potassium hydradde in 200 ml. of
nothanol vas heated to reflu= and 105 g. of dibrmodifluoroethano

(0.5 nolo) was added dropdsoe. A large amount of salt mas forncd and
a quantity of liquid mas recovered faa the cold trap. Distillation
of this liquid yielded 16 go of material boiling at -2 An infrared
spectrum of this liquid was of little help in identifying the sub-.
stance and its constitution ;is not Imonm.

2. Dibrcodifluoromthae iedth Sodiun Diothyl IIslanate

Preshly cut codiun (11.5 g., 0.5 moalo) was added to 250 nl. of
absolute othanol at such a rate as to maintain reflu. Diothyl nalonate

(00 g., 0.5 nole), rodistIlled at reduced plrssure, was added to tho


- ~...~ -_rtC ._.:


sodium ethoide solution with stirring. To this solution at ref1m:

tcnpcrature was added 105 g. of CTF23r (0.5 mole) over a period of
one hour. A brown solid precipitated and roflux Vas continued for one

At the end of this tno, the solid was filtered and the filtrate
.as distilled. The alcoholic distillate yielded no organic material
on pouring into cold water swile the still residue gave a crystalline
solid on addition of icowater. Mo solid vwa separated, rccrystallizod
frc2 Ihexane, and dried, n.p. 73-74O Th e roltinC point of thid solid
was not depressed on nixing it with the material derived fr=a the
reaction of sodium diothyl nalonate and CF.-rGFClDr. In addition, the
infrared opoctra of the solids obtained from these two experiments wror

Anal. Called. for CU200: C,52.0; II, 6.92. Found: C, 52.27;

I, 6.37.

3. 3,3-Difluoro-3-brcm~ propene with Alcoholic Potassium Illdrcmido
A solution of 44 g. of C05 potassiun Ihydroide (0.67 mole) in
150 ml. of ethanol was heated to reflin: in a 200 ml. flask and 50 g,
of CFrCylsC 2 (0.33 nolo) vas added dropiise. A vhte solid as
procipitated and a transparent gelatinous substance formed on the surP
face of the reaction :'txturo.
At the end of the reaction, the content of the flask were dis-
tilled to give two fractions: fraction 1, bhp. 30-4 CF2'CI=ClI21,

12.2 g.; fraction 2, 65, methanol. The methanol distillato gave no
insoluble portion on diluting with cold water. The rosidue in the

lask wasi filtered to give 34 C. of solid wiich contained 5Z'
potanaoiu bra-ide (0./48 molo). The reminder can be only potao-
siun fluoride (0281 molo).

D. Reactions Attampting to fTap the Carbanions CF B- or CFClDrCF

I. Reaction of CFgBr2 vith Alcoholic Potassium Hydroxido in the
Presence of Bensaldehydeo

In a 500 ml. three-necked flask equipped with reflum condonsor,
mechanical stirrer, and an addition funnel, iwep placed 105 g. of CF2J3r
(0.5 molo) and 53 gS of bensldehyde (0.5 mole). A solution of 35 g,
of potassium hydroaide (0.5 mole) and 75 tl* of methanol was dropped
eloaly into the reflidng mixture of CFr.2 and bensaldehydo. After
eight hours, the flask was cooled d and the solid matter was filtered
frxo the contents. The solid was dissolved in water and acidified with
concentrated hydrochloric acid until no further precipitation occurred.
After drying in an oven, 28 g, of material, n.p., 122-230o (Denzoic
acid, n.p., 1220)
Tho filtrate yielded 30 g. of a liquid boiling at 24O (CF23r2
b.p., 22-24) and methanol which gave no insoluble organic layer on
washing with cold watcr.

2. Reaction of CF2r2 %Ath Alcoholic Potassium Ilydroxide in the
Presence of Acetone
Potassium hydroxide (20 g, 0*3 Cole) was dissolved in 75 mlt of
ethanol and added dropi~se to a refluxing mixture of 50 g, of CF2Dr2
(0.25 mole) and 75 mt. of acetone. After addition had been cacpleted

the r=tture was distilled directly from the reaction flask through a
20 cm column. A fraction boilin at 220 (12 g.) as collected in
addition to 130 ml. of liquid boiling frma 56--6f0 which yielded no
organic portion on washing with cold water.

3. Action of CF2DrCFClDr with Alcoholic Potassiun Hydroide
in the Presence of Ethyl Acetate

CF2BrCFC1Br (13S g., 0.5 mole) and ethyl acetate (44 g.0 0.5 molo)
there placed in a 500 mal three-necked flask and treated dropwise with
33 g. of potassium hydroxido (0.5 mole) dissolved in 90 ml, of nethanol.
After six hours of rofl dng, the reaction mixture was distilled to
yield an alcoholic distillate boiling over the rango of 55-750. his
liquid was poured into cold water and the denser organic layer was
separated and dried over calcium chloride. Distillation gave 84.5 g.
of a luid boiling at 92-93, 5 1.4S265. For CF;prCl1r, b.p. 93.5;

145 .278.

4. Pcaction of CF2rCFGCDr with Alcoholic Potassium Hydroxide in the
Presence of Ethyl Trifluoroacetato
CFDrCFCI3r (233 g., 0.5 mole) and ethyl trifluoroacetate
(100 g., 0.7 rnol) were placed in a 500 ml. three-necked flask and
treated droprise with 66 g. of potassium hydroxide (1,0 mole) dissolved
in 130 ml. of methanol. The solution was distilled directly fran the
reaction flask at the end of the reaction to give the following: fraction
1, b.p. 60-65; fraction 2, b.p. 68-74e. Fraction 2 we dissolved in
water and extracted with ether. There was no residue on evaporation of

the solvent. Fraction 1 was poured into cold water, the insoluble
layer separated, washed, dried over calciuIn chlrido, and redistilled.
Fraction bp. weight 2

1 67-70 12.0 g. 1.3487
2 74 14.0 g. 1.4194
3 92 32.0 g. 1.4285

For CG0CF2C1H F b;P p.V, 1.3340
For CF2BrCFCIBr: b.p, 93.5$ 5 1 4278

5. Reaction of CF p2BrOFOIdB th Alcoholic Potassiun IHdroidde in the
Presence of Acetone
Potassium hydroxide (66 g. 1.0 mole) was dissolved in 130 ml, of
methanol and added dropwiso to a refluidng mixture of 338 g. of
CF2BrCFCIBr (0.5 mole) and 75 ml. of aetone. Aat the end of the
reaction, the mixture was filtered to yield 95 g. of potassium bro-
aide and a filtrate which gavo no insoluble organle layer on pouring
into cold uatcr.
In the cold trap was found 37 g. of a liquid which added bromine
readily and boiled at -23, For CPFFCPCl, b.po -26

6. Reaction of CFBrCFC1Br with Alcoholic Potassium Hydroxide in the
Presence of Propionaldohyde
Potassium hydroxide (66 e., 1.0 nole) was dissolved in 130 ml. of
nothanol and added dropwise to a refluxing mixture of 138 g. of
CF2IrCFClB (0.5 mole) and 40 g. of ropdonaldehyde (0.7 mole). At the
end of the reaction, the mixture was filtered to yield 84 g. of salt

and a filtrate Uihich was poured into cold water to give an insoluble
organic layer. This material was washed several times with cold water,
dried over calciti chloride, and distilled to give two fractions:
fraction b.p 70-72 23.5 g. r 1.3392; fraction 2, b.p. 92,
33 .o, nD 1.4245.
For CIOCF2CIIFC1: b.p. 700 5 13340
For CFlroCF.lBr: b.p. 93.50 1.4278
In the cold trap was found 7.2 g. of a liquid which added brcmino
readily and boiled at -23o For CF2mCFCI, b.p. -2C~

7. Reaction of CF CICFCLI and Zinc in the Presence of Ethyl Acotate
In a 500 ml. three-necked flask wre placed 88 g. of ethyl acetate
(1.0 mole) and 50 g. of powdered cine (0.79 mole). This mixture as
cooled to 00 in an ico-bath and stirred vigorously while 100 g. of
CF C1CFC1I was added dropwso. After addition was canpletod, excess
since was filtered and the filtrate distilled. A fraction boiling at
77 and with a refractive index of 13723 was collected.
For othyl acetate: bp. 77 n91 1.3722.
A liquid whichh was found in the cold trap (25.5 g.) was bUbbled
through 35 g. of bramine. Ecess bromine was removed by washing with
sodi~n thiosulfate solution and the resulting organic product was dried
over calcium chloride. There was obtained in this manner 45 g. of a
liquid with the following properties: b.p., 92-94; nP 1.4255.
Reported for CFjBrCFCClr: b.p., 93.50 5 1.427.


General Procidure
All runs tire performed in a solvent cystcn consisting of tr

parts of diocane purified by the method of Fieser (310) and one part
of distilled later. The use of this solvent gave more reasurable
rates than cdald be obtained tith an alcohol colvant. All solutions
iere thcmostatod before dsizig in a constant tcWperature bath idhich
could' bo taintaind within I 0.01 of the desired setting.
The reaction vessel consisted of a 250 mil. volumetric fla:c
painted black to rdnnizo photol ytic dccoaposition of the halocarbons.
A side arm uas provided on the neck of this flask wich facilitated
raoval of samples through a pipotto by the application of positive

pressure into the flask. Loss of solvent by volatilisation was also
rniniAsod by this technique,
Into the reaction flask tvre pipottod 50 as of purified diomano
and 25 ml. of standardized aqu=ouso odiua hydroxide solution. Stan-
ddi rdation was performed under the iame conditions which wre anployed
to analyse each reaction sample. 1o accomplish this, the stock solu-
tion of sodium hydroaide 'as diluted ,Aith three parts of water, a 5 ml.
aliquot portion pipotted into 10 al. of access standard hydrochloric
acid and the resulting solution back-titrated uith standard base. The


halocarbon was voeighed into a glass ampoule and dropped into a black-

wallod flask containing 25 Bma of distilled water and 50 ml. of

purified diacsane. Both solutions were then placed in the thermostat-
ically controlled bath.
To start the reaction, 25 msi of the halocarbon solution was
pipetted into the solution of sodium~ lydromido contained in the
reaction flask and mixed well. Zero tine was taken when the pipotte
was half epty. Five milliliter Eoaples were rcioved at intervals
and reaction iras stopped by introducing the sample into 10 ml. of
standard hydrochloric acid. The blow-out proceduie was used and tinm
was marked at the half-empty point.
For CFBrCFr, the volatility of this compound necessitated a
modification of the above procedure. Samples of the halocarbon were
scaled into thin-walled glass bulbs uhich were in turn sealed in Pyrex
test tubes containing the base solution and a few glass boads. After
thcrmostating, the tube was shaken to break the glass bulb and start
the reaction. The tube was opened after the desired time interval and

the contents erer poured into excess standard acid and back titrated.
The stoichiometry of the reaction under consideration has not
been conclusively demonstrated but several simultaneous acid-base and
argentinxtric titrations have indicated that four moles of base react
for each nole of the dibramide. This stoichiamotric relation has
been c ployed in developing the folloduin kinetic depression hAich was
used in calculating the second order rate constants for these reactions.


2.303 b(a-4%)
k2 205L- Fo
t(a-4b) a(b-x)

A plot of time, t, verus log (a-4x) / (b-x) wflJ give & straight
limn whose elope in 2.303 / k2(a-4b). Frca this relation, k2 nayr
be determined.


b [CFICC21po 0.0395 I
T 30.20 o 0.03
ime (ia-4x) (b-x) (a-4x)
(secomns) zao2 x 3o A-)

115 9.63 3.54 0.4346
243 8.65 3.29 0.4198
440 7.81 3.08 0.4=04
683 7.13 2.91 0.3892
1055 6.36 2.72 0.3689
1703 5.49 2.50 0.3417
1900 5.22 2.43 0.3321
2485 4.57 2.27 0.3039
3100 4.11 2.15 0.2814
3735 3.73 2.05 0.2599

X2 2.2 x 10-2 noles/Iee.


a 2000 -
Sslope -2.3 x 10
A 0%

1000 -

26 30 35 40
Log (a-4x) / (b-x) x 102
Fig. 1.-Second order plot: reaction of CF2BrCC12Br with soditi
hydroxide in water-dioxane (1:2). a [Oj]o 0.1125 M;
b [CF2BrCC2Br] 0.0395 M; T 30.20 .01.



a =[Olrd 0,1175 11
b FrPFCIr oA 0.0327 MT
T w 30.20 1 0.010
The (a-4) (t(a) (a-4t)
(seconds) .02 0 log-

84 l,59 3.23 0.5549
235 fL.2 3.31 0.5533
460 30.72 3.01 0.5516
692 10.42 2.93 0.5510
10i3 10.06 2.84 0.5493
1600 9.63 2.73 0.5475
2540 80.87 2 0.5431
3605 7.97 2.31 0.5362
4cW0 7.53 2.21 0.5324
6243 6.94 2.06 0.5275

1c, 7-. x lCO4 nojNes/ ec.



Saslope -2,21 x 105


o 1 1 ."
53 54 55
Log (a-4x) / (b-x) x 102

Fig. 2.--Second order plot: reaction of CF2BrCFC1Br with
sodiun hydraxide in vater-diaman (1:2). a Eo]0o 0.1175 M;

b CF2B rCFC1 0.0327 M; T 30.20 0.01.


[~ old A*o00944 11
b ft EF2P1CF#]rO 0 .097 14'
T 0 30.20 O.C02

Thio compound vas unroactive under the conditions cmxpoyod
and consequonttl no rate constant is available.




The thermal decomposition of benzoyl peroxide in various solvents

has been extensively studied by a number of workers (44). Although
certain of these studies (36) have been primarily concerned with sol-
vent effects on the rate of decacposition, others (1,7133,18) have

emphasized the products obtained in order to determine the fate of
the free radicals so formed. Boeseken and Gelissen (3) reported the
only instance of the use of a polytaloalkane as the solvent in their
investigation of the decamposition of bensoyl peroxide in carbon
tetrachloride. Among the products isolated were carbon dioddo,

para-trichloromethylbenzoic acid, herachloroethane, chlorobenaene, and
phthalic acid. These compounds may be accounted for by equations
I through VII.

0- 0
C6HCOO* ..-- --. *"CH C+2 00
*C6H5+ *CC00 --- C06H50 + *CCCl3 I

2 *CC -------- C26 17

*CC + CF00* ----- CCAlOC00O* V
+ H*

CC3C61c4OO' + H' ----- CCcfHcooH VI

CC'CHooH -imsc CH4(COCeOcI)2 v

As part of the program being carried out in these laboratories
to study the peroxide induced addition of halofluoroalkanes to olefins,
it appeared of interest to canine the decomposition of benzoyl per-
ccide in the presence of certain of these halogen compounds.
The ethane derivative CF2frCFClBr was selected for one such
study both because of its convenient preparation and in view of its
ease of reaction with olefins by free radical mechanisms as established
by Lilyquist (31) and by Gilman (12). In general, the molar ratios
of reactants employed were one mole of the alkane and one-half of
peroxide. Table 8 gives the products formed in same typical reactions
by the thermal decomposition of benzoyl peroxide in CF2BrCFClBr*
The following equations can account for the observed products although
they are not necessarily the only possibilities.

c6HC0OCCA 5 --- w 2 06HC00- VIII
0 0

(c0HCOO.)cage C--. .c + *02 I

*C5 *- CF2BrCFaBr ------- -
COH5Br + *CFClGFjr

*CFCICF2Br + C061COO ------------- XI

CF2BrCFC61 COO. + H*
CF2BrCFC1C6H4CO* + H* -- ----- XII

Conspicuous for their absence are the coupled products
(CF2BrCFCl)2 and C615CFC1CF2Br vAich could be formed by the coupling
of radicals:

2 CF2BrCFC -------(CFBrCFC1) XI

*CFC0CF?2r + *C *A- C-H-CFC1CF2fBr Xi1

The compound shown in equation XIV would be particularly desirable as
a convenient route to o( v p ,p -trifluorostyrene.
DeTar and Weis (8) found that the decomposition of bis- 6 -phenl-
valeryl peroxido in carbon tetrachloride yielded no 1,1,1-trichloro-
5-phenylpentano although they were able to obtain the coupled products
hexachloroothano and 1,8-dipheryloctane:

c(c ooc (c ----) cn xv
0 0

2 C6H5(C(2).4COo

C0H5(CH2)4c00* C065(CHc)4 + c02 xyi

C6Ul5(cH2)4. + c4 x---- xvn
c6115(Cg2)3cu2cl + *CC13



Reactants Products HIols $ C02

D enzoyl Perosdde 0.25 100
CFP2rCFC1Br 0.50

C00 0.265 53
CF2frC6FCC 4CO0OH 0.126 25

06H 5r 0.23
Bensoyl Poroxide 0.25 66.7

CFwrFClalr 0.25
C61fCOOl 0.125 33.3
C02 0.1 26

ClCO5000 0.032 10
CFPrCFClc6l COOH 0o1 26

CeHpBr 0.085
Denzo1l Pamroid 0.125 100
CFprCFCIBr 0.25

064(C F3)2 0*5
Cq16C00 0.016 6.5
CF2DrCFClICI04COO 0.0o 32
c6H5fr 0.94

2 *CC5 ---- 2 XVII

2 C6H5(CH2)4" -- C- 0615(1C2)cC6I5 X

ced6 (1 *.. + *CC -- xx

The failure for the reaction represented by equation XX to occur
corresponds in the present instance to the absence of any C6IIC1FrCFC Dr
among the products when CP2rCFCIBr is the solvent for the deacopooi-
tion of benaoyl porocidd. DTar and Weis (8) hypothesized that the
4-phenyl-l-butyl radical cabination was a "cage" reaction (35) and
that any radicals escaping froa the cage reacted rapidly with carbon
tetrachloride so thatat any time the concentration of free 4-phenyl-
1-butyl radicals was negligible.
Although these postulates may explain the absence of the cross-
coabination product, CFjrFCPC61"U, in the lnmediate study, the failure
to observe any of the svlple ca binations, (CF2CCFC1)2 and biphenyl,
may be attributed to the difficulty in distinguishing such mall amounts
in the coq.plo reaction rxture.
Isolation of the alcylated bensoic acid was at first unoxpectcd
but its formation appears reasonable on the basis of equations XI and
XII. The structure of this material was proved not only by a deter-
nination of the silver and neutralisation equivalents but also by con-
version to terophthalic acid by dehalogenation and oxidation. The
terophthalic acid was characterized by its bis-(para-nitrobenzyl) ester.

his reaction may be classified as a free radical aromatic sub-
stitution by way of a secondary radical. After the transformation had
been characterized, a review of the literature revealed a number of
studies on free radical aranatic substitution. In 1934, Grieve and
Iey (33) decomposed sodium benconediazoato in a mlaxturo of nitrobonscne
and toluene and found that nitrobenzne was phenylated in nuch greater
proforonco to toluene under competitive conditions. They also ob-
served that ortho-nitrotoluone thon phenylated by the decomposition
of sodium banonediaoate gave only 4-nitro-3--mcthlbiphonryl

C0I6I5'20 a --- -- *C6115 53I

.c6H5 102 ----------- x


Later, Hey (18) decaiposod benzoyl pcraidoe in chlorobenseno obtaining
4-chlorobiphenyl, and in nitrobenseno obtaining both the A-nitro and
the 2-nitrobiphenyl.derivativos.
Recently, Augood, a al.(1) established partial rate factors for
phenylation of several substituted aromatic compounds under competitive
conditions. For ePample, among the mono=alogenated bensenes, the
relative rates toward phenylation are: C6Eg I ) >CIL r >) C5Cl ) C6IIF;
all are activated with respect to bonzeno and the proportions of isomer
stand in the order: ortho > para) > eta.

It is to be noted, however, that the studies referred to above
are concerned with primary radicals, in particular the phonyl radical,
in contrast to the secondary radical which is of more significance in
the present study.

Although these investigations of aromatic free radical substi-
tution are still in progress, a number of new concepts are being formu-
lated with respect to free radical substitution of the aromatic nucleus.

Improved techniques for product identification have made it possible

to account quantitatively for ortho, meta, and para isoaers and thereby
to arrive at certain conclusions regarding directive influences of
substituent groups .

Relevant to this are the studies by Dannly (7) on free radical
aromatic substitution who has shown that the electronic characteris-
tics of not only the aromatic nucleus but also the substituting free
radical are important in determining orientation.
It is of interest in connection with the present investigation
that the most easily isolated isoner in many of the reactions recorded
in the literature,and in several cases the only product reported,is
the para isomer. However, the other ring isomers have been identified
by more refined techniques such as spectrophotometric analysis.
Similarly, the decomposition of bensoyl peroxide in CF2BrCFClBr gave

a 50;1 yield of the para alkylated benzoic acid as shown by its conver-
sion to terephthalic acid. None of the ortho or neta iscmcr was

Efforts to extend the applicability of this reaction have met
with only limited success. For example, the addition of excess benzoic

acid to the reaction flask resulted in an increased yield of the
alkylated acid. However, an excess of benzene as co-solvent for the
peroxide decomposition gave none of the desired alkylated benzene.
Another attempted modification involved the decomposition of the

silver salt of perfluoropropionic acid. Inasmuch as the thermal
decomposition of these salts undoubtedly yields intermediate free

radicals (30), it was expected that alkylation could be accomplished
by heating the salt in the presence of an aromatic compound

2 CF3CF2COOAg -2 o p 2 CFCF2* + Agg20 mII
+ c02 +CO

CYF3F2* + C8% -- CF3F2C6H H li XXIV

Decomposition of silver perfluoropropionate in the presence of

benzene gave only unreacted benzene and a small amount of low-boiling

liquid -hich was probably perfluorobutane formed by the combination of
perfluoroethyl radicals. The use of nitrobensene as solvent led to
a violent explosion lhich dcnolished the autoclave and auxiliary

It appears likely that the reaction of peroxides with haloalhanes
may have considerable usefulness in introducing fluorine-contaird.nn side
chains on the aromatic ring. There is at present no satisfactory
method for preparing perfluoroalkylbenscno derivatives, for instance,

and a possible route is the decomposition of peroxides in perfluoro-
alkyl iodides:

C6OOCc ----F 2 C6H5* + 2 CO2 XXV

C6* Rf f--1-- -- C6 + x5l

RfCH* 00+* + H* ---- C6H.4C XIXTII

The product isolated in substantial yields frame the decaoposition
of bonzoyl poroxide in CF2BCFCBr CFrCFC1Ce000IIeOfI, may provide a
useul intermediate for the preparation of do, s/-trifloarostyrenev

CFirCFlC OO4 000 al CF2PCaFC64P H XXIX
CF2CFC6CO a -c no CgrCnOC65 + co02

This ccagpound has been, prepared previously by a lengthy synthesis (46)
involving the following steps:

CF2C10o0Ta 21--3 CF20C10C1 XxXI

C6CocFe2l M-----5- C6HeCCF2Cl XXII
C6iCC1CFC ---2c -- C6 5C lCF2PC xxxxv
6>CF"1CF2C1 alcohol V C Hc XXXV

The decomposition of benzoyl peroxide in CHZBrCCl2Br was under-

taken primarily for the purpose of explaining the observation that

certain ytdrogen-containing haloalanes do not add to olefins in the

presence of peroxides (27). It was felt that perhaps side reactions

deplete the source of initiating free radicals and that it would

therefore be of interest to determine the products of the decomposition

of benzoyl peroxide in quantities larger than the usual catalytic


The identification of all the products has not been completely
achieved and the failure to obtain an accurate materials balance has
hindered the successful interpretation of this reaction.

The small amount of benzoic acid (0.02 mole) associated with the
relatively large quantity of bromobenzene produced (0.30 mole) and the

absence of benzene do not indicate a preferential attack on the hydrogen

atom of this compound. On the other hand, it is apparent that the bro-
mine atom is readily accessible to attack ly the pheryl radical so that

the reaction is thus far analogous to that employing CFPrCFCIBr.

However, the product analysis does not definitely reveal the course of

reaction from this point.

There is some evidence that one possible path may involve beta

elimination of a bromine atom to give vinrlidene chloride:


*CC12CHIr ----- CCI23C2 + Br" XXXVII

ScIMhrling and mest (37) and West and Schnerling (47) have shown
that polychloroothano radicals undergo beta eliminations of this type.
For example, these workers reacted polyhalamethanes with saturated
hydrocarbons in the presence of peroxides and observed exchange of
hydrogen and halogen (47):

.OORt ---- 2 RO* ( R t-C H9 or XXXVI

RIto* + M --- RH + RB XXX3I
-* + CC14 C1 *CC1 XL

*CCL RH --- CIIC R ,otc XLI

However, when CCl3CC- was used, CC2olCC2 was obtained rather than the
expected CC3CIIKC2 A beta elimination of the inteoediate free radical
was proposed as a plausible explanation:


*CC 12CO --- CC12CC12 + *C1 XLII

"*1 + RH -1 -------- I1 + *R, etc. XLIV

Similarly, when hydrocarbon radicals reacted Swth chloroolefins, the
products .'ere alkylatod chloroolefins, RICXCX2 where X Cl or 1 (37)-
Again, the reaction mechanism proposed involved beta elimination fro
the intermdiate freo radical:

'*oCiC0CIR --- CX~I~iCCR + X' XLVI

Although in the present instance only a sa0lI amount of C%"=CC
(0.036 mole) vau isolated, the ready poflgerUabitty of this material
nay lead to the formation of teloaers and polyiners. However, there is
no definite ta evidence f for this although there was obtained
12 g. of a non-volatile viscous residue.
This does not constitute an adequate explanation for the failure
to obtain addition products by reaction of CFBrCHI~r with lefins in
the presence of peroxides (27). Beta elimination of brawino fros
*CIHlCF2Dr would not be expected in view of the difficulty of hoaolytie
sciesion of the carbon-bramine bond. Thus the inteaediate radical
derived fran CF2WCFBr is suffcient3y stable to give good yields of
addition products with olefine.
A nore thorough study of these hydrogen-containing haloethaes is
necessary to clarify their reaction with peroddes.


A. Decomposition of Bensoyl Peroxide in CF2BrCFClBr

General Procedure:

A reaction vessel was arranged consisting of a 500 ml. three-

necked flask fitted with a mechanical stirrer, ice-water reflu

condenser, and a connecting auxiliary flask for the addition of solid
bonzoyl peroxide. To the open end of the reflux condenser was con-

nected a vapor absorption train which was composed of a calcium chloride

tube, Dry Ice-acetone cold trap, two weighed Ascarite tubes for the

absorption of carbon dioxide, and a terminal calcium chloride tube.

The general procedure used in these decomposition experiments

involved heating the flask containing the CF2BrCFCIBr and ary other

components on a steam cone and adding the peroxide in five gram incre-

ments over a period of several hours. At the end of this time, the

system was aspirated with C02-free air in order to sweep all the

liberated carbon dioxide into the Ascarite tubes.

I. With CF2rCFCIBr Alone

a. Decomposition of the peroxide: Benzoyl peroxide (60.5 g.,
0.25 mole) was added in the manner described to 138 g. of CF2BrCFC3Br
(0.5 mole) at steam bath temperature. At the end of the reaction,

11.67 g. of carbon dioxide (0.265 mole) had been absorbed by the Ascarite
tubes. The reaction mixture was extracted four times with 50 ml. por-
tions of 12% sodium bicarbonate solution. The organic layer was

ccparated and dried while the aqueous solution was acidified to yield

45 go of solid material. Extraction idth hot water gave 2 g. of a
vbwte crystalline solid, m.p. 90, *wich was not identified. The
remainder of the solid was recrystallized froa petroleum ether to give

40 g, of crystals, n.p. 108-1090

Anal. Calcd. for C9HfrClFP 02t silver equiv., 158.5; neut. equiv.,
317. Found: silver equiv., 163; neut. equiv., 313.
The organic liquid was distilled through a 25 ca. column with protruded
packing to give 36.7 g. of unreacted CF2BrCFCIBr (0.333 mole) and
36.0 g. of braaobensone (0.23 mole) There were also obtained much
Smaller quantities of unidentified high boiling materials. Fourteen
crans of an unidentified liquid was found in the cold trap.

b. Preparation of trifluorovinylbenzoi acid, CF2.FC6H0COOII:
In a 200 ml. three-necked flask fitted *ith stirrer, reflux condenser,
and addition funnel were placed 6.1 g. of powdered zinc, 0.5 g. of
zinc claoride, and 25 nl. of absolute ethanol. The product obtained in
the previous c -erioent (m.p. I08-1090), 27 g., was dissolved in 100 ml.
of absolute ethanol and added dropise to the rofluding alcoholic
suspension of zine. A vigorous reaction occurred and the reflux rate
was controlled by careful addition of the ample.
At the end of the reaction, the contents of the flask were poured
into a beaker of cold rater and a voluminous hitte solid precipitated.
The solid wam filtered and washed uith water until a negative halide ion
test was givon by the mwshings. The product was dried in an oven at

85. Yield, 33 g., n.p. 156-157.

Anal. Calcd. for C9HF3o2: neut. equiv., 202; C, 53.46; H, 2.47;

Found: neut. equiv., 205; C, 53.71; Hi, 2.51.

c. Preparation of terephthali acid: In a 500 ml. beaker our-
rounded by ice--ater were placed 9 g. of the product obtained by

dehalogonation frcO the previous reaction, 8.4 go of sodium bicarbonate,

and 50 iml of water. Potassium permangafteI (9.04 g.)was dissolved in

water and added slodly with stirring to the ice-cold solution of olefin.
After addition was completed and the evolution of gas had ceanod, the
mixture was allowed to ccme to room temperature and excess potassium

permanganate was destroyed with sodium bisulfite. ll*~aneso dioxide
was removed from the solution by filtration after which the clear
filtrate was acidified to yield 4 go of a white precipitate. This
precipitate was filtered and dried in an oven.

d. Preparation of the bis-(para-nitrobenayl) ester of terephthalic
acid: In an eight inch test tube were placed 300 nag. of the acid iso-
lated in the previous reaction, a drop of phenolphthalein solution, and
enough 5% sodium bicarbonate solution to render the mixture slightly pink

in color. Para-nitrobensyl bromide (400 ag.) and 8 ml. of ethanol were
added and the solution was refluxed for two and one-half hours on a
steam conoe After the solution had cooled, 2 ml, of water wa added
and the ester filtered. The product was washed with two 4 ml. portions

of 5% sodium bicarbonate solution, then with water, and rccrystallized
froa ethanol, m.p. 262-2630.

Reported melting points of biM-(para-nitrobenyl) enters of the
isoneric bensene dicarbaxylic adcds: phthalic acid, 154.-1550; Iso-
phthalic acid, 20e2; terephthali acids 26a3

2, DIecoposition of Btensoy Peroxide fn CF BrCFClBr and Bensole Acid
A oixturo of 69 g. of CF2BrCFClBr (0.25 mole) and 15 g. of bensoic
acid (0.125 inle) was placed in a 200 al. three-necked flask provided
with a mechanical stirrers, solids addition flask, and a refaux condenser
to which was connected an Atarite tube for the absorption of carbon
dioxide. Bensoyl percd.do (30.2 go, 0.125 mole) was added in incro-
ments as the contents of the reaction flask were heated on a steam cone.
At the end of the reaction, the system ws swept with dry, C02-free
air in order that all the liberated carbon diaride might be absorbed
in the traps. II this manner, 4.45 &. of the gas (0.1 mole) was
accounted for.
he reaction mixture was Retracted with three 100 imr portions of
1 K sodi hydraxide solution. lhe organic liquid was dried over
calcium chloride and the aqueous portion was acidified to yield a
white solid, Extraction with hot water gave 3.9 go of crystal3ine
material, n.p. 120; reported for bensoic acid, m.p. 122. he mater-
ial insoluble in hot water was dried to give 31.2 g. of the compound
while a same regeystatised froa petroleum ether mnatod at 1080
2he organic liquid was ditilled to give two fractions with the
follomiog characteristics: fraction 1; b.p. 88-9%, n 1.4254, 15.1 g.
traction 2, bp. 145-1560, ~~ 1.5437, 33.3 g. Fraction 1 is
CF2DrCFC1Dr and fraction 2 is bramobensene on the basis of these con-

stants. Based on unrecovered CF2BrCFClBr, the yield of alkylated
product, CF2BrCFClC6H4C00, is 5Q~0

3. Decomposition of Benzoyl Peroxide in CF BrCFCBIr and bis-
In a 500 ml. three-necked flask, 69 g. of CF2BrCFClDr (0.25 mole)
was dissolved in 107 g. of bis-(trifluorasethyl)bensene (0.5 mole) and
reflu:ed on a steam cone. To this refluring solution was added
30.2 g. of bonzoyl peroxide (0.125 mole) over a period of two hburs
and heating ias continued overnight. The reaction ditui-c ias e~tractod
with dilute alkali to give an organic portion and a basic extract. fhe
latter was acidified and the resulting solid was washed with boiling
water. On cooling, the washings yielded 2 g. of white crystals,
n.p. 122 (bensoic acid, *mp, 122) while the water-insoluble solid
melted at 107. Tanty-fivo grams of this compound (CF2BrCFClC6HjCOOH)
was obtained for a yield of 41S based on unrecovered CF2BrCFC1Br.
STe organic portion was distilled to give three fractions uith
the following characteristics: fraction 1, bp. 90-96o, nI 1.30;
15.1 g,; fraction 2, b.p 100-1150 ( flat .10), 0 1.3852, 108.5 g.t
fraction 3, b.p. 154-156, 0 1.5495, 21 8 g.o Fraction 2 represents
a quantitative recovery of bis-(trifluoronothyl)ben~cno while fraction
1 and fraction 3 are OF2 rCFCDr and briobnenno, respectively.

B. Decomposition of SiLver Per

1. Decoaposition of Silvwr Porfluoropropionate in Benseno

A 300 ml~ stainless steel autoclave was charged with 200 aml of

benzene and 68 g. of silver porfluoropropionato (0.25 mole) and heated

at 250 for four hours. At the end of this time, the vessel was
vented into a Dry ice-acetone cold trap to give a snail amount of low-

boiling material. The autoclave was opened and the contents were

diatilled to give 150 ml. of a liquid boiling at 80. o other

material was obtained.

2. Dccamposition of Silver Perfluoropropionate in IIitrobenseno

The autoclave Was charged with 68 g. of silver porfluoropropionate

(0.25 mole) and 100 ml. of nitrobensene. tMion the temperature of the

vessel reached 250, a violent explosion demolished the autoclave and

auwdliary oquipaont.

C. Docomposition of Bensoyl Peroxide in CIlP CCjr
1- I J I .... ._ ... ... . I .. ..

The reaction vessel waas sembled fra a 250 nl. three-necked

flask equipped with a mechanical stirrer, solids addition flask, and

a liquid temperature thermnmter. A alow stream of nitrogen obtained

frame the tank without further treatment was passed into the reaction

flask and through the vapor trap system consisting of a Dry ico-acetone

cold trap, a calcium chloride tube, and two weighed Asoarite tubes for

the absorption of carbon dioxide.

In a typical decomposition, 60 g. of bensoyl peroxide (0.25 mole)
was added in 5 g. increments over a period of three hours to 128 g. of
CI2BrCC2 Dr (0.5 mole) which was stirred vigorously and maintained at
a temperature of 5. The reaction mixture was heated for two hours
at this tperature after the addition of peroxide had been completed.
At the end of this time, 8.83 g. of carbon dioxide (0.2 mole) had been
absorbed in the Ascarite tubes.
Froa tho cold trap there was obtained 3.5 g. of a liquid with the
following properties: bp. 35-3 i 5 1.4310. This liquid added
brain readily and on standing formed an opaque solid nass. Reported
for CI1CC=: b.p. 37, 25 14292.
The contents of the reaction vessel (172.5 g.) were steam distilled
to give 111.5 g. of distillate and a pasty, non-volatile residue. After
drying the distillate over caleiu chloride, redistillation gave 46.26 g.
of brmobonszen (0.29 mole), b.p. 50-550/ 13 mn., m.p. of 2,4-dinitro-
derivative 71, reported m.p. 710o and 56.70 g. of unreacted CI jrC r
(0.22 mole), b.p. 65/ 13 mn..
The non-volatile residue from the steam distillation was dissolved
in ether and extracted with three 50 ml. portions of 11 sodimn hydrox-
ide. The aquoous basic portion was acidified and the precipitate was
cAtractcd rith hot rater. On cooling, the hot water extract yielded
2.5 c. of a solid which melted at 120-122c. Reported for benzoic acid,
m.p 1220. The hot water insoluble portion solidified on cooling to give
6*8 g. of material.


On concentrating the ether portion, 8 g. of a rhite crystalline

solid was obtained uhile further removal of solvent gave 12 g. of a
viscous liquid residue. The crystalline material neltod at 1020

after recrystallisation from ether and gave no qualitative test for

halogen on fusing with sodium. A mixed nrlting point with pure
bonzoyl peroxide was not depressed and the infrared spectrum was
identical with that obtained fran an authentic sample of benyl

In the course of this research, the reaction of CF2rCFC1Br

with alcoholic potassium hydroxide has boon found to give the olefin,
CF2CFC1, and saturated ethers of the type ROCF2CHFC1. It has been
proposed that the ethers arise by the addition of alcohol to the
intermediate olefin.
The mechanism of this reaction has been investigated and it
appears likely that an initial abstraction of a positive bromoniun
ion is involved. However, a kinetic study of the reaction of three
dibromides, CFBrCC31Br, CFrrCFC3Br, and CF2BrC2Br, was insufficient
to reveal the mechanism of the subsequent beta halide ion clinination.
The order of reactivity CF'BrCClBr > CF2frCFCB3r > CFBrCF2Br indi-
cates that the ease of beta elimination is a rate determining factor.
Other basic reagents such as sodium diethyl malonate, diethyl

amine, and the Grignard reagent also reacted readily with CFBrCFC]Br.

In addition, it was found that other vicinal dihalides such as

CFj2rCC12Br, CF2BrCF2Br, CF2CICFClI, and CF CCBrCF2 r were also dehal-
ocenated by alcoholic potassium hydroxide.
An attempt to utilize this reaction in syntheses failed when no
addition products were obtained between an intermediate carbanion of
the haloalkane and carborWl compounds. It is possible that the mecha-
nism of the reaction does not involve a carbanion of finite lifetime.
The decomposition of bensoyl peroxide in CF3rCFC1Br led to a


novel alkylation reaction whereby the acid CFPBrCFCl-C6HC000H was

prepared in substantial yields. It is suggested that this compound
mzy be a useful intermediate for the preparation of o( ,, -trifluoro-
styrene by dehalogenation followed by decarboaylation. The reaction
also holds some promise as a method of introducing perfluoroa3~yl side
chains on the aromatic nucleus tb the decomposition of bezoyl per'-
axide in perfluoroalkyl iodides. Attempts to alkylate the aromatic

ring of co-solvents such as bonzone and bis-(trifloaromelthl)benzene
were musuccesful. Likewise, the decomposition of silver perfluoro-
propionate in aromatic solvents failed to give alkylated products.
The decomposition of benzoyl poroaide in CH2BrCC12Br gavo sub-
stantial quantities of bromobenzene and there was some indication that
virylidone chloride, CH20CC12, was foromd. IIowrover, the failure of

this lydrogen-containing alkane to give addition products with olofins
in the presence of peroxides could not be explained on the basis of the
product analysis.


1. Atmood, D. R., J. J. G. Cadogan, D. H. Hey, and G. H. 1Yillians,

JO Chen. Soc, 3412 (1953)
2. Banus, J., H. J. ITeeleus, and R. I. Haszeldine, J. Chem. Soc.,

60 (1951)

3. Boescl:en, J. and H. GSlisscn, Rec. Tray. Chim., (, 869 (192.\)

4. Bolas, T. and C. E. Groves, J. Che. Soc., 2 164;

ibid, 783

5. Conrad, I.: and I. Guthzeit, Ber., 2, 28.;1 (C184)

6. Corley, 1. S., Final Surmar-y report IYo. 8(A), Feriod July 1-

Septceder 30, 1952, Polaroid Corporation, Canbridgo 39,

7. Dannlcy, Ralph, J. A,. Chei. Soc., L.53 (1954)

8. DeTar, DeLos F. and Claus Ueis, J. An. Chen. Soc.,

4296 (1956)
9. Dk;-on, Stanley, E. I. duPont do I:creurs & Compa~t/, Priv ate

corrunication to Dr. Paul Tarrant, University of Florida

10. Ficoor, t. F., "Ec:pcri-cnts in Organic Chcrnistry'

D. C. Hcath & C--2any, B3ooton and J.ew Yo-:, 1935
11. Fried, Helvin, M. S. Dhesis, Univcrsit:' of Florida, 1954

12. Gilian, E., I1. S. Ticsis, Univcrsity of Florida, 1954

13. Grieve, Vh. S. M. and Donald H. Hey, J. Chea. Soc., 1797 (1934)
14. Hanford, I. E. and G. I. Uigby, U. S. Patent 2,409,274 (1946)

15. Ilartuan, P. F. H. G. Sellers, and D. Turnbull,

J. An. Chem. Soc., 92, 2416 (19147)
16. IasZeldine, R. II., J. Chei. Soc., 4259 (1950)

17. Zlaszeldine, R. N. and B. R. Stccle, J. Chem. Soc., 1199 (1953)
18. sc;-, Donald H., J. Chae. Soc., 1966 (1934)

19. Hine, J. and W. D, Drader, J n.n.Chen. Soc., 3964 (1953)
20. Iir.o, J. and W. D. Brader, J. An. Chean. Soc., 2 3886 (1955)

21. Hurwitz, M. D. and W. T. Miller, Abstracts of Papers, 114th

llcetir; Acerican Chemical Society, lashinrton D. C.

August, 1948, p. 4 L
22. Inhold, C. K. and I. J. Poccll, J. Chcn. Soc., 1222 (1921)

23. Ingold, C. K*, 'tStructure and I:cclhaism in Organic Chemistr'-"

Cornell Uhiversity Press, Ithaca, N. Y. 1953, p. 310
24. Incold, C. K., ibid, p. 331
25. Ingold, C. K., ibid, p. 420
26. InGold, C. K., ibid. p. 422

27. Johnson, Robert, Unpublished reports for Offico o0f Uaval

Research Contract 1017 (00), University of Florida
28. lharasch, M. S. and R. J. Danlemy, JO. Or. Chem., 0s, 406 (1945)

29. I2iarasch, 1. S., E. V. Jansen, and W. H. Urry,
J. Am. Chen. Soc., 6a, 1100 (1947)

30. lirshcnbaun, A. D., A. G. Strcn, and 1. Hauptschein,
J. An. Chle. Soc., 7, 3141-5 (1953)

31. Lilyquist, R, Ph.D. Dissertation, University of Florida,

32. Lovelace, Alan UI., Ph. D. Dissertation, University of Florida,

33. cDcc, E. T. and R O0. Bolt, Ind. En7. Chan., 412 (1947)

34. Tcf, V-, Ann., ga., 330 (1699)
35. Hjyco, R. It., J. Am. Che,. Soc., Z, 2042 (1955)

36. TIoal:i, K. and P. D. Bartlett, J. Am. Chem. Soc., 8., 1686

37. Schnerling, Louis and J. P. West, J. Am. Chian. Soc.,

2o 2015 (1949)
38. Simons, J. I1. (editor), "Fluorine Chemistr;y, Vol. I,
Chapter 14 by U. H. Pearlson, Acadamic Press Inc., N. Y., 1950
39. Simons, J. H., ibid, Vol. II, Chapter 6 by J. H. Simons

40. Skell, P. S. and C. R. Hauser, J. Am. Chem. Soc., 6, 1661 (1945)
41. Tarrant, P. and H. Brown, J. An. Chem. Soc., ~2 5831 (1951)
42. Tarrant, P. and J. Young, J. Am. Chemn. Soc., a, 2432 (19!..9)

43. Tarrant, P. and D. Warner, J. Am. Chem. Soc., 1, 1624 (1954)
4A. Toboll!-y, A. V. and R. B. I:csrobian, "Organic Peroxides",
Interscience Publishers Inc., Iew York, N. Y., 1954

45. Urry, W. If. J. R. Eisznor, and J. W Wilt, J. Am. Chai. Soc.,

2, 918 (1951)
46. Warner, Dale, M. S. Thesis, University of Florida, 1950
47. Ucst, J. P. and Louis Scherlin;, J. Am. Chem. Soc., P, 3525 (1950)

48. Wolosinsfki, T., S. G. Cahen, and P. J. Scheuer, J. Am. Chean
So_., L 3439 (1949)


Floyd Edward Bentley was born on Eovmber 16, 1932 at Richmond,
Virginia, the cocond of five children of Rev. and Mrs. Floyd Thomas

Bentley. He vas educated in the secondary schools of Virginia and

graduated as valedictorian frca Jefferson Senior High School in

Roanoko, Virginia in June, 1949.

Ie received the B. S. degree in chemistry from Randolph-Ilacon
College in Ashland, Virginia in June, 1953 and began graduate study

in organic chemistry at Case Institute of Technology in Cleveland,

Ohio. The work prerequisite to the II. S. degree as completed in
February, 1955 and the degree was awarded to him by that institution
in June, 1955. The author entered the University of Florida in
February, 1955 and undertook the program leading to the Ph. D. degree

in organic chemistry. He has been a National Science Foundation
Fellmo for two years and worked for one semester on a project opon-
sored by the Office of ITaval Research.

The author is a member of Phi Beta Kappa, Omicron Delta Kappa,
Chi DOta Phi, Sigma Xi, and the American Chemical Society, He is

married and has one child.

This dissertation was prepared under the direction of the Chair-
man of the candidate's Supervisory Ccamittee and has been approved
by all meubors of the committee. It was submitted to the Dean of the
College of Arts and Sciences and to the Graduate Council and was
approved as partial fuhlfiltent of the requirements for the degree of
Doctor of Philosophy.

June 3, 1957

Dean, College of Arts and Scionceo

Dean, Graduate School



^ ^2IL~



Print Page 1 of 2

From: lib-ufdissertations (ufdissertations@uflib.ufl.edu)
To: 'ebentley7@sbcglobal.net'
Date: Tuesday, March 10, 2009 3:07:47 PM
Subject: UF Libraries:Digital Dissertation Project

Dear Dr. Floyd E. Bentley,

The George A. Smathers Libraries at the University of Florida has initiated a project to
retrospectively digitize and make available on the Internet any dissertation written by a
University of Florida doctoral candidate and accepted by the University of Florida It is our goal
to make the documents fully text searchable and easily harvested by Internet search engines,
allowing the full breadth and scope of scholarship produced at the University of Florida to be
made available across the world quickly and easily. The Library is bearing the full cost of the

We would like to add your dissertation, Some reactions of 12-Dibromochlorotrifluoroethane.,
published in 1957, to the project. In order to do so we need a signed, legal original Internet
Distribution Consent Agreement for our files. If you want your dissertation included in the
project, please print the Consent Agreement on the page below, sign it and mail it back to the
Libraries at the address listed.

We will keep you informed of the progress of your dissertation as it works its way through the

If you have any questions, please reply to UFdissertations@uflib.ufl.edu.

Thank you,

Cathy Martyniak, Project Coordinator
Christy Shorey, Project Technician

The Foundation fo r The Gaator Natfion

Internet Distribution Consent Agreement

In reference to the following dissertation:

AUTHOR: Bentley, Floyd
TITLE: Some reactions of 12-Dibromochlorotrifluoroethane. (record number:



Print Page 2 of 2

I, //d y l /LEf re/ d 8n ,-/fey as copyright holder for the aforementioned
dissertation, hereby grant specific and limited archive and distribution rights to the Board of Trustees of
the University of Florida and its agents. I authorize the University of Florida to digitize and distribute
the dissertation described above for nonprofit, educational purposes via the Internet or successive

This is a non-exclusive grant of permissions for specific off-line and on-line uses for an indefinite term.
Off-line uses shall be limited to those specifically allowed by "Fair Use" as prescribed by the terms of
United States copyright legislation (cf, Title 17, U.S. Code) as well as to the maintenance and
preservation of a digital archive copy. Digitization allows the University of Florida or its scanning
vendor to generate image- and text-based versions as appropriate and to provide and enhance access
using search software.

This grant of permissions prohibits use of the digitized versions for commercial use or profit.

Signature of Copyright Holder

FIoy C E1uW27rd /5^^/y
Printed or Typed Name of Copyright Holder/Licensee

Personal information blurred

Date of Signature

Please print, sign and return to:
Cathleen Martyniak
UF Dissertation Project
Preservation Department
University of Florida Libraries
P.O. Box 117008
Gainesville, FL 32611-7008

http://us.mg201.mail.yahoo.com/dc/launch?.partner=sbc&.rand=Ooi4cmapuh5p6 ... ,/10/2

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs