Some reactions of the 5 perfluoroalkyltetrazoles

PDF ( 4 MBs ) ( .pdf )

SOME REACTIONS OF THE

5-PERFLUOROALKYLTETRAZOLES













By
ROBERT JAMES KASSAL


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY













UNIVERSITY OF FLORIDA
December, 1963











ACKNOWLEDGMENTS


The author wishes to extend his thanks to his

research director, Dr. H. C. Brown, for his guidance

throughout this investigation. He also wishes to thank

the other members of his committee, Dr. G. B. Butler,

Dr. W. M. Jones, Dr. J. D. Winefordner and Dr. W. V. Wilmot,

for their aid and encouragement.

The author is indebted to Dr. W. S. Brey for N.M.R.

measurements and their interpretation which were used to

identify several new compounds.

The success of this work was in no small way due to

the patience and encouragement of the author's wife,

Barbara, and his children, Christopher, Kenneth and Cynthia.












TABLE OF CONTENTS


ACKNOWLEDGMENTS . . . . . . . .

LIST OF TABLES. . . . . . . . . .

LIST OF FIGURES . . . . . . . . .

PRELIMINARY REMARKS . . . . . . . .

Chapter

I. INTRODUCTION . . . . . . . ..

Statement of the Problem . . . . .

II. DISCUSSION . . . . . . . . .

Preparation and General Properties of 5-Per-
fluoroalkyltetrazoles . . . . ..

Reactions of the 5-Perfluoroalkyltetrazoles
Below the Thermal Ring-Opening
Temperature . . . . . . . .

Synthesis of 2,5-bis(perfluoroalkyl)-
1,3,4-oxadiazoles. . . . . .

Synthesis of 1,5-bis(5-perfluoroalkyl-
1,3,4-oxadiazolyl-2)perfluoropropanes.

Synthesis of 5,5'-bis(perfluoroalkyl)-
2,2'-bi-1,3,4-oxadiazoles. . . .

Reaction of 5-perfluoropropyltetrazole
with phosgene. . . . . . .

Reaction of 5-perfluoropropyltetrazole
with perfluorobutyronitrile: synthesis
-of 3,5-bis(perfluoropropyl)-1,2,4-
triazole . . . . . . . .


iii


Page

ii

vii

viii

ix



1

7

9


9


12


12


15

18


21



26







Page


Reactions of 5-Perfluoroalkyltetrazoles at
Temperatures Above the Thermal Ring-
Opening Temperature . . . .3 . 30

Reactions with perfluoroalkylnitriles:
synthesis of 5,5-bis(perfluoroalkyl)-
1,2,4-triazoles. .. .. . .. . 0

Reactions with perfluorobutyne-2:
synthesis of 3,4,5-tris(perfluoro-
alkyl)pyrazoles. . . . . .. 32

Reactions of the perfluorobutyro-
nitrilimine fragment under mild
conditions . . . . . . . 37

Thermolysis of 5-perfluoropropyltetrazole
with no acceptor molecule present. . 38

Infrared Spectra and Acid Strengths of the
Perfluoroalkyl Substituted Pyrazoles,
Triazoles and Tetrazoles. . .. . .. 45

Infrared spectra. . . . . . . 45

Acid strengths. . . . . . . 53

III. EXPERIMENTAL ........ . . . 55

Source of Materials. . . . . . . 55

Method of Molecular Weight and pKa Determi-
nation of the 5-Perfluoroalkyltetrazoles
and 3,4,5-tris(Perfluoroalkyl)pyrazoles 56

Preparation of 5-Perfluoroalkyltetrazoles. 56

5-Perfluoromethyltetrazole. . . . 56

5-Perfluoropropyltetrazole. .. . . 57

1,3-bis(5-Tetrazolyl)perfluoropropane . 60

Preparation of the Disodium Salt of
Bitetrazole . . . . . . . 61

Hydrolytic Stability of 5-Perfluoropropyl-
tetrazole . . . . . . . 62

iv







Page


Oxidative Stability of 5-Perfluoropropyl-
tetrazole . . . . . . . .

Synthesis of 2,5-bis(Perfluoroalkyl)-1,3,4-
oxadiazoles by Reaction of 5-Perfluoro-
alkyltetrazoles with Perfluoroacyl
Chlorides . . . . . . . .

Synthesis of 2,5-bis(perfluoropropyl)-
1,3,4-oxadiazole . . . . . .

Synthesis of 2-perfluoropropyl-5-per-
fluoromethyl-l,3,4-oxadiazole. . .

Synthesis of 2,5-bis(perfluoromethyl)-
1,3,4-oxadiazole . . . . .


S 63



S 63


S 63


S 64


S 65


Synthesis of 1,3-bis(5-Perfluoroalkyl-l,3,4-
oxadiazolyl-2)perfluoropropanes . . . 66

Synthesis of 1,3-bis(5-perfluoropropyl-
1,3,4-oxadiazolyl-2)perfluoropropane . 66

Synthesis of 1,3-bis(5-perfluoromethyl-
1,3,4-oxadiazolyl-2)perfluoropropane . 67

Synthesis of 5,5'-bis(Perfluoroalkyl)-2,2'-bi-
1,3,4-oxadiazoles . . . . . . 68

Synthesis of 5,5'-bis(perfluoropropyl)-
2,2'-bi-l,3,4-oxadiazole . . . . 68

Synthesis of 5,5'-bis(perfluoroethyl)-
2,2'-bi-l,3,4-oxadiazole .. . . .. 69

Synthesis of 5,5'-bis(perfluoromethyl)-
2,2'-bi-l,3,4-oxadiazole . . . . 70

Attempted Isolation of l(2)-Perfluorobutyryl-
5-perfluoropropyltetrazole. . . . . 71

Reaction of 5-Perfluoropropyltetrazole with
Phosgene. . . . . . . . 74

Reaction of 5-Perfluoropropyltetrazole with
Perfluorobutyronitrile Under Mild Con-
ditions: Synthesis of 3,5-bis(Perfluoro-
propyl)-l,2,4-triazole. . . . . . 77








Page


Attempted uncatalyzed reaction of 5-
perfluoropropyltetrazole with per-
fluorobutyronitrile . . . .. 77

Boron trifluoride etherate catalyzed
reaction of 5-perfluoropropyltetra-
zole with perfluorobutyronitrile. . 78

Gaseous boron trifluoride catalyzed
reaction of 5-perfluoropropyltetra-
zole with perfluorobutyronitrile. . 79

Reaction of 5-perfluoropropyltetrazole
with perfluorobutyronitrile in the
presence of gaseous hydrogen chloride 81

Attempted Reaction of 5-Perfluoropropyl-
tetrazole with Perfluorobutyne-2 in the
Presence of Gaseous Hydrogen Chloride. 84

Reactions of 5-Perfluoroalkyltetrazoles
Under Thermal Ring Opening Conditions. 84

Synthesis of 3,5-bis(perfluoropropyl)-
1,2,4-triazole in a static system . 84

Synthesis of 3,5-bis(perfluoroalkyl)-
1,2,4-triazoles in a flow reactor . 85

Synthesis of 3,4,5-tris(perfluoroalkyl)-
pyrazoles by reaction of perfluoro-
butyne-2 with 5-perfluoroalkyltetra-
zoles . . . . . . . . 90

Reactions of perfluorobutyronitrilimine
on a cold finger. . . . . . 93

Thermolysis of 5-perfluoropropyltetra-
zole with no acceptor molecule present 98

Deuteration of the Perfluoroalkyl Substi-
tuted Tetrazoles, Triazoles and Pyra-
zoles. . . . . . . . .. 99

IV. SUMMARY .. . . .. .. . . .. 101

BIBLIOGRAPHY . . . . . . . . . 104

BIOGRAPHICAL SKETCH. . . . . . . . 107











LIST OF TABLES

Table Page

1. 2,5-bis(Perfluoroalkyl)-1,3,4-oxadiazoles . 13

2. 1,3-bis(5-Perfluoroalkyl-1,3,4-oxadiazolyl-2)-
perfluoropropanes . . . . . . .. 17

3. 5,5'-bis(Perfluoroalkyl)-2,2'-bi-1,3,4-
oxadiazoles . . . . . . . . . 19

4. The Effects of Hydrogen Chloride on the
Reaction of 5-Perfluoropropyltetrazole with
Perfluorobutyronitrile. . . .. . . 29

5. Synthesis of 3,5-bis(Perfluoroalkyl)-1,2,4-
triazoles in a Flow Reactor . . . . . 32

6. N.M.R. Spectral Data of 3,4,5-tris(Perfluoro-
alkyl)pyrazoles . . . . . . . .. 36

7. N.M.R. Spectral Data of 2-H-Perfluorobutene-1 40

8. pKa Values of Acidic Azoles . . . . . 54


vii












LIST OF FIGURES


Figure Page

1. Infrared spectra of: (A) 5,5'-bis(perfluoro-
methyl)-2,2'-bi-1,3,4-oxadiazoreT (B) 5,5'-
bis(perfluoroethyl)-2,2'-bi-1,5,4-
oxadiazole; (C) 5,5'-bis(perfluoropropyl)-
2,2'-bi-1,5,4-oxadiazole . . . 20

2. Infrared spectra of: (A) N,N'-bis(perfluoro-
butyrimidoyl chloride)urea; (B) 2-(per-
fluorobutyrimidoyl chloride)amino-5-per-
fluoropropyl-1,5,4-oxadiazole; and (C)
N-(perfluorobutyrimidoyl chloride)perfluoro-
butyramide . . . . . . . . 22

3. Infrared spectra of: (A) 2-H-perfluoro-
butene-1; and (B) 3,6-bis(perfluoropropyl)-
1,2-dihydro-1,2,4,5-tetrazine. . . . 42

4. Infrared spectra of: (A) 5-perfluoromethyl-
tetrazole; (B) 5-perfluoropropyltetrazole;
and (C) 1,3-bis(5-tetrazolyl)perfluoro-
propane. . . . . ..... .. 47

5. Infrared spectra of: (A) 5,5-bis(perfluoro-
propyl)-1,2,4-triazole; (B) 3-perfluoro-
propyl-5-perfluoroethyl-1,2,4-triazole; and
(C) 3-perfluoropropyl-5-perfluoromethyl-
1,2,4-triazole . . . . . . . 48

6. Infrared spectra of: (A) 3-perfluoropropyl-
4,5-bis(perfluoromethyl)pyrazole; and (B)
3,4,5-sris(perfluoromethyl)pyrazole. ... 49
70 Apparatus for attempted isolation of 1(2)-
perfluorobutyryl-5-perfluoropropyltetrazole. 72

8. Flow reactor for 3,5-bis(perfluoroalkyl)-
1,2,4-triazole synthesis . . . . . 86

9. Apparatus for preparation and reaction of
perfluorobutyronitrilimine . . . . 94


viii













PRELIMINARY REMARKS


All temperatures reported are uncorrected and in

degrees Centigrade.

Nuclear magnetic resonance measurements were

obtained using the Varian high-resolution spectrometer and

are expressed in parts per million (p.p.m.). Infrared

spectra were obtained using the Perkin-Elmer Infracord

recording spectrophotometer. Ultraviolet spectra were

obtained using the Beckman Model DK-2 recording spectro-

photometer.

Chromatographic analyses were obtained using a

Perkin-Elmer Model 154 Vapor Fractometer.

Elemental analyses were determined by the

Schwarzkopf Microanalytical Laboratory, 56-19 37th Avenue,

Woodside 77, New York.

The term RF appears frequently in this report and is

used to denote a perfluoroalkyl radical.











CHAPTER I


INTRODUCTION


The tetrazoles are a class of organic heterocycles

characterized by the presence of a five membered unsaturated

ring consisting of one carbon and four nitrogen atoms. The

parent compound, tetrazole, like the 5-substituted tetra-

zoles, exists in tautomeric forms I and II, and while the

N=N N --N
H-C H-C
N--NH N-N
H

I II

ring is sometimes referred to as "1-H tetrazole" the

hydrogen is apparently equally bound to all the nitrogen

atoms of the ring. This structure is apparent from the

facts that methylation of its salts with methyl iodide

produces both 1- and 2-methyl derivatives2 while methyla-

tion of the free tetrazole with diazomethane produces the

1-methyl derivative3 and acylation gives the 2-acyl

tetrazoles.

A more correct way of depicting the ring would be

as follows:







2


R- C( H
N- N

This structure would explain the complex N-H absorption in

its infrared spectra.

It would appear to the uninitiated that the tetra-

zoles would be relatively unstable due to the high nitrogen

catenation. This expected instability is shown when the

tetrazole ring is linked to a chain of nitrogen atoms;

tetrazolyl-5 azide explodes in aqueous solutions of concen-

trations greater than about 2 per cent at 0o.5 However,

the majority of tetrazoles are reasonably stable and a few,

such as 5-(benzoylamino)tetrazole, are stable at temperatures

approaching 5000.6 Tetrazole salts in general have been

found to be more thermally stable than the free tetrazole.

The tetrazole ring normally survives such reactions as

oxidation, reduction, decarboxylation and hydrolysis of its

substituents.

Only a few tetrazoles have achieved practical

importance. The heavy metal salts of a number of the less

stable tetrazoles have been described in patents as being

useful components in priming and initiating compositions

for explosives.7-9 1,5-Pentamethylene tetrazole (Metra-

zole) is extensively used as a general cardiac and respira-

tory stimulant.










The first tetrazole was prepared by BladinlO in 1885

by diazotization of -N-phenyl-C-cyanoformamidrazone.


SNH
NC-C


C6H5


N=N


+ HNO2 -- NC-C I
N -N-C6H5


A similar method employing nitrous acid and aryl

hydrazidines has become important in the synthesis of

5-aryl tetrazoles. The reaction is presumed to proceed via

an intermediate imidazide which rearranges to give the

tetrazole.


7NH


NH-NH2 + HNO.


NH N==-N
-- Ar-C -N ----Ar-C
S. N-NH


The reaction of hydrazoic acid or its salts with

organic nitriles produces 40-90 per cent yields of 5-aryl

or alkyl tetrazolesll and is also believed to proceed through

an intermediate imidazide.


RCN + HN 3


N= N
--N RC
N-NH


The syntheses of two 5-perfluoroalkyl-substituted

tetrazoles have been reported using the azide-nitrile

reaction.2,12


Ar -C










In 1929, Stolle13 reported that 5-aminotetrazole

formed the exocyclic N-acetyl derivative (III) when gently

warmed with acetic anhydride. With vigorous heating of

the reaction mixture, nitrogen was eliminated and the

1,3,4-oxadiazole (IV) formed.

N=N 0 N= -N
-NH-C + (CH3C0)20--> CH -C-N-C
N NH H N- N

SIII

reflux (CH CO)20
8 hrs

O 0
CH -C-N-C C-CH + N
H \ \ 3 2
N--N

IV


Recently the synthesis of 1,3,4-oxadiazoles by the

acylation of 5-aryl and alkyl tetrazoles has been examined

by Huisgen and his coworkerso14,15 They suggest the

following mechanism for the reaction.1 The majority of










/N=N N=N
R-C + R'COC1--- R-C 0
N--NH N-N-C -R'





S-N-N2 11
0 o 0 NN 2
R-C=N-N=C-R' -- R-C=N-N-C-R' <- R-C C-R'
R-C N=N'

VI V


0
R-C C-R'
N-N

4
available data indicates that VI forms directly from the

acylated tetrazole but kinetic data leave open the

possibility that the slow step is the transformation of V

to VI. In either case, VI is assumed to be the immediate

precursor of the oxadiazole.

Huisgenl6 also investigated an analogous reaction

with imidoyl chlorides to give 5,4,5-trisubstituted-1,2,4-

triazoles and proposed a mechanism analogous to that in the

tetrazole-acylchloride reaction. With only a few exceptions,

these syntheses of oxadiazoles and triazoles produced 75-100

per cent yields.










R"
N=N N NN N "
R-C + C-R'--- R-C N
SN-NH C1 N-N-C-R'


-N2
R"
N
R-CEN-N-C-R'

R" R"
N N
R-C C-R' R-C=N-N=C-R'
N-N R
R"
N
R- -N=N-C-R'

In the past several years, there has been considerable

interest in this laboratory in heterocycles containing

perfluoroalkyl substituents. One class of heterocycles, the

2,4,6-tri-s(perfluoroalkyl)-s-triazines exhibited excellent

thermal stability and synthesis of fluorocarbon polymers

containing s-triazine rings in the backbone was studied.17

The polymers reported suffered from the inherent drawback

of nonlinearity.

2,5-bis(Perfluoroalkyl)-1,3,1-oxadiazoles and 3,5-

bis(perfluoroalkyl)-l,2,4-triazoles were previously

synthesized in this laboratory. The oxadiazoles were

prepared by cyclo dehydration of a bis(perfluoroacyl)hydra-

zine18 and the triazoles by dehydration of an N-acyl









hydrazidine formed by the reaction of a 2,5-bis(perfluoro-

alkyl)-1,5,4-oxadiazole with ammonia.19 These heterocycles

also possessed good thermal stability and when incorporated

into a polymer structure probably would not lead to

branching or crosslinking.

In view of the 75-100 per cent yields of the

acylation reactions of the 5-substituted tetrazoles, a

similar method employing ditetrazoles and diacid chlorides

was considered as a synthetic route to fluorocarbon

polymers containing 1,5,4-oxadiazole rings in the backbone.

However, the information available on the reactions of the

5-perfluoroalkyltetrazoles was insufficient to determine

their suitability for use in polymerization reactions. It

was considered advisable, therefore, to investigate

initially those reactions of the 5-perfluoroalkyltetrazoles

which might furnish a route to other simple heterocyclic

systems. Information gained might then be translated to

polymer producing reactions.


Statement of the Problem

The primary purpose of this problem was to investigate

some of the ring-opening reactions of the 5-perfluoroalkyl-

tetrazoles as a route to the formation of new heterocycles

and to note the effect of the perfluoroalkyl group on the











reactions. Since some of these reactions were carried out

at temperatures above the thermal ring-opening temperature

of the pure 5-perfluoroalkyltetrazole, a further objective

was to show the mechanism of the thermal ring-opening

reaction and to trap and identify the intermediate by

reacting it with other materials to give more stable, known

species.

As the investigation proceeded, several new hetero-

cycles were prepared. A further objective of this

investigation therefore was to compare their spectra and

to compare the acidic properties of those heterocycles with

hydrogen attached to the ring nitrogen and to show the

effect of the perfluoroalkyl group on the acid strength.











CHAPTER II


DISCUSSION


Preparation and General Properties of

5-Perfluoroalkyltetrazoles


The perfluoroalkyl nitriles are readily attacked by

nucleophiles such as ammonia,20 hydrazine21 and hydrogen
22
sulfide. These reactions produce the stable perfluoro-

alkyl amidines, hydrazidines and thioamides, respectively,

by the following general mechanism:

/N NH
R5-C5N R F-C R F-C
B B-H B
B-H

where B = NH 2-, N2H -, SH-

A similar reaction with azide ion in anhydrous media results

in formation of the imidazide (VII) which rearranges

spontaneously to the tetrazole.

N:0 /N- N
RF -N -- -C R-C
E N3 N--' N
VII
VII










Prior to the appearance of Norris' paper describing

the synthesis of 5-perfluoromethyltetrazole in anhydrous

acetonitrile, 5-perfluoropropyltetrazole and 1,3-bis(5-

tetrazolyl)perfluoropropane were prepared in the present

study by reaction of the appropriate nitrile with sodium

azide in a methylcellosolve-acetic acid mixture using a

method similar to that described by Mihina and Herbst11

for the preparation of 5-alkyl and 5-aryl substituted

tetrazoles. The methylcellosolve, dried by azeotroping

with benzene, contained traces of water which hydrolyzed

part of the perfluoroalkyl nitrile to the amide. The amide

was readily removed from 1,3-bis(5-tetrazolyl)perfluoro-

propane by recrystallization of the product from perfluoro-

butyric acid. Perfluorobutyramide could not be readily

removed from 5-perfluoropropyltetrazole, however, and

purification of this tetrazole required preparation of its

silver salt, from which the amide could be removed by

washing with acetone. After regeneration and distillation,

5-perfluoropropyltetrazole of high purity was obtained and

slowly crystallized on standing at room temperature.

Norris' method, adapted to the preparation of

5-perfluoropropyltetrazole, gave a product of slightly lower

purity which did not crystallize at room temperature; how-

ever, the method is less laborious than the longer route

involving preparation of the silver salt.










The 5-perfluoroalkyl substituted tetrazoles are

colorless, odorless strong acids with the characteristic

sour taste of an acid. In contact with the skin, they

produced a chemical burn similar to that produced by glacial

acetic acid. Pure 1,3-bis(5-tetrazolyl)perfluoropropane

slowly decomposed at 1700 while 5-perfluoromethyltetrazole

and 5-perfluoropropyltetrazole decomposed at about 1800.

As has been observed with 5-aryltetrazoles,23 the sodium

salts of 5-perfluoropropyl and 5-perfluoromethyl tetrazole

were more stable thermally than the free tetrazole and

decomposed slowly at 2450 and 2580, respectively. The

latter salt detonated at 3000.

Silver 5-perfluoropropyltetrazole undergoes a slow

decomposition at 2900 which becomes very rapid, but not

explosive, at 3500. Both silver 5-perfluoropropyltetrazole

and disilver 1, -bis(5-tetrazolyl)perfluoropropane were

found to be stable to bright sunlight. Mihina and Herbst11

reported several silver tetrazoles were light sensitive but

also noted that they occluded silver nitrate. This

sensitivity may be due to traces of occluded silver nitrate,

for if reduction of silver ion is to occur, a corresponding

oxidation of the tetrazole must also take place and tetra-

zoles are generally characterized by excellent oxidative

stability.










The 5-perfluoroalkyltetrazoles are stable to both

hot aqueous acids and bases. The former is not surprising

since the tetrazoles are strong acids themselves. In

alkaline solutions, the tetrazole anion forms; since this

anion is a nucleophile itself, it is not readily attacked

by another nucleophile.

5-Perfluoropropyltetrazole was found to be stable to

both alkaline and acidic potassium permanganate at 1000.

The tetrazole ring in general is stable to oxidizing agents.

If the ring is substituted with the p-aminophenyl or the

thiol group, these substituents may be removed by oxidation

without ring rupture. Since the perfluoroalkyl group is

not easily oxidized, no reaction of 5-perfluoropropyltetra-

zole occurred with potassium permanganate, even at 1000.

The infrared spectra of the 5-perfluoroalkyltetra-

zoles are shown in Figure 4.


Reactions of the 5-Perfluoroalkyltetrazoles Below

the Thermal Ring-Opening Temperature


Synthesis of 2,5-bis(perfluoroalkyl)-l,3,4-oxadiazoles

The synthesis of 2,5-bis(perfluoroalkyl)-1,,,4-

oxadiazoles was conveniently accomplished by reaction of a

5-perfluoroalkyltetrazole with a perfluoroacyl chloride.

The reactions were carried out in sealed glass tubes since










the temperatures employed were above the boiling points of

the perfluoroacyl chlorides. In small batches, the crude

products were readily purified by separating volatile and

non-volatile components on the vacuum system. Table 1

shows the 2,5-bis(perfluoroalkyl)-l,3,4-oxadiazoles prepared

by this reaction.

TABLE 1

2,5-bis(PERFLUOROALKYL)-1,3,4-OXADIAZOLES


N=N 0 0
R -C + R' F-Cl RF-C C-R' + N2 + HC1
N- NH N-N

Nitrogen Oxadiazole
RF R Per Cent Per Cent B.P.
of Theory Yield
C F C F not determined 25 1240

C F CF 100 82 980

CF3- CF3- 100 94 650
3 3


The yield of 2,5-bis(perfluoropropyl)-l,3,4-

oxadiazole was lower than the others in Table 1. This

reaction was carried out early in this study before a good

procedure for purification of 5-perfluoropropyltetrazole

was found and optimum reaction conditions established. It

is believed that 90 per cent yields of this product would

also be obtained using pure, dry tetrazole under the re-

action conditions used in the synthesis of the other

oxadiazoles.










An attempt was made to isolate the intermediate

2-perfluorobutyryl-5-perfluoropropyltetrazole,

/N=-N
CF -C O
CN--N-C-C 37

by acylating the silver salt of the tetrazole at 0-250 in

tetrahydrofuran. Even under these mild conditions, the

acylated tetrazole was found to be unstable and rearranged

spontaneously with nitrogen elimination to give 3,5-bis(per-

fluoropropyl)-1,5,4-oxadiazole in 95 per cent yield. The

instability of the 2-perfluoroacyl-5-perfluoropropyltetra-

zoles was in sharp contrast to that of the 2-acyl-5-aryl-

tetrazoles which require a temperature of 110-1400 for

rearrangement to the oxadiazole. The latter proceeded via

the following mechanism:424


/N N -N2
Ar-C I 0 >
NN-N-C-R


Ar-C -N=N -C -R


GO
G .. .. II
Ar-C=N-N=C -R

0

Ar-C C-R
zN-N









However, since the high electronegativity of the

perfluoroacyl group markedly weakened the tetrazole ring,

the ring opening and rearrangement to the oxadiazole

probably proceeded as follows:

N==N 0 N==N ]
RF-C I + C-R'F -HC R -C 0
N NH Cl N--N -C-R'F





,- -- '. -oF
R F-C C-R' F R F-C C-R'
N-N L N-N
e

-N2
0
RF-C C-R'
SN-N

Since nitrogen leaves with its electrons from an electro-

positive carbon, its leaving may be assisted as diagrammed

above.

Synthesis of l,3-bis(5-perfluoroalkyl-1,3.,4-oxadiazolyl-2)-

perfluoropropanes
The reaction discussed above for the synthesis of

5,5-bis(perfluoroalkyl)-l,3,4-oxadiazoles was extended to
the synthesis of 1,3-bis(perfluoroalkyl-1,3,4-oxadiazolyl-2)-

perfluoropropanes. Two routes were available for this









synthesis, the use of a monotetrazole with a diacyl chloride

(A) and a difunctional tetrazole with a monoacyl chloride

(B). 1,3-bis(5-Perfluoropropyl-1,3,4-oxadiazolyl-2)per-

N=N 0 0
II II
(A) 2 RF-C N- + C1-C-(CF2)3-C-C1
N-NH




RF-C C-(CFP2) -C C-RF + 2 HC1 + 2 N2
N-N' N-N



N=N N=N 0
(B) CN-(CF ) -CI + 2 RFC
HN-N N-NH Cl

fluoropropane was prepared by methods A and B in sealed

tubes. In theory, both reactions should occur with the same

ease but in practice, method B was preferred since it

resulted both in higher yields and a more easily purified

product. There were several reasons for this. First, the

monotetrazoles were slightly hygroscopic while 1,3-bis(5-

tetrazolyl)perfluoropropane was not. Also, using method A,
the perfluoroglutaryl chloride was exposed to the atmosphere

during weighing, resulting'in some hydrolysis, while in










method B the perfluorobutyryl chloride was measured in the

vacuum system and condensed into the reaction tube with no

contact with the atmosphere. Further, since it is usually

advantageous in the reaction of a monofunctional compound

with a difunctional compound to use a slight excess of the

monofunctional material to promote higher yields, the

excess perfluoroacyl chloride employed in method B was

simply removed under reduced pressure while the excess

5-perfluoroalkyltetrazole in method A required more elabo-

rate techniques. Table 2 shows the 1,3-bis(5-perfluoro-

alkyl-l,3,4-oxadiazolyl-2)perfluoropropanes prepared by

methods A and B.

TABLE 2

1,3-bis(5-PERFLUOROALKYL-1,3,4-OXADIAZOLYL-2)-

PERFLUOROPROPANES


Metl



C37-

CF,-


aj


0 0
R -C C-(CF2) -C NC-R
N-N N-N

Nitrogen
Evolution Oxadiazole
aod Per Cent Per Cent
of Theory Yield
A 81 30

3 85 93

3 100 96


M.P.


37.0-37.8

36.2-57.5

44.1-44.8









The infrared spectra of the 1,3-bis(5-perfluoro-

alkyl-1,3,4-oxadiazolyl-2)perfluoropropanes showed the same
weak absorption at 6.35 and 6.40 1, assigned to cyclic C=N
stretching that was found in the 3,5-bis(perfluoroalkyl)-

1,3,4-oxadiazoles.18

Synthesis of 5,5'-bis(perfluoroalkyl)-2,2'-bi-1,3,4-

oxadiazoles
The synthesis of the 5,5'-bis(perfluoroalkyl)-2,2'-

bi-l,3,4-oxadiazoles was accomplished by two synthetic
routes analogous to A and B shown on page 16. The yields

N==N 0 0
II II
(C) 2 RF-C + C1-C-C-C1
N- NH
S-2 HC1
-2 N2


/\ cc\

N-N 5 N-N


-2 NaC1
0// N N -N -2 N
(D) 2 R -C + Na+ ;, c-C Na N
Cl N"'N N

from method C were much lower than D for the same reasons
given on page 16 for method A. The reaction occurred by the

same mechanism described for the synthesis of the 3,5-bis(per-
fluoroalkyl)-l,3,4-oxadiazoles described on page 15o Sodium









chloride rather than hydrogen chloride was eliminated in

method D. Table 3 shows the 5,5'-bis(perfluoroalkyl)-2,2'-

bi-1,3,4-oxadiazoles which were prepared. These compounds

showed several interesting differences from the compounds

containing isolated oxadiazole rings. The bioxadiazoles

were very high melting by comparison and showed surprising

resistance to hydrolysis by hot aqueous alkali while the

1,3-bis(perfluoroalkyl-1,3,4-oxadiazolyl-2)perfluoropro-

panes were destroyed in hot water. The infrared spectra of

the bioxadiazoles, shown in Figure 1, were also different.

The band at 6.35 .p did not appear in the spectra of the

bioxadiazoles while another considerably stronger band

appeared at 6.81 ) and was probably due to the extended

conjugated system, C=N-N=C-C=N-N=C.

TABLE 3

5,5'-bis(PERFLUOROALKYL)-2,2 -bi-1,5,4-OXADIAZOLES


O 0
R -C C-C C-R
N-N N-N
Nitrogen
Evolution Bioxadiazole
R Per Cent Per Cent
F Method of Theory of Theory
C3F7- C 90 55 163.5-
C37- D 100 75 165.0-
C2F5- D 100 93 182.0.
CF D 84 75 193.1
n -propyl alcohol solution.
In i-propyl alcohol solution.


*


i.P. max Log E
-164.50
-165.80 234.0 4.09
-182.80 232.5 4.10
-193.9 230.5 4.06


*













4000 3 0I 20I I 100 900 8
,0


/\ /o / 1
CP.-C \C--c C-CP .1
S I\\ _# \\ _// I
N-N N -N
-.2
Mull Kel F oil
-.3
.4
-.5
.6
-.7
11.C
IOCORD \ 1


80


'60


040
z

-20


0


12 13 14 15


1500


1000 900 800


3 4 5 6 7 8 9 10 11
WAVELENGTH (MICRONS)


12 13


1000 900 800
I


/\ /\
C F7-C C-C C-C3F7 1
S7 \\ // \\ // \ /
N-N N-N
Mull Kel F oil -.2

-.3

.4:
.5'
.6
.7
1.c
C .z Nr CORD
-1.5


1 i i 1 i I I I
5 6 7 8 9 10 11
WAVELENGTH (MICRONS)


12 13 14


Fig. 1.-Infrared spectra of: (A) 5,5'-bis(perfluoro-
methyl)-2,2'-bi-l,3,4-oxadiazole; (B) 5,5 -'is(perfluoro-
ethyl)-2,2'-bi-l,3,4-oxadiazole; (C) 5,5'-bisperfluoro-
propyl)-2,2'-bi-1,3,4-oxadiazole.


IUU r


I I I I I t I 1 I
3 4 5 6 7 8 9 10 11
WAVELENGTH (MICRONS)


4000 3000
4000 3000


1 n i) I , I , , , . . r


2000


'60
<
t-
I-
E40
z
< -

-20-


A


/o\ /0\
C,2?-C C--C C-
2 \\ // \\ II
N-N N-N
Mull Kel P oil


I- -_INFHACOI


C25 .1


.20
0



-.5 m
.6
.7

1.0

S15
15


4000 3000
100 i i i


2000


UJ
'60
<


040-
z

20-
-20-


3 4


I


I
4000 3000


2000


1000 900 800


1










The bioxadiazoles also showed strong ultraviolet

absorption in the 230-234 my region (see Table 5) while the

2,5-bis(perfluoroalkyl)-1,3,4-oxadiazoles showed no
18
absorption maxima in the 220-540 my region.8 The absorption

in this region is attributed to the increased conjugation.

Reaction of 5-perfluoropropyltetrazole with phosgene
Two moles of 5-perfluoropropyltetrazole were found

to react with one mole of phosgene in a closed system to

give a white solid product tentatively assigned the structure

0 C1
CF -C C-NH-N C-C F7
N-N

VIII

on the basis of it's infrared spectra (see Fig. 2B) which

showed a broad absorption at 5.1-3.4, due to N-H stretching

and a sharp band at 6.10.u due to exocyclic C=N stretching.

The cyclic C=N absorption of the oxadiazole ring appeared

as a weak band at 6.534u. The band at 6.821u was due to

N-H deformation and shifts to a higher wavelength when the

compound was deuterated. The chlorine was surprisingly

stable and did not react with water but did react with

dilute aqueous base.

Under the reaction conditions, part of VIII reacted

further with the evolved hydrogen chloride to give










I
4000 3000 2000
100 ,


1500 .


1000 900 800 700


Cl 0 C1
C F7-C.N-N-C-N-N.C-C- F7
S3 H
Mull Kel F oil










A -NFRACORD


80


'60
z
I-

r40


-20


0


12 13 14 1
12 13 14 1


3 4 5 6 7 8 9 10 11 12 13 14 15
WAVELENGTH (MICRONS)

Fig. 2.-Infrared spectra of: (A) N,N'-bis(perfluoro-
butyrimidoyl chloride)urea; (B) 2-(perfluorobutyrimidoyl
chloride)amino-5-perfluoropropyl-1,3,4-oxadiazole; and
(C) N-(perfluorobutyrimidoyl chloride)perfluorobutyramide.


S' I I I I i
3 4 5 6 7 8 9 10 11
WAVELENGTH (MICRONS)








C1 0 C1
I II I
C3F7C=N-NH-C-NH-N=C-C3F

IX

The infrared absorption of IX (see Fig. 2A) was very

similar to that of VIII. The bands due to N-H stretching

were more defined and occurred at 2.94, 3.20 and 5.41 ).

The N-H deformation band occurred at 6.69 ). The bands

assigned to N-H stretching and deformation shifted to higher

wavelengths when the compound was deuterated. The C=0 and

C=N stretching absorptions occurred at 5.70 and 6.11 u,

respectively. The experimental molecular weight and ele-

mentary analysis of IX were in good agreement with the

proposed structure. To prove that IX was derived from VIII,

a sample of VIII isolated from the reaction mixture was

heated in the presence of gaseous hydrogen chloride in a

sealed tube and thereby converted to IXo Since the re-

action of a perfluoroalkyl substituted 1,5-4-oxadiazole

with hydrogen chloride has not been reported, a sample of

2,5-bis(perfluoropropyl)-l,5,4-oxadiazole was heated at 210-

2250 in the presence of a large excess of hydrogen chloride

and produced a white solid with an infrared spectra similar

to that of IXo This product was assigned structure X

resulting from the ring opening addition of hydrogen chloride

to the oxadiazole. Only low yields of X have been obtained

which may be due to reversibility of the reaction. In a









0 01 0
C 3F-C C-C F + HCI2100 3 CF7N-NH-C-C3F7
37 ~ 3 2100 7 37
N-N
X

preliminary experiment, a low yield of X was also produced

from the reaction of N,N'-bis(perfluorobutyryl)hydrazine

with phosphorous pentachloride. The chlorine atom of X

was also not readily attacked by water at room temperature.

The infrared spectra of X (see Fig. 2C) showed bands due

to N-H stretching at 3.08 and 3.26 ) and the N-H deforma-

tion band at 6.73 The absorption due to C=0 stretch

occurred at 5.61 ) which was lower than found in the per-

fluoroalkylamides. This shift was due to the electro-

negative imidoyl chloride substituent. The C=N absorption

of X occurred at 6.01 u.

In a further experiment designed to establish the

role of hydrogen chloride in the reaction of phosgene with

5-perfluoropropyltetrazole, equivalent amounts of phosgene

and tetrazole were heated in a sealed tube in the presence

of four equivalents of hydrogen chloride. The infrared

spectra of the crude product showed it to be nearly pure

IX from which pure IX was recovered in 55 per cent yield.

No VIII was recovered.

Additional work is required on the reaction of the

perfluoroalkyltetrazoles with phosgene to determine whether

the route of the reaction is represented by E or F below.










N=N 0 N
E) R -C + C-C-Cl +
N-NH HN


C-RF -2HC/


-N

-N


N=N N=N
R -C O C-R
N- N-C-N N

-2N2
+HC1


Cl 0 Cl
RF-C =N-NH-C-NH-N = C-RF

XII


N= N
F) R -C
N -NH


0
II
+ C1-C-C1


0 Cl
HcO 1 -, ,C-NH-N=C-R
N-N
XI

N=N
t-HC R -C O0
-HC1 F NCI
N -N-C-C1


0
RF-C C-Cl
XN-N
XIII


N-N


-N2
XII <--- XI <
HC1 +HC1


/


0 C-R
R -C C-N -- N
N -N
N-N


-N2


RFCN4H
-HC1










If the 2-perfluoroalkyl-5-chloro-l,3,4-oxadiazole (XIII)

can be synthesized, the chlorine should be reactive, as

in the chlorotriazines or 2-chloropyridine for example, and

may then be reacted with the tetrazole following route F.

If successful, this reaction would be good evidence that

the reaction of phosgene with 5-perfluoropropyltetrazole

occurs by route F which is the route to be preferred since

it eliminates the need of the multicentered reaction in the

first step of E. The structure of XI may be definitely

established if the chlorooxadiazole XIII will react with

a perfluoroacyl hydrazide. The resulting hydrazide

substituted oxadiazole on reaction with phosphorous penta-

chloride in an open system should give XI.

Reaction of 5-perfluoropropyltetrazole with perfluorobutyro-

nitrile: synthesis of 3,5-bis(perfluoropropyl)-1,2,4-

triazole

The reaction of 5-perfluoropropyltetrazole with

perfluorobutyronitrile was attempted in a sealed tube at

temperatures of 100-1355 but was not successful.

In view of the pronounced ring destabilizing effect

shown by the electronegative perfluorobutyryl group, it

appeared that a strong Lewis acid might complex with the

electrons of the tetrazole ring and cause it to rupture at

relatively low temperatures. The dipolar fragment produced

could then add to the perfluorobutyronitrile.










/N=N -N2 ..
C3 F-C > [ CC3 -NNH > 3F7C=N-N-H]
S N-NH I CF CN
BF3 3F7C

H
N
CF -C C-C F
N-N


Boron trifluoride-etherate was found to produce a low

yield (11.6 per cent) of 3,5-bis(perfluoropropyl)-l,2,4-

triazole when heated at autogeneous pressure in a sealed

tube with one equivalent of 5-perfluoropropyltetrazole and

five equivalents of perfluorobutyronitrile. However, based

on the nitrogen evolution, only one tetrazole molecule out

of four which underwent ring cleavage added to the nitrile

to produce a molecule of triazole.

Gaseous boron trifluoride formed a stable 1/1 adduct

with 5-perfluoropropyltetrazole at autogeneous pressure in

a sealed tube. When heated at 1350 in a sealed tube, the

adduct broke liberating free 5-perfluoropropyltetrazole and

boron trifluoride. No nitrogen was evolved. At low

pressure, the adduct gave off boron trifluoride leaving the

5-perfluoropropyltetrazole unchanged. It is apparent that

boron trifluoride exerted no pronounced influence on the

thermal stability of 5-perfluoropropyltetrazole.










Two reactions were carried out in which one equiva-

lent of 5-perfluoropropyltetrazole was heated with two

equivalents of perfluorobutyronitrile in the presence of 10

and 1 mole per cent boron trifluoride. At 10 per cent

boron trifluoride, 3,5-bis(perfluoropropyl)-1,2,4-triazole

was obtained in 10 per cent yield while at 1 per cent boron

trifluoride a 6 per cent yield was obtained.

Anhydrous hydrogen chloride was also investigated

as a catalyst for promoting a reaction between perfluoro-

butyronitrile and 5-perfluoropropyltetrazole. After

prolonged heating at 1250 in a sealed tube with a large

excess of hydrogen chloride, the tetrazole was recovered

unchanged. However, in the presence of three equivalents

of perfluorobutyronitrile the yields of 3,5-bis(perfluoro-

propyl)-1,2,4-triazole increased with increasing amounts of

hydrogen chloride (see Table 4). The reaction probably

proceeded through the unstable perfluorobutyrimidoyl

chloride (XIV) and followed a path analogous to that

proposed for the oxadiazole formation.










TABLE 4

THE EFFECTS OF HYDROGEN CHLORIDE ON THE REACTION OF

5-PERFLUOROPROPYLTETRAZOLE WITH PERFLUOROBUTYRONITRILE


Mole Ratio
No. HC1/C F7CN
3 7


Nitrogen
Evolution
Per Cent
of Theory


3,5-bis(Per-
fluoropropyl)-
1,2,4-triazole
Per Cent of Theory


1 0.1 49.2 23.5

2 2.0 83.0 35.0

3 4.0 100.0 42.3


C F CN + HC1
37


C 3F 7CN4H'
-HCl


XIV


N=N
F0 C NH
SN- N-C-C3



(1) N2
(2) ring
closure


N
C3F"C C-CF
N-N


Increasing the hydrogen chloride concentration would be

expected to shift the equilibrium to the right in favor of

the imidoyl chloride formation. A similar intermediate has

been proposed in the HC1-catalyzed trimerization of










perfluoroacetonitrile.25 A series of 4-aryl substituted

1,2,4-triazoles was prepared from a similar reaction

employing the more stable N-aryl benzimidoyl chl-orides.16

In order to eliminate any possibility that the effect

of the hydrogen chloride is one of activating the tetrazole

(rather than the nitrile) and that it is an activated

tetrazole molecule which reacts with the carbon-nitrogen

triple bond, reaction 3 in Table 4 was repeated replacing

the perfluorobutyronitrile with an equivalent amount of

perfluorobutyne-2. No nitrogen was evolved and the tetrazole

was recovered unchanged. If the hydrogen chloride served to

some way activate the tetrazole, the tetrazole would have

reacted with the butyne as it did with the nitrile. The

fact that this reaction did not take place is further

evidence that the hydrogen chloride acts upon the nitrile

as shown.



Reactions of 5-Perfluoroalkyltetrazoles at Temperatures

Above the Thermal Ring-Opening Temperature


Reactions with perfluoroalkylnitriles: synthesis of 3,5-

bis(perfluoroalkyl)-l,2,4-triazoles

The thermolysis of 5-perfluoropropyltetrazole in the

presence of excess perfluorobutyronitrile at autogeneous

pressure in a sealed tube produced low yields (16 per cent)










of 3,5-bis(perfluoropropyl)-1,2,4-triazole. The product
resulted from the addition of the nitrile to the dipolar

nitrilimine fragment (XV) which was produced by thermolysis

of the tetrazole. A similar structure has been proposed

N= N
/N N -2 4- ** eG
C3F7-C 200-2400'> C-N=N-H -- C C=N-N-H]
N -N-NH XV
C F7CN

H
N
CF -C C-C F
37 k 3F
'N-N
for the fragment resulting from the thermolysis .of

2,5-diphenyltetrazole.

In the sealed tube reaction, the tetrazole decomposed

in the liquid phase while the nitrile was in the gas phase

and the reaction occurred only at the interface or with

the nitrile dissolved in the liquid tetrazole. In order to

increase the yield of this reaction, the tetrazole was heated

in the vapor phase in the presence of the nitrile using the

flow reactor shown in Figure 8. The yield of 3,5-bis(per-

fluoropropyl)-l,2,4-triazole was increased by this method

to 57 per cent. Table 5 shows the results of extending

this reaction to the synthesis of triazoles with other

perfluoroalkyl substituents.









TABLE 5

SYNTHESIS OF 3,5-bis(PERFLUOROALKYL)-1,2,4-TRIAZOLES

IN A FLOW REACTOR

H
N= N /N
RFC + R' CN R -C C-R'
N-NH 2 N-NF

R R'F Reaction Yield of Triazole
Temperature Per Cent of Theory
C3F7- C3F7- 270-275 57

C3F7- C2F5- 290-2950 42

CF3- C3F7- 3000 35*

CF CF3- 3000 30*
3 3
Yields were estimated using a gas-liquid chromato-
graph.

The 3,5-bis(perfluoroalkyl)-l,2,4-triazoles produced

by the reaction of nitrile and tetrazole in a flow tube are

difficult to purify, therefore this reaction cannot be

considered a good synthetic method.

The infrared spectra of the 3,5-bis(perfluoroalkyl)-

1,2,4-triazoles are shown in Figure 5o

Reactions with perfluorobutyne-2: synthesis of 3,4,5-

tris(perfluoroalkyl)pyrazoles

The gas phase thermolysis of 5-perfluoroalkyltetra-

zoles in the presence of perfluorobutyne-2, carried out in

the flow reactor (see Fig. 8) produced 3,4,5-tris(perfluoro-

alkyl)pyrazoles. The reaction was similar to that described









for the formation of the triazoles except the dipolar

nitrilimine fragment added to a carbon-carbon triple bond

rather than to the carbon-nitrogen triple bond. In contrast

to the triazoles, the pyrazoles were easily purified by

standard techniques.


R -C E--- [RF-C-N=iNH -- RF-C=N-NH]
"N--NH
N--NI CF C=CCF

CF3


RF = CF7-;CF- RF-C C-CF
N-NH

XVI

3-Perfluoropropyl-4,5-bis(perfluoromethyl)pyra-
zole was produced in 67 per cent yield from 5-perfluoro-

propyltetrazole and perfluorobutyne-2. It is a colorless
liquid boiling at 1780 and has a disagreeable odor.

3,4,5-tris(Perfluoromethyl)pyrazole was produced
from 5-perfluoromethyltetrazole and perfluorobutyne-2 in

24 per cent yield. It is a white volatile solid, m.p.

70.5-72.5 and has the same characteristic odor of the
unsymmetrical pyrazole. The infrared spectra of the 3,4,5-

tris(perfluoroalkyl)pyrazoles are in Figure 6.








34

At the initiation of this work, it was not known

if the 5-perfluoroalkyltetrazoles would undergo thermal

ring cleavage with elimination of the 2,3- or the 3,4-

nitrogen atoms. The fragment from either mode of cleavage

could add to a nitrile to produce a triazole but the

addition product with perfluorobutyne-2 would result in an

imidazole (XVII) from 2,3-nitrogen elimination and a

pyrazole (XVI) from 3,4-elimination as shown below.











CF3

/c%
N-NH


XVI


CF CGCCF
3 3


SFCN


NI'l -N 131
N= N
R-C I
N-NH
2,35


elimination


elimination


H
N

R -N '
N--N


CF C2CCF
3 3 H
H
/N
RF-C C-CF

SNIC-CF3

XVII


H
R /N

N


H
NO

N










N.M.R. spectra were obtained on the 5-perfluoro-

alkyltetrazole-perfluorobutyne-2 reaction products and as

seen from Table 6 definitely establish the pyrazole structure.

TABLE 6

N.M.R. SPECTRAL DATA OF 3,4,5-tris(PERFLUOROALKYL)-

PYRAZOLES

CF (a)
C
CF -CF -CF -C C-CF (b)
3 2 2\ / 3
N-NH
(c) (e) (d)

Group p.p.m.* No. of Peaks Splitting Area
(c.p.s.)

a -21.1 6 73. 1.0
b -14.9 4 7.2 1.0
c + 4.1 5 9.1 1o0
d +32.3 5 (or 7) approx. 0.7
10
e +48.3 4 approx. 0.7
7.6

H single broad peak, 7 +4.97 p.p.m.


CF (a)

(b) /C (b)
CF -C C-CF
S\\ / 3
N-NH


a -20.0 7 6.9 1.00
b -14.7 broad 2.08

Ref. CF COOH
5










The proton nuclear magnetic resonance spectrum of

3-perfluoropropyl-4,5-bis(perfluoromethyl)pyrazole was

obtained and indicated that the hydrogen was on a nitrogen.

The equivalency of the 3 and 5 methyl'groups in 3,4,5-

tris(perfluoromethyl)pyrazole established the mobility of

the H atom, e.g.,

CF CF
C /C
CF -C C-CF CF -C C-CF3
N-N N-N
H H

The definite establishment of the pyrazole structure con-

clusively demonstrated that the 5-perfluoroalkyltetrazole

rings opened with elimination of the 3,4 nitrogen atoms to

give the nitrilimine XV.

Reactions of the perfluorobutyronitrilimine fragment under

mild conditions

In order to further establish the existence of the

perfluorobutyronitrilimine fragment in the vapor phase

thermolysis reactions of 5-perfluoropropyltetrazole, the

tetrazole was distilled at low pressure into a short,

tubular furnace and the thermolysis products condensed on

a liquid nitrogen cooled cold finger (see Fig. 9). The

material with which the nitrilimine fragment was to react

was also condensed on the cold finger. The reactions took









place at an undetermined temperature by allowing the cold

finger to warm slowly to room temperature.

3,5-bis(Perfluoropropyl)-1,2,4-triazole resulted

from the reaction of perfluorobutyronitrilimine with

perfluorobutyronitrile and 3-perfluoropropyl-5-perfluoro-

methyl-1,2,4-triazole from reaction with perfluoroaceto-

nitrile. The reaction of perfluorobutyne-2 with perfluoro-

butyronitrilimine produced 3-perfluoropropyl-4,5-bis(per-

fluoromethyl)pyrazole and reaction of the fragment with

water gave perfluorobutyrhydrazide. The formation of the

heterocycles from this reaction has already been discussed.

The hydrazide is produced as follows:
OH 0
E I 1
C3F7C=N-NH + H20 -- C3F7C=N-NH2 = C F7C-NH-NH2


From this series of experiments it was concluded

that perfluorobutyronitrilimine is the reactive species

in triazole and pyrazole synthesis described above. Further,

perfluorobutyronitrilimine has a finite lifetime and is

sufficiently reactive to attack perfluoroalkylnitriles,

perfluorobutyne-2 or water at or below room temperature.

Thermolysis of 5-perfluoropropyltetrazole with no acceptor

molecule present

The literature contains few reports of the decompo-

sition, due to purely thermal effects, of 5-aryl-substituted

tetrazoles and none of 5-alkyl tetrazoles. Pinner26










observed the formation of 3,5-diphenyl-l,2,4-triazole and

3,6-diphenyl-l,2,4,5-tetrazine when 5-phenyltetrazole was

heated above its melting point. In a more thorough study,

Huisgen observed, in addition, the formation of 2,4,6-

triphenyl-s-triazine, 4-amino-3,5-diphenyl-1,2,4-triazole

and 3,6-diphenyl-l,2-dihydro-l,2,4,5-tetrazine. These

thermolyses were carried out at about 2000 and the products

indicated that the tetrazole ring opened both by loss of

hydrazoic acid to produce benzonitrile and by loss of the

5,4 nitrogen atoms to produce the phenylnitrilimine fragment.

When 5-perfluoropropyltetrazole was passed at low

pressure through a hot tube heated to about 4000, and the

thermolysis products quickly removed from the heated zone,

the reaction followed a course different from that reported

for the decomposition of the 5-aryltetrazoles at atmospheric

pressure. No formation of 3,5-bis(perfluoropropyl)-1,2,4-

triazole was observed and the lack of red color in the crude

products indicated the absence of 3,6-bis(perfluoropropyl)-

1,2,4,5-tetrazine.a
The infrared spectra of the crude gaseous products

showed the presence of only trace amounts of perfluoro-

butyronitrile. A group of bands occurred at 4.85, 8.02 and


a3,6-bis(Perfluoroalkyl)-1,2,4,5-tetrazines,
presently under investigation by other workers in this
laboratory, have an intense red color.









9.62-9.67 ) doublett). These bands may be due to tetra-

fluoroallene27 but due to the small amount present, this

component was not separated. Another band in the infrared

spectra of the crude products occurred at 4.70 1. This

band was also due to a minor component which was not

separated, but it is interesting to note that it is in the

diazoalkane region and may be due to 1-diazo-l-H-perfluoro-

butane.

The major product of the thermolysis was identified

as 2-H-perfluorobutene-1, and was obtained in 45 per cent

yield. The N.M.R. spectral data shown in Table 7 were used

to determine the structure of the olefin.


TABLE 7

N.M.R. SPECTRAL DATA OF 2-H-PERFLUOROBUTENE-1

CF -CF2-CH=CF2

Group p.p.m.* Area

F .-4.3 0.85
F -2.4 0.91
CF3 +11.6 2.72
CF2 +36.5 1.82

H Doublet, tripled as expected; '= 5.18

Reference CF COOH
3










The infrared spectra (see Fig. 3A) of the olefin showed a

strong band at 5.66yp due to C=C stretching.

A small amount of yellow 3,6-bis(perfluoropropyl)-

1,2-dihydro-l,2,4,5-tetrazine was recovered from the non-

volatile products of the thermolysis. The dihydrotetrazine

was identified by its infrared (see Fig. 3B) and ultra-

violet spectra which are very similar to 3,6-bis(l-H-per-

fluoroethyl)-l,2-dihydro-1,2,4,5-tetrazine prepared by

Carboni and Lindsey.28 5,6-bis(Perfluoropropyl)-l,2-

dihydro-l,2,4,5-tetrazine was easily oxidized to 3,6-bis(per-

fluoropropyl)-1,2,4,5-tetrazine by nitric acid.

The reaction probably proceeds as follows:
H H
/N--N
N--N
/ N
C3F7C \C-C3F

/N=N -N2 O
C3 70 390-395 c 3F7C3N-NH]
N NH XV 1-3 hydride
shift

C3F7CHN2



-N2
H
CF -CF2-CHECF2 CF -CF2-CF2-C:
carbene insertion
between andB
carbons












































Fig. 5.-Infrared spectra of: (A) 2-H-perfluoro-
butene-1; and (B) 3,6-bis(perfluoropropyl)-1,2-dihydro-
1,2,4,5-tetrazine.










The thermolysis of 5-perfluoropropyltetrazole

initially forms the nitrilimine fragment, XV, as shown by

the addition reactions of this product with perfluoroalkyl

nitriles and perfluorobutyne-2. This behavior is analogous

to the reported first step of the decomposition of 5-phenyl-

tetrazole. The observed difference in products must occur

after the nitrilimine is formed and there appear to be two

major factors involved. First, 5-phenyltetrazole was

pyrolyzed at atmospheric pressure while 5-perfluoropropyl-

tetrazole was pyrolyzed at low pressures. Therefore, the

phenylnitrilimine fragments, once formed, were in closer

proximity and had the greater chance of combining to give

the observed sym-tetrazines and N-amino-3,5-diphenyl-l,2,4-

triazole (from rearrangement of 3,5-diphenyl-l,2-dihydro-

tetrazine under the influence of heat).

The second difference may be seen from a considera-

tion of the relative stabilities of the phenyl and

perfluorobutyronitrilimine fragments. Phenylnitrilimine

can be stabilized by the following 14 resonance forms:








S10
Cc=N=iNH


CF-C=N-N H C-N=NH CC C=N =NH C=N=NH













-C FN-C=N-NH
C =phe ni -NH Cth l=Nee NHe


t

etc

Perfluorobutyronitrilimine can have only 5 canonical forms
as follows:


C 3F 7C=N-NH C3 F 7C-N=NH 0- 3 CF 7-C=N=NH -, C3 F 7-CN-NH

XV XVI I
C 3F 7-C=N-NH


Thus the phenyl group contributes to the stabilization of
phenylnitrilimine while the electronegative perfluoropropyl
group serves to activate perfluorobutyronitrilimine through
XV. Therefore, in perfluorobutyronitrilimine a hydride ion
can then easily transfer to the carbon to give 1-diazo-l-H-
perfluorobutane which will eliminate nitrogen at the reaction









temperature to give the carbene. Alternatively, a concerted

1-3 hydrogen shift with simultaneous nitrogen elimination

may produce the carbene directly from XVI. The carbene then

inserts between the o< andR carbon atoms to give the

observed product, 2-H-perfluorobutene-1. Similar o< -,

insertion reactions have been reported in the decomposition
29
of a number of diazoalkanes.2


Infrared Spectra and Acid Strengths of the Perfluoro-

alkyl Substituted Pyrazoles, Triazoles and Tetrazoles


Infrared spectra

The perfluoroalkyl substituted pyrazoles, triazoles

and tetrazoles are acids and can exist in the tautomeric forms

zN N Z N --N
N=N N-N
R -CI RF-C I
N-NH N- N
H
H
N N N
R -C C--R'F R -C C-R R F-C C-R'
N-N N-N N-N
H H
R" R"
IF I F
C c
RF-C ^C-R' R -C C-R'
N-N N-N
H H










The N-H stretching absorption of these three classes

of perfluoroalkyl substituted heterocycles were similar.

From Figures 4, 5, and 6 it is evident that the absorption

bands were broad and fell in the 3.0-3.5 p region generally

associated with hydrogenic stretching of associated

carboxylic acids.30 The broad absorption of the perfluoro-

alkyl-substituted heterocycles was attributed mainly to

intermolecular association and partly to absorption of the

various tautomeric forms.

1,3-bis(5-Tetrazolyl)perfluoropropane had a hydrogenic

stretching absorption (see Fig. 4C) which was broader and

had more peaks than that of the monotetrazoles (Fig. 4A and

4B). This was probably due to the contribution of intra-

molecular hydrogen bonding which does not occur with the

other heterocycles. A model of 1,3-bis(5-tetrazolyl)per-

fluoropropane indicates that the rings can be on top of one

another with a hydrogen atom between them. Interesting

information supporting this assumption was found in following

the exchange of hydrogen for deuterium in 1,3-bis(5-tetra-

zolyl)perfluoropropane-d2 when it was exposed to the

atmosphere. The deuterated compound appeared to exchange

one deuterium for hydrogen within 12 hours at about 50 per

cent relative humidity. After that time, no further

appreciable exchange was noted even after several days. It

appeared that one deuterium was tightly held in a cage

















































Fig. 4.-Infrared spectra of: (A) 5-perfluoromethyl-
tetrazole; (B) 5-perfluoropropyltetrazole; and (C)
1,3-bis(5-tetrazolyl)perfluoropropane.










I CM". L
4000 3000 2000 1500 1000 900 800 700
100 I 0.0
H


N-N







3 4 5 6 7 8 9 10 11 12 13 14 15.4
WAVELENGTH (MICRONS)









100 --"----------------------------------- 0.0
40 -.7





S60- /\.". KIV ot] -.5m
< 2 0
I 0 1 I I I .-
3 A 5 6 7 8 9 10 11 12 13 14 15
WAVELENGTH (MICRONS)

4000 3000 2000 1500 1000 900 800 700
100 0.0
H
rPO8 o -.15
N-N
'60 Mull Kel P oil .2o
60.0
.3"


.6
20 .7
1.0
0 INFPACOPD..- 1.5
3 4 5 6 7 8 9 10 11 12 13 14 15
WAVELENGTH (MICRONS)

4000 3000 2000 1500 1000 900 800 700
1001'I"'""1'. I I 0.0



80 C 37 7-C ,-CP3 .
N-N
p u o e Mull Kel F oil
SI.6

240 -.7



1.0
S1NFPACOPD 1r'1.5
0-1 ---7-------T-- .1 1----------- I-- -
3 4 5 6 7 8 9 10 11 12 13 14 15
WAVELENGTH (MICRONS)

Fig. 5.-Infrared spectra of: (A) 3,5-bis(perfluoro-
propyl)-1,2,4-triazole; (B) 3-perfluoropropyl-5-perfluoro-
ethyl-1,2,4-triazole; and (C) 3-perfluoropropyl-5-
perfluoromethyl-1,2,4-triazole.



















1000 900 800


Fig. 6.-Infrared spectra of: (A) 3-perfluoropropyl-
4,5-bis(perfluoromethyl)pyrazole; and (B) 3,4,5-
tris perfluoromethyl)pyrazole.


1
4000 3000
10 I I - -


2000









between the two tetrazole rings and could not exchange

readily. Deuterated forms of the other heterocycles showed

rapid exchange of hydrogen for deuterium. The replacement

was noticeable spectroscopically after a few minutes

exposure to the atmosphere.

The cyclic C=N absorption of 5-perfluoropropyl-

tetrazole (see Fig. 4B) appeared as a weak shoulder at

6.62 The N-H deformation absorption occurred at 6.75 .

A band at 7.09 ) in sodium 5-perfluoromethyltetrazole has

been attributed to N=N absorption.31 A similar band in

5-perfluoropropyltetrazole occurred at 7.17 1 and was

probably due to the N=N group. In 5-perfluoropropyltetra-

zole-d the bands at 6.62 and 7.17y retained their positions

and relative intensities while the 6.75 band shifted into

the C-F region, 7.8-9.0 ). The silver salt showed a cyclic

C=N absorption at 6.75.,.

Free 5-perfluoromethyltetrazole showed a relatively

intense band at 6.59, (see Fig. 4A). This has been
31
attributed to a C-CF3 stretching frequency. The band

at 7.16 u was due to N=N absorption and the N-H deformation

band was at 6.75 u. The deformation frequency appeared at

7.73 in 5-perfluoromethyltetrazole-d. No band assignable
to cyclic C=N stretching absorption occurred but in light

of its position and intensity in 5-perfluoropropyltetrazole,

it may be obscured by the C-CF3 absorption.







51

The infrared spectra of 1,5-bis(5-tetrazolyl)per-

fluoropropane was somewhat different from the 5-perfluoro-

alkyltetrazoles. Six distinct peaks were observed in the

N-H stretching region while the monotetrazoles had only

three. 1,3-bis(5-Tetrazolyl)perfluoropropane appeared to

have two bands due to N-H deformation at 6.89 and 7.09 p.

These disappeared on deuteration while a shoulder at 7.15,

remained as a weak peak. Consistent with the previous

assignments, the shoulder at 7.15 l was assigned to N=N

absorption. 1,3-bis(5-Tetrazolyl)perfluoropropane-d2

showed two new bands at 6.59 and 7.70 p. The latter is

due to N-D deformation and the former, surprisingly,

appeared in the cyclic C=N region. No bands in the 6.5-

6.6u1 region appeared in the spectra of 1,5-bis(5-tetra-

zolyl)perfluoropropane, even in concentrated mulls. The

broad weak absorption at 5.75,- was due to the so-called

immonium element.19 Since it did not occur in the mono-

tetrazoles, it probably results from intramolecular

association.

The infrared spectra of the 3,5-bis(perfluoro-

alkyl)-1,2,4-triazoles (see Fig. 5) have already been

discussed19 and only a few additional comments are in order.

The infrared spectra of the deuterated triazoles indicated

the bands in the 7.15-7.15 1 region were due to N-H









deformation while the bands in the 6.86-6.90 u region were

due to the cyclic C=N element. As pointed out by Brownl9

this was an unusually long wavelength for C=N stretching.

The spectra of 3-perfluoropropyl-5-perfluoromethyl-1,2,4-

triazole (see Fig. 5C) shows a band at 6.60u not found in

the other triazoles. This, as in 5-perfluoromethyltetra-

zole, was assigned to C-CF3 stretching.

The infrared spectra of 3-perfluoropropyl-4,5-

bis(perfluoromethyl)pyrazole and 3,4,5-tris(perfluoro-

methyl)pyrazole (see Fig. 6) were similar and will be dis-

cussed together. Bands at 6.61 and 6.58,u respectively

were in the region previously assigned to C-CF3 stretching

and were probably due to the C-CF3 stretch in the pyrazoles.

The N-H deformation absorptions occurred at 7.03 and 6.99,'

respectively and the remaining two bands, 6.31 and 6.72,u

in 3-perfluoropropyl-4,5-bis(perfluoromethyl)pyrazole, and

6.24 and 6.78 in 3,4,5-tris(perfluoromethyl)pyrazole were

due to the ring unsaturation. These bands arose from C=C

and C=N stretching and in view of the positions of C=N

absorption noted in the other acidic azoles discussed above,

it appeared reasonable that the bands at the longer wave

lengths arose from C=N and those at the shorter wave length

from C=C stretching. However, in 3,5-dimethylpyrazole

bands at 6.05 and 6.44 y were assigned to C=C stretching










and a band at 6.28 u was assigned to C=N stretching.32

Acid strengths

The insertion of sp hybridized nitrogen atoms into

the pyrrole ring shows a progressive acid strengthening

effect (see Table 8). This effect has been attributed to

both the inductive attraction of electron density from the Tr

cloud by the nitrogen and its resonance stabilization of

the anion formed by loss of the acidic hydrogen.3 The

I I I

-c /C- - -C C- -C C-
/ _H / \
N-NH N-N N-N


effect of each sp nitrogen is approximately equal to that

of a conjugated nitro group.

The replacement of hydrogen bound to a carbon by a

perfluoroalkyl group produced a similar and only slightly

less pronounced effect on the acid strength. From Table 8,

it is seen that introduction of an Osnitrogen decreases

the pKa by 4-5 units and the average effect per perfluoro-

alkyl group is in the 3-5.5 pKa units range. This

illustrates the extreme electronegativity of the perfluoro-

alkyl group and the transmission of its effect to the










TABLE 8

pKa VALUES OF ACIDIC AZOLES


pKa of
Parent
Compound*


pKa of
Perfluoro-
alkyl
Substituted
Azole


Average Change
in pKa Per
Perfluoroalkyl
Group


Pyrazole 14 5.1 3

1,2,4-
Triazole 10.1 2.7-3.1** 3.5

Tetrazole 4.9 1.7*** 3.2

Albert, ref. 23, p. 143.
**
Brown and Cheng, ref. 19.
Norris, ref. 2, reports the pKa of 5-perfluoromethyl-
tetrazole is 1.14. This value was obtained by a method which
corrects for the high dissociation of strong acids.



nitrogen acids in a manner and degree similar to that

occurring in the carboxylic acids.a













aH. L. Henne and C. J. Fox, J. Am. Chem. Soc., 72,
2323(1951) give pKa for CF COOH = 0.23. That of acetic acid
is 4.73. The difference is 4.5 pKa units.


Azole
Rinn












CHAPTER III


EXPERIMENTAL


Source of Materials


Perfluoroacetic, perfluoropropionic and perfluoro-

butyric acids were purchased from Columbia Organic

Chemicals Company, Columbia, South Carolina. Perfluoro-

glutaronitrile and perfluoroglutaryl chloride were supplied

by Hooker Chemical Company, Niagara Falls, New York.

Perfluorobutyne-2 was purchased from Peninsular ChemResearch,

Gainesville, Florida. The sodium azide (Eastman purified

grade) was used as received.

Perfluoroacetonitrile, perfluoropropionitrile and

perfluorobutyronitrile were prepared by dehydration of the

respective perfluoroalkylamides as described by Swarts34
35
and Gilman and Jones.

The Fisher anhydrous reagent grade acetonitrile used

was further dried by refluxing it at least three hours over

granular barium oxide and distilling it out of the barium

oxide mixture. The apparatus was vented through a dry

ice-cooled trap to protect against contamination of the

acetonitrile by atmospheric moisture.










Method of Molecular Weight and pKa Determination

of the 5-Perfluoroalkyltetrazoles and 3,4,5-

tris(Perfluoroalkyl)pyrazoles


The molecular weights of the tetrazoles and pyrazoles

were determined by titrating a weighed sample of the compound

with OlO000 N NaOH. The samples were dissolved in methanol-

water solution with the exception of 5-perfluoromethyltetra-

zole and 1,3-bis(5-tetrazolyl)perfluoropropane, which are

water soluble and were titrated in water solution. During

the titration, the pH was determined with a Beckman Model G

pH meter.

The pKa values of the compounds were read from the

pH meter at the half-neutralized point where pH equals pKa.


Preparation of 5-Perfluoroalkyltetrazoles


5-Perfluoromethyltetrazole

5-Perfluoromethyltetrazole was prepared by treating

perfluoroacetonitrile with sodium azide in anhydrous

acetonitrile as described by Norris.2 The reaction was slow

using acetonitrile dried with barium oxide but was

effectively catalyzed by a trace of glacial acetic acid.

The tetrazole was obtained in 54 per cent yield, b.p. 82-830

(5.1 mm.). Molecular weight calculated for C2HF3N4: 138.

Found by titration against 0.1000 N NaOH: 138; pKa, 1.70.










5-Perfluoropropyltetrazole

A one liter 5-neck flask was equipped with a gas inlet

tube, magnetic stirrer and dry ice-cooled reflux condenser.

The flask was charged with sodium azide, 53.75 g. (0.52

mole), dry methyl cellosolve, 120 ml. and glacial acetic

acid, 150 ml. in that order. The flask was immersed in an

ice bath and the sodium azide slurry was stirred vigorously.

Perfluorobutyronitrile, 116 g. (0.595 mole), was added slowly

beneath the surface of the liquid by means of the gas inlet

tube at a rate adjusted to give a slow reflux of perfluoro-

butyronitrile. When addition of the nitrile was complete,

the flask was heated at 700 for one hour. After cooling,

the reaction mixture was poured slowly into 200 ml. of 20 per

cent hydrochloric acid and stirred. The tetrazole separated

as a slightly viscous oil. The aqueous layer was removed and

the heavier oil layer was washed ten times with dilute

hydrochloric acid, then once with water and dried under

reduced pressure to give 157 g. of crude 5-perfluoropropyl-

tetrazole.

The crude tetrazole was dissolved in 1500 ml. of

acetone and added with stirring to a solution of 98 g. of

silver nitrate in 1500 ml. of distilled water. The voluminous

silver 5-perfluoropropyltetrazole formed rapidly and was

removed by filtration, washed with water and acetone and dried

at reduced pressure. This procedure removed a carbonyl-

containing contaminant which cannot be removed by distillation.










Free 5-perfluoropropyltetrazole was regenerated from

the silver salt by addition of gaseous hydrogen chloride to

a slurry of the salt in one liter of anhydrous diethyl ether.

(Complete conversion of silver 5-perfluoropropyltetrazole to

the free tetrazole and silver chloride is easily detected by

a rapid settling of the precipitated silver chloride, when

stirring is discontinued, to give a clear solution of

tetrazole in ether. The silver tetrazole settles slowly and

any cloudiness in the solution is indicative of incomplete

conversion to the free tetrazole.)

The ether solution of 5-perfluoropropyltetrazole was

filtered to remove the silver chloride, concentrated under

reduced pressure, then transferred to a distillation flask.

A few grams of P205 were added and the tetrazole was

distilled from P205, yield 96.2 g., 77.6 per cent, b.p. 600

(0.02 mm.), m.p. 31-350.

Anal. Calcd. for C4HF7N : C, 20.17; H, 0.41;

F, 55.89; N, 23.53; mol. wt., 238. Found: C, 20.42;

H, 0.55; F, 56.00; N, 23.75; mol. wt., 241 (by titration

against 0.1000 N NaOH); pKa, 1.73.

The synthesis described above was carried out before

Norris's2 paper on the synthesis of 5-perfluoromethyl-

tetrazole appeared. Since Norris's method is the more

convenient of the two methods, a slightly modified version

was used to reprepare 5-perfluoropropyltetrazole.











Sodium azide, 30.0 g. (0.462 mole), was placed in a

3-neck flask equipped with a magnetic stirrer, a dry ice-

cooled reflux condenser, a dry nitrogen sweep to exclude

air and a gas entry tube. Anhydrous acetonitrile, 300 ml.,

and glacial acetic acid, 2 ml., were added. Perfluoro-

butyronitrile, 95.0 g. (0.487 mole), was admitted under the

surface of the liquid. The reaction was exothermic and the

perfluorobutyronitrile addition rate was adjusted to maintain

a reaction temperature of 55-600. Toward the end of the

addition, heat was applied to maintain the reaction

temperature. Heating was continued for thirty minutes after

the nitrile addition was complete. The solution was filtered

while hot, concentrated to 250 ml. by boiling under slightly

reduced pressure and then cooled. The first crop of white,

deliquescent needles of sodium 5-perfluoropropyltetrazole

was removed by filtration. Further concentration of the

acetonitrile solution produced a second crop of crystals.

The salt was dried under reduced pressure at 600 for 24

hours and at 850 for 24 hours. The total yield of sodium

5-perfluoropropyltetrazole was 105.5 g., 86.5 per cent.

The free tetrazole was generated by slurrying the

sodium salt in anhydrous ether and bubbling in gaseous

hydrogen chloride. The sodium chloride formed was removed by

filtration and the ether solution was concentrated under

reduced pressure. The 5-perfluoropropyltetrazole was










distilled at 87-880 (3.5 mm.); yield, 83.0 g., 75.4 per cent.

This product did not solidify and was apparently not as pure

as the tetrazole purified by means of the silver salt.

1,3-bis(5-Tetrazolyl)perfluoropropane

Sodium azide, 8.68 g. (0.133 mole), glacial acetic

acid, 7.95 ml., and methylcellosolve, 30 ml., were placed in

a 100 ml. flask equipped with a dry ice-cooled reflux

condenser and nitrogen sweep. The contents of the flask

were frozen; the flask was pumped free of air and perfluoro-

glutaronitrile, 10.0 g. (0.0495 mole), was condensed in

through the vacuum system. The reactants were warmed to 00

and stirred for thirty minutes. The mixture was then heated

at 70 for fifteen minutes, refluxed for fifteen minutes,

and cooled to room temperature. Acetone, 25 ml., and

concentrated hydrochloric acid, 8 ml., were added to the

reaction mixture and the solids formed were removed by

filtration. An additional 30 ml. of acetone and 7 ml. of

concentrated hydrochloric acid were added and the mixture

was refiltered. The solvent was removed under reduced

pressure and the crude product was redissolved in acetone

and filtered. Most of the solvent was again removed under

reduced pressure and the solid 1,3-bis(5-tetrazolyl)per-

fluoropropane was precipitated by the addition of 60 ml. of

benzene. The yield of crude product was 11.58 g. (81 per










cent), m.p. 156-157.50. The product was crystallized twice

from perfluorobutyric acid to give a pure, white solid,

m.po 159.8-160.50. 1,3-bis(5-Tetrazolyl)perfluoropropane

decomposed slowly at 1700.

Anal. Calcd. for C5H2F Ng: C, 20.83; H, 0.67;

F, 39.58; N, 38.88; mol. wt., 288. Found: C, 21.03;

H, o084; F, 36.69; N, 39.28; mol. wt., 287 (by titration

against 0.1000 N NaOH); pKa, 1.70.


Preparation of the Disodium Salt of Bitetrazole


The procedure for the synthesis of the disodium salt

of bitetrazole was reported by Friederich36 in a patent and

is given here since the patent may not be available.

Sodium cyanide, 50 g. (1.04 mole), sodium azide,

65 g. (1.00 mole), and distilled water, 600 ml., were added

to a two liter 3-neck flask fitted with a water-cooled

reflux condenser, a stirrer, a 500 ml. addition funnel and

a thermometer. The mixture was stirred vigorously and 55 g.

of manganese dioxide was added with continued stirring. A

solution consisting of sulfuric acid, 100 g., glacial acetic

acid, 80 g., and distilled water, 200 g., was prepared and

added to the flask over a one hour period during which the

temperature was maintained at 10-200.

The mixture was slowly heated to 1000 over four











hours and maintained at 1000 for three hours and then cooled

to room temperature. The black manganese bitetrazole was

removed by filtration and washed with water. The bitetra-

zole was stirred with 56 g. of Na2CO3 dissolved in 150 ml.

of H20. Sufficient water was added to make a thin paste and

the mixture was filtered hot through a fine filter to remove

manganese carbonate. The aqueous solution was cooled to 0

for 24 hours and the first crop of hydrated disodium

bitetrazole was removed by filtration. Three additional

crops were obtained by further concentration and filtration.

The four crops were combined and crystallized three

times from water. Dehydration was effected at 800 under

reduced pressure to yield 51.7 g. (56.8 per cent) of

disodium bitetrazole.


Hydrolytic Stability of 5-Perfluoropropyltetrazole


5-Perfluoropropyltetrazole, 0.50 go (0.0021 mole),

and 6 ml. 1 N NaOH solution (0.0065 eq.) were placed in a

10 ml. flask equipped with a reflux condenser. A gas buret

was connected to the reflux condenser. The solution was

refluxed five hours. No gas was evolved. 5-Perfluoropropyl-

tetrazole, 0.42 g. (84 per cent), was recovered by acidifi-

cation with concentrated hydrochloric acid.

The procedure was repeated using 6 ml. of 1 N










hydrochloric acid and 0.50 g. of 5-perfluoropropyltetrazole.

No gas was evolved. The recovered tetrazole weighed 0.47 g.

(94 per cent).


Oxidative Stability of 5-Perfluoropropyltetrazole


Approximately 0.2 ml. of 5-perfluoropropyltetrazole

was dissolved in 1 N sodium hydroxide and a few drops of

4 per cent potassium permanganate were added. No dis-

coloration of the permanganate occurred at room temperature

nor within ten minutes at 1000.

Approximately 0.2 mlo of 5-perfluoropropyltetrazole

was shaken with a potassium permanganate solution acidified

with nitric acid. No reaction occurred at room temperature

nor after 30 minutes at 1000.


Synthesis of 2,5-bis(Perfluoroalkyl)-l,3,4-oxadiazoles

by Reaction of 5-Perfluoroalkyltetrazoles with Per-

fluoroacyl Chlorides


Synthesis of 2,5-bis(perfluoropropyl)-l,3,4-oxadiazole

5-Perfluoropropyltetrazole, 0.51 g. (0.0021 mole),

was weighed in a previously constricted heavy-wall glass

tube. The tube was connected to the vacuum system and

pumped free of air. Perfluorobutyryl chloride, 0.50 g.










(0.0022 mole), was condensed into the tube. The tube was

sealed and heated at 900 for six hours and at 1100 for 11.5

hours. The tube was cooled in liquid nitrogen and opened

to the vacuum system; nitrogen, 0.00105 mole (50% of

theory), was recovered. The nitrogen was removed and the

reaction tube was resealed and heated 21 hours at 1250. The

tube was reopened and the volatile portion of the reaction

mixture was transferred to the vacuum system. A rough

separation in the system gave 0.55 g. (24.9 per cent) of

nearly pure 2,5-bis(perfluoropropyl)-1,3,4-oxadiazole.

Synthesis of 2-perfluoropropyl-5-perfluoromethyl-1,3,4-

oxadiazole

5-Perfluoropropyltetrazole, 2.00 g. (0.0084 mole),

was weighed in a previously constricted heavy-wall glass

tube. The tube was connected to the vacuum system and

pumped free of air. Perfluoroacetyl chloride, 1.36 g.

(0.0103 mole), was condensed into the tube; the tube was

sealed and heated ten hours at 1500 and 50 minutes at 150.

The tube was cooled in liquid nitrogen and opened to the

vacuum system. Nitrogen, 0.24 g. (0.0084 mole), was

re...vered.

The volatile oxadiazole was transferred through the

vacuum system to another tube to separate it from the non-

volatile products. Hydrogen chloride, produced as a

by-product, and residual perfluoroacetyl chloride were










removed to the vacuum system, leaving 2.03 g. (87 per cent

yield) of 2-perfluoropropyl-5-perfluoromethyl-1,3,4-

oxadiazole, b.p. 980. The infrared spectra of the oxadiazole

showed the weak, characteristic oxadiazole bands at 6.30

and 6.38 u.

Synthesis of 2,5-bis(perfluoromethyl)-l,3,4-oxadiazole

5-Perfluoromethyltetrazole, 1.47 g. (0.0107 mole),

was weighed in a previously constricted heavy-wall glass

tube. The tube was connected to the vacuum system, pumped

free of air, and perfluoroacetyl chloride, 1.69 g. (0.0128

mole), was condensed in the tube. The tube was sealed and

heated at 950 for 30 minutes and at 1050 for 30 minutes.

The temperature was then raised to 2000 and maintained for

30 minutes. The tube was cooled in liquid nitrogen and

opened to the vacuum system. Nitrogen, 0.0107 mole, was

recovered.

The hydrogen chloride and residual perfluoroacetyl

chloride were removed through the vacuum system. A small

amount of a nonvolatile product was separated by transferring

the oxadiazole to another tube by means of the vacuum system.

The yield of 2,5-bis(perfluoromethyl)-1,3,4-oxadiazole was

2.07 go (95 per cent), b.p. 650. The reported b.p. is 650.










Synthesis of 1,3-bis(5-Perfluoroalkyl-l,3,4-

oxadiazolyl-2)perfluoropropanes


Synthesis of 1,3-bis(5-perfluoropropyl-l,3,4-oxadiazolyl-2)

perfluoropropane

From 5-perfluoropropyltetrazole and perfluoroglutaryl

chloride. 5-Perfluoropropyltetrazole, 3.06 g. (0.0129 mole),

and perfluoroglutaryl chloride, 1.78 g. (0.00643 mole), were

weighed in a previously constricted heavy-wall glass tube.

The tube was sealed and heated for 18 hours at 125-130,

then cooled in liquid nitrogen and opened to the vacuum

system. Nitrogen, 0.292 g. (0.0104 mole, 81% of theory),

was recovered.

The crude liquid product was washed with a dilute

potassium carbonate solution to dissolve any residual

tetrazole or acid chloride. The desired product precipi-

tated as a white solid and was rapidly filtered, washed

with water and dried at reduced pressure. Sublimation at

reduced pressure produced lol6 g. (30% of theory) of 1,3-

bis(5-perfluoropropyl-l,3,4-oxadiazolyl-2)perfluoropropane,

m.p. 37.0-37.8.
From 1,3-bis(5-tetrazoyl)perfluoropropane and

perfluorobutyryl chloride. 1,3-bis(5-Tetrazoyl)perfluoro-

propane, 5.00 g. (0.0173 mole), was weighed in a previously

constricted heavy-wall glass tube. The tube was connected










to the vacuum system and pumped free of air. Perfluorobutyryl

chloride, 8.08 g. (005345 mole), and 25 ml. of anhydrous

methylene chloride were condensed into the tube.. The tube

was sealed and heated at 1250 for 6.5 hours with continuous

agitation, then cooled in liquid nitrogen and opened to the

vacuum system. Nitrogen, 0.818 g. (0.0292 mole, 85% of

theory), was recovered. The methylene chloride was removed

under reduced pressure, leaving 10.06 g. (95.4 per cent

yield) of pale yellow 1,3-bis(5-perfluoropropyl-l,3,4-

oxadiazolyl-2)perfluoropropane, m.p. after crystallization

from hexane, 56.2-37.5.
Anal. Calcd. for C13F20N402: C, 25.02; F, 60.85;

N, 8.97. Found: C, 25.50; F, 60.73; N, 8.95.

Synthesis of 1,3-bis(5-perfluoromethyl-l,3,4-oxadiazolyl-2)-

perfluoropropane
1,3-bis(5-Tetrazolyl)perfluoropropane, 5.00 g.

(0.0175 mole), was weighed in a heavy-wall glass tube. The
tube was attached to the vacuum system and held under

reduced pressure for 2.5 hours to remove traces of moisture.

Anhydrous methylene chloride, 25 ml., and trifluoroacetyl

chloride, 5.80 g. (0.044 mole), were condensed into the tube

through the vacuum system. The reaction tube was sealed and

heated with agitation at 125-1355 for ten hours, then cooled

in liquid nitrogen and opened to the vacuum system. Nitrogen

evolution was found to be quantitative.









All of the volatile materials were removed from the

'tube under reduced pressure to leave 7.42 g. of white solid

contaminated with a trace of yellow liquid. The solid

product was sublimed under reduced pressure three times to

give pure, white 1,3-bis(5-perfluoromethyl-l,3,4-

oxadiazolyl-2)perfluoropropane melting at 44.1-44.80,


Synthesis of 5,5'-bis(Perfluoroalkyl)-2,2'-

bi-1,l,4-oxadiazoles


Synthesis of 5,5'-bis(perfluororopoyl)-2,2'-bi-1,3,4-

oxadiazole

From 5-perfluoropropyltetrazole and oxalyl chloride.

5-Perfluoropropyltetrazole, 2.50 g. (0.0105 mole), was

weighed in a heavy-wall glass tube. The tube was attached

to the vacuum system, pumped free of air, and 2 ml. of dry

methylene chloride and 0.667 g. (0.0053 mole) of oxalyl

chloride were condensed in through the vacuum system. The

tube was sealed and heated at 1000 with agitation for 5.5

hours and then cooled in liquid nitrogen and opened to the

vacuum system. Nitrogen evolution was found to be 90 per

cent of the theoretical amount. The volatile materials

were removed from the tube under reduced pressure and the

remaining solid product crystallized twice from toluene and

sublimed under reduced pressure to give 1.37 g. (55 per cent









yield) of 5,5'-bis(perfluoropropyl)-2,2'-bi-l,5,4-

oxadiazole melting at 163.5-164.50.

From disodium bitetrazole and perfluorobutyryl

chloride. Dry disodium bitetrazole, 3.30 g. (0.0181 mole),

was weighed into a previously constricted heavy-wall glass

tube, and the tube was pumped free of air. Methylene

chloride, 15 ml., and perfluorobutyryl chloride, 11.1 g.

(0.0478 mole), were condensed in the tube which was sealed

and heated at 950 for four hours and at 1250 for three

hours with constant agitation. On opening the tube,

approximately the theoretical amount of nitrogen was found

to have formed. The crude product was recrystallized twice

from toluene. Sublimation at reduced pressure gave 6.51 g.

(75.4 per cent yield) of 5,5'-bis(perfluoropropyl)-2,2'-
bi-l,3,4-oxadiazole, m.p. 165.0-165.80.

Anal. Calcd. for C10F14N 02: C, 25.33; F, 56.12;

N, 11.80; 0, 6.75. Found: C, 25.55; F, 56.00; N, 11.70.

Synthesis of 5,5'-bis(perfluoroethyl)-2,2'-bi-1,3,4-

oxadiazole

Dry disodium bitetrazole, 5.00 g. (0.0275 mole), was

weighed into a previously constricted heavy-wall tube. The

tube was attached to the vacuum system, pumped free of air

and 25 ml. of dry methylene chloride and 12.0 g. (0.066

mole) of perfluoropropionyl chloride were condensed in the










tube. The tube was sealed, heated 15 hours at 1000 and

four hours at 1250 with agitation and then cooled in liquid

nitrogen. On opening the tube to the vacuum system,

nitrogen was found to have evolved quantitatively. The

reaction mixture was concentrated by removing part of the

solvent under reduced pressure, then washed with water to

dissolve the sodium chloride; the crude product was

separated by filtration and washed with water. A crude

yield of 9.62 g. (93.4 per cent) of 5,5'-bis(perfluoro-

ethyl)-2,2'-bi-1,3,4-oxadiazole, m.p. 182-184.80 was

obtained. The product was crystallized twice from toluene,

m.p. 182-182.80.

Synthesis of 5,5'-bis(perfluoromethyl)-2,2'-bi-1,3,4-

oxadiazole

Dry disodium bitetrazole, 6.00 g. (0.033 mole), was

weighed into a previously constricted heavy-wall glass tube.

The tube was attached to the vacuum system, pumped free of

air, and 25 ml. of methylene chloride and 91l g. (0.069

mole) of perfluoroacetyl chloride were condensed in the tube.

The tube was sealed and heated at 100-1050 for 17 hours and

at 1280 for three hours with constant agitation. A yield of

84 per cent of the theoretical amount of nitrogen was

obtained on opening the tube. The reaction mixture was

mixed with water, the solid product removed by filtration










and washed three times with water to remove the sodium

chloride. A crude yield of 6.81 go (75.4 per cent) of 5,5'-

bis(perfluoromethyl)-2,2'-bi-l,3,4-oxadiazole was obtained.

After two crystallizations from toluene, the product

melted at 193.3-194.6.


Attempted Isolation of 1(2)-Perfluorobutyryl-5-

perfluoropropyltetrazole


The apparatus shown in Figure 7 was constructed for

this reaction. The N-acetylated tetrazole, if formed, was

expected to be easily hydrolyzed; therefore, this apparatus

was designed to allow the reaction to be carried out under

strictly anhydrous conditions, including filtration of the

reaction mixture and solvent removal from the final product,

without exposure of the product to the atmosphere.

Preparation of silver 5-perfluoropropyltetrazole.

5-Perfluoropropyltetrazole, 10.0 g. (0.041 mole), was dis-

solved in 20 ml. of acetonitrile and 7.15 g. (0.0384 mole)

of silver nitrate was added to the solution. The mixture

was warmed with stirring to bring it into solution. On

cooling, silver 5-perfluoropropyltetrazole precipitated and

was removed by filtration. The filtrate was concentrated and

a second crop of the salt was removed. The total yield of

the silver salt was 13.2 g. or 91 per cent, based on the

amount of tetrazole used.























stopcock


receiving tube


S 24/40







medium sintered glass
filter


stopcock
S 24/40




reaction tube




.magnetic stirring bar


Fig.7.-Apparatus for attempted isolation of 1(2)-
perfluorobutyryl-5-perfluoropropyltetrazoleo











Acylation of silver 5-perfluoropropyltetrazole with

perfluorobutyryl chloride. Silver 5-perfluoropropyltetra-

zole, 2.50 g. (0.0072 mole), was placed in the reaction

tube of the apparatus shown in Figure 7. The apparatus

was attached to the vacuum system and the salt thoroughly

dried under reduced pressure. The tube was then cooled

with liquid nitrogen and 10 ml. of tetrahydrofuran, freshly

distilled from barium oxide, was condensed into the tube

after passing through a phosphorous pentoxide drying tube.

The solvent was degassed and 3.34 g. (0.0143 mole) of per-

fluorobutyryl chloride (100 per cent excess) was condensed

into the tube. The mixture was allowed to warm slowly to

room temperature with continuous stirring. The apparatus

was then inverted and the pressure in the receiving tube

slowly reduced to promote filtration of the product from

the precipitated silver chloride. Nitrogen was generated

in the reaction indicating the acylated tetrazole rearranged

to 2,5-bis(perfluoropropyl)-l,3,4-oxadiazole under the

reaction conditions.

Solvent and volatile products, 15.31 g., were removed

from the receiving tube through the vacuum system and

analyzed by gas-liquid chromatography using a 1/4 inch

column two meters in length containing a silicone oil liquid

phase on a diactomaceous support (Perkin-Elmer Column C).

Helium, at 0.14 ml. per second, was used as a carrier gas










with a column temperature of 1170. 2,5-bis(Perfluoropropyl)-

1,3,4-oxadiazole was retained 6.13 minutes and the tetra-

hydrofuran 10.88 minutes after injection. The determination

of the yield 2,5-bis(perfluoropropyl)-l,3,4-oxadiazole was

based upon the integrated areas of a chromatogram of a known

solution of 2,5-bis(perfluoropropyl)-l,3,4-oxadiazole in

tetrahydrofuran produced under the same conditions. This

method indicated a 2.78 g. or 94.5 per cent yield of 2,5-

bis(perfluoropropyl)-l,3,4-oxadiazole.


Reaction of 5-Perfluoropropyltetrazole with Phosgene

Reaction with no added hydrogen chloride. 5-Per-

fluoropropyltetrazole, 5.00 g. (0.021 mole), was weighed in

a previously constricted heavy-wall tube. The tube was

attached to the vacuum system, pumped free of air, and

10 ml. of dry methylene chloride and 1.04 g. (0.0105 mole)

of phosgene were condensed in the tube. The tube was sealed

and heated seven hours at 1350, cooled in liquid nitrogen and

opened to the vacuum system. Nitrogen, 0.0197 mole (93.4%

of theory), was recovered. The tube contents were allowed

to warm to room temperature and the solvent and volatile

products were removed at reduced pressure, leaving the crude

solid products in the tube.

The crude solids were dissolved in anhydrous ether










and separated in an alumina column with ether as an eluent.

Crude N,N'-bis(perfluorobutyrimidoyl chloride)urea, 0.85 g.,

m.p. 98-1030, was the first product eluted. A white solid,

probably 2-(perfluorobutyrimidoyl chloride)amino-5-

perfluoropropyl-l,5,4-oxadiazole) 1.49 g., m.p. 86-880,

was obtained after continued elution with ether and subse-

quently with methanol. The crude N,N'-bis(perfluoro-

butyrimidoyl chloride)urea was further purified by passage

through a 22 inch alumina packed column with ether as

solvent and eluent. After evaporation of the ether, the

product was sublimed twice at reduced pressure to give

0.51 g. of pure N,N'-bis(perfluorobutyrimidoyl chloride)urea,

m.p. 105-1070.
Anal. Calcd. for C9H2C12F14N40: C, 20.81; H, 0.39;

Cl, 13.68; F, 51.25; N, 10.79; 0, 3.08; mol. wt., 519o

Found: C, 21.18; H, 0.87; Cl, 13.85; F, 53.74; N, 10.95;

mol. wt., 527 (by freezing point depression in dioxane).

The product assumed to be 2-(perfluorobutyrimidoyl

chloride)amino-5-perfluoropropyl-l,3,4-oxadiazole, 0.2 g.,

was placed in a heavy-wall tube, air in the tube was removed

through the vacuum system and 0.5 g. anhydrous hydrogen

chloride admitted with the tube cooled by liquid nitrogen.

The tube was sealed and heated at 1400 for seven hours, then

opened to the vacuum system. The remaining hydrogen chloride

was removed at reduced pressure leaving a light colored crude










product in the tube. Without removing the tube from the

vacuum system, the crude product was heated and caused to

sublime to the upper portion of the tube. The infrared

spectra of the sublimate showed it to be nearly pure N,N'-

bis(perfluorobutyrimidoyl chloride)urea. The yield was not

determined.

Reaction in the presence of added hydrogen chloride.

5-Perfluoropropyltetrazole, 1.00 g. (.0042 mole), phosgene,

0.208 g. (0.0021 mole), hydrogen chloride, 0.307 g. (0.0084

mole) and 1.5 ml. of anhydrous methylene chloride were

sealed in a heavy-wall tube and heated ten hours at 950

and seven hours at 1600. The tube was cooled in liquid

nitrogen and opened to the vacuum system and 0.0032 mole

of nitrogen was recovered. The tube was warmed to room

temperature and the solvent and residual hydrogen chloride

removed under reduced pressure. The infrared spectra of

the crude product, a pale amber solid mixed with a small

amount of oil, showed it to be almost pure N,N'-bis(per-

fluorobutyrimidoyl chloride)urea. It was purified by

eluting with ether from an alumina column to give 0.601 g.

(55.2 per cent yield) of pure N,N'-bis(perfluorobutyrimidoyl
chloride)urea, m.p. 105-106.50.

Reaction of 2,5-bis(perfluoropropyl)-l,3,4-oxadiazole

with hydrogen chloride. 3,5-bis(Perfluoropropyl)-1,3,4-

oxadiazole, 0.50 g. was weighed in a heavy-wall tube, the









tube connected to the vacuum system and pumped free of air.

Hydrogen chloride, 0.75 g. was condensed in the tube and

the tube sealed and heated at 2100 for 24 hours and 2250

for 20 hours. The tube was cooled in liquid nitrogen and

opened to the vacuum system. The residual hydrogen

chloride and other volatile products were removed under

reduced pressure leaving 0.03 g. of a white solid product,

presumably N-(perfluorobutyrimidoyl chloride)perfluoro-

butyramide in the tube, m.p. 110-112. The product was

found to be stable to water at room temperature but is

attacked by aqueous base.


Reaction of 5-Perfluoropropyltetrazole with Perfluoro-

butyronitrile Under Mild Conditions: Synthesis of 5,5-

bis(Perfluoropropyl)-1,2,4-triazole


Attempted uncatalyzed reaction of 5-perfluoropropyltetrazole

with perfluorobutyronitrile

5-Perfluoropropyltetrazole, 1.32 g. (0.00555 mole),
was weighed in a previously constricted heavy-wall glass

tube which was then attached to the vacuum system and pumped

free of air. After 5 ml. of dry methylene chloride and

1.08 g. (0.00555 mole) of perfluorobutyronitrile were

added through the vacuum system, the tube was sealed and

heated 15.5 hours at 1000. When the reaction tube was










cooled in liquid nitrogen and opened to the vacuum system,

no nitrogen was found and the tetrazole and nitrile were

recovered unreacted.

Boron trifluoride etherate catalyzed reaction of 5-perfluoro-

propyltetrazole with perfluorobutyronitrile

5-Perfluoropropyltetrazole, 0.66 g. (0.0028 mole),

and 0.60 ml. of boron trifluoride etherate were placed in

a previously constricted heavy-wall glass tube. The tube

was attached to the vacuum system, cooled in liquid nitrogen,

pumped free of air and perfluorobutyronitrile, 2.71 g.

(0.0159 mole), was condensed into the tube. The tube was

sealed and heated five hours at 650, two hours at 100,

two hours at 1500 and then cooled in liquid nitrogen and

opened to the vacuum system. Nitrogen, 0.0545 g. (0.00123

mole), or 44 per cent of the theoretical amount was

recovered. The tube was warmed to room temperature and the

remaining boron trifluoride-ether complex was pumped out.

The crude product, 1.12 g., was analyzed by gas-liquid

chromatography using the Perkin-Elmer Column "C" at 1460

with helium as a carrier gas flowing at 1.16 ml. per second.

3,5-bis(Perfluoropropyl)-1,2,4-triazole in the product was

identified by comparison of the chromatogram to one produced

by a known sample of the triazole under identical conditions.

The retention time of the triazole was 4.05 minutes. The










yield, estimated from the integrated area under the curves

of the chromatogram, was 0.13 g. or 11.6 per cent.

Gaseous boron trifluoride catalyzed reaction of 5-per-

fluoropropyltetrazole with perfluorobutyronitrile

Formation of the boron trifluoride-tetrazole adduct.

5-Perfluoropropyltetrazole, 0.50 g. (0.0021 mole), was

weighed in a doubly constricted, heavy-wall glass tube.

The tube was connected to the vacuum system and pumped free

of air. Boron trifluoride gas, 0.143 g. (0.0021 mole), was

measured in the vacuum system at 155 mm. The tube containing

the tetrazole was opened to the vacuum system containing the

boron trifluoride. No reaction or solution of the boron

trifluoride in the tetrazole was apparent within 15 minutes.

The boron trifluoride was then condensed into the tube and

the tube was sealed. On warming the tube to room tempera-

ture, a white, solid, boron trifluoride-tetrazole adduct

slowly formed. The tube was heated six hours at 600. Some

melting, but no other apparent reaction had occurred,

therefore, heating was continued for 2.5 hours at 90. The

tube was then cooled in liquid nitrogen and opened to the

vacuum system; no nitrogen had evolved.

The tube was resealed and heated two hours at 1350

and then rapidly cooled to liquid nitrogen temperature.

Considerable gas condensed on the tube wall. When warmed

to room temperature, the product was liquid but slowly











reacted to re-form the original solid adduct. The tube was

re-cooled in liquid nitrogen and opened to the vacuum system.

No nitrogen was found. When warmed to room temperature at

low pressure, the adduct slowly liberated boron trifluoride,

leaving the relatively nonvolatile 5-perfluoropropyl-

tetrazole unchanged.

Reaction of 5-perfluoropropyltetrazole with per-

fluorobutyronitrile in the presence of 10 mole per cent boron

trifluoride. 5-Perfluoropropyltetrazole, 0.50 g. (0.0021

mole), was weighed in a heavy-wall glass tube; the tube was

connected to the vacuum system and pumped free of air. Boron

trifluoride, 0.0143 g. (0.00021 mole), and perfluorobutyro-

nitrile, 0.84 g. (0.0042 mole), were condensed in the tube

and the tube was sealed and heated six hours at 1000. The

tube was opened to the vacuum system and the residual

nitrile pumped off. The crude product was dissolved in

ether and extracted with 1 N sodium hydroxide solution to

separate the triazole product and remaining tetrazole from

any nonacidic products. The alkaline solution was acidified

with concentrated hydrochloric acid to give 0.35 g. of a

colorless oil which was analyzed by gas-liquid chromatography

using the Perkin-Elmer Column "C" at 1220.with helium as a

carrier gas flowing at 1.36 ml. per second. The major

components of the oil were identified as 3,5-bis(perfluoro-

propyl)-l,2,4-triazole, 23.7 per cent, and 5-perfluoro-

propyltetrazole, 63.0 per cent, by comparison of the









chromatogram to one produced by authentic samples under the

same conditions. The retention time of the triazole was

9.84 min., and the tetrazole 32.80 min. after air. The

yield of 3,5-bis(perfluoropropyl)-l,2,4-triazole was

calculated to be 9.8 per cent.

Reaction of 5-perfluoropropyltetrazole with per-

fluorobutyronitrile in the presence of one mole per cent

boron trifluoride. The procedure described above was

repeated using 2.00 g. (0.0084 mole) of 5-perfluoropropyl-

tetrazole, 3.28 g. (0.0168 mole) of perfluorobutyronitrile

and 0.00562 g. (8.4 x 10-5 moles) of boron trifluoride.

The reaction tube was heated 11 hours at 1000, then

opened to the vacuum system and 25 per cent of the

theoretical amount of nitrogen was recovered. The crude

product, 2.12 g. analyzed as described above, contained a

6.2 per cent yield of 3,5-bis(perfluoropropyl)-l,2,4-

triazole.

Reaction of 5-perfluoropropyltetrazole with perfluoro-

butyronitrile in the presence of gaseous hydrogen chloride

Heating 5-perfluoropropyltetrazole with excess

hydrogen chloride. 5-Perfluoropropyltetrazole, 0.50 g.

(0.0021 mole), was weighed in a constricted heavy-wall glass

tube, the tube was connected to the vacuum system and pumped

free of air. Anhydrous hydrogen chloride, 1.89 g. (0.0517

mole), was condensed in the tube and the tube was sealed










and heated 14 hours at 1250. The tube was cooled in liquid

nitrogen and opened to the vacuum system. No nitrogen was

liberated. The tube was warmed to room temperature and

pumped free of hydrogen chloride. The residual liquid was

identified as 5-perfluoropropyltetrazole by its infrared

spectrum.

Reaction of 5-perfluoropropyltetrazole with per-

fluorobutyronitrile in the presence of gaseous hydrogen

chloride. Three experiments in which 5-perfluoropropyl-

tetrazole reacted with perfluorobutyronitrile in the

presence of gaseous hydrogen chloride were performed under

identical conditions with the exception of the amount of

hydrogen chloride present. Three constricted heavy-wall

glass tubes of approximately 50 ml. capacity were prepared

and 0.50 g. (0.0021 mole) of 5-perfluoropropyltetrazole

was placed in each tube. Each tube was then connected to

the vacuum system, pumped free of air and 1.28 g. (0.0065

mole) of perfluorobutyronitrile was added. The following

amounts of gaseous hydrogen chloride were placed in the

tubes: tube 1, 0.024 g. (0.00065 mole); tube 2, 0.48 g.

(0.0126 mole); tube 5, 0.96 g. (.0252 mole). The tubes

were sealed and heated 11 hours at 1550 with constant agita-

tion. The tubes were then immersed in liquid nitrogen and

opened to the vacuum system. The following amounts of free

nitrogen were recovered from the tubes (yields werebased on











the evolution of one mole of nitrogen per mole of tetrazole):

tube 1, 0.0289 g. (0.00105 mole or 49.2 per cent); tube 2,

0.0491 g. (0.00175 mole or 85.4 per cent); tube 5, 0.0595 g.

(0.00212 mole or 100 per cent).

The tubes were pumped free of hydrogen chloride

leaving the crude products-mixtures of white solids and

colorless liquids-in the tubes. The solid portion of the

crude product, mainly 3,5-bis(perfluoropropyl)-l,2,4-

triazole, was separated from the liquid by quickly mixing

the crude product with methylene chloride and filtering

immediately. The residue was dissolved in ether, filtered

and the ether allowed to evaporate. From tubes 2 and 3

were obtained 0.18 g. and 0.30 g. respectively of almost

pure 3,5-bis(perfluoropropyl)-l,2,4-triazole; no triazole

was separated with this procedure from tube 1.

The filtrates from the above filtrations were concen-

trated and analyzed by gas-liquid chromatography and

estimated to contain 0.20, O.11, and 0.055 g. of 3,5-bis(per-

fluoropropyl)-l,2,4-triazole from tubes 1, 2, and 3,

respectively. The total yields were: tube 1, 0.20 g. or

25.5 per cent; tube 2, 0.29 g. or 35.0 per cent; tube 3,

0.36 g. or 42.2 per cent.










Attempted Reaction of 5-Perfluoropropyltetrazole

with Perfluorobutyne-2 in the Presence of Gaseous

Hydrogen Chloride


5-Perfluoropropyltetrazole, 0.50 g. (0.0021 mole),

was weighed in a heavy-wall glass tube, the tube was

attached to the vacuum system and pumped free of air.

Perfluorobutyne-2, 1.02 g. (0.0065 mole), and hydrogen

chloride, 0.96 g. (0.0252 mole), were condensed into the

tube. The tube was sealed, heated for 11 hours at 1550,

then cooled in liquid nitrogen and opened to the vacuum

system. No nitrogen was found in the tube and the tetrazole

was recovered after pumping off the hydrogen chloride and

perfluorobutyne-2.


Reactions of 5-Perfluoroalkyltetrazoles Under

Thermal Ring Opening Conditions


Synthesis of 5,5-bis(perfluoropropyl)-l,2,4-triazole in a

static system

5-Perfluoropropyltetrazole, 1.00 g. (0.0042 mole),

was weighed in a previously constricted heavy-wall glass

tube, the tube attached to the vacuum system and pumped

free of air. Perfluorobutyronitrile, 1.64 g. (0.0042 mole),

was condensed into the tube and the tube was sealed and

heated at 2000 for one hour and 2400 for 1.5 hours. The










tube was then cooled in liquid nitrogen, opened to the

vacuum system and 0.147 g. (0.0053 mole) of nitrogen was

recovered. The tube was warmed to room temperature and

the volatile materials removed under reduced pressure. The

dark viscous residue in the tube was dissolved in diethyl

ether and removed from the tube. After removal of the

solvent, the oil was heated and 0.28 g. of yellow crystals

sublimed from it. After resubliming at 80 (75 mm.), the

slightly yellow product melted at 85-970 and was identified

as 3,5-bis(perfluoropropyl)-1,2,4-triazole (crude yield

16.5 per cent). The crude product was further purified by

dissolving it in aqueous potassium carbonate and extracting

the solution with diethyl ether. The alkaline solution was

then acidified with hydrochloric acid and the precipitated

triazole removed by filtration. After resubliming at

atmospheric pressure, the melting point was 103-105. The

reported melting point is 110-1110.19

Synthesis of 3,5-bis(perfluoroalkyl)-l,2,4-triazoles in a

flow reactor

The following series of reactions between 5-perfluoro-

alkyltetrazoles and perfluoroalkylnitriles was carried out

in the flow reactor shown in Figure 8. The nitrogen was

dried by passage through a copper helix immersed in a dry

ice-acetone bath.














0
"* 0'd
Soo
H 0 -


'do-


H -


rl4

O .4


0 D 0 O 4



Un D 0 +


r-'



b0
(-




*H
U)
r-l






00)

H,
fl


.86


U)
-4o











































O
C.)





0


O
i-I
0








co













4-
.4
r0)





rcl

0




CH
?rx










I-
0n
*p

ri


1-i




0 )

0 .H

r-4 -+
,-.4


kcuj
'dZ










Synthesis of 3,5-bis(perfluoropropyl)-l,2,4-triazole.

5-Perfluoropropyltetrazole, 0.45 g., was placed in the

vaporization pot and the pot attached to the inlet of the

pyrolysis apparatus. The furnace was preheated to 2650 and

the hot tube flushed simultaneously with perfluorobutyro-

nitrile and nitrogen to insure complete removal of air and

the presence of the nitrile in the hot tube when the

tetrazole entered the hot zone.

The tetrazole reservoir was then heated to 1580 to

begin vaporization of the tetrazole and the nitrogen and

perfluorobutyronitrile flows were adjusted to 0.04 liter

per minute and 4.0 g. per hour, respectively. After 20

minutes the furnace temperature was raised from 2650 to

2710 to affect more complete decomposition of the tetrazole.

At the same time, the reservoir temperature was increased to

1650 to increase the vaporization rate of the tetrazole and

the nitrile flow rate was increased to 6.5 g. per hour.

After 70 minutes of operation of the apparatus, the tetrazole

reservoir temperature was raised to 1800 and the furnace

temperature maintained at 273-2750. The tetrazole had

completely vaporized and the reaction was completed after

110 minutes. The nitrile flow and furnace temperature were

maintained for an additional 15 minutes and the nitrogen

sweep was maintained for 20 minutes.










The product was removed from the exit tube of the

pyrolysis apparatus with diethyl ether and the ether

removed by evaporation. Sublimation of the product under

reduced pressure produced 0.476 g. of a white solid, melting

at 68-810, contaminated with a small amount of liquid. An

additional 0l10 g. of solid product was recovered from the

dry ice trap to give a total crude yield of 0.486 g. A

portion of the crude product was further purified by

absorbing the liquid impurity on filter paper, then sub-

liming the dried solid at reduced pressure, to give a product

melting at 109-1100 whose infrared spectra was identical to

that of a known sample of 3,5-bis(perfluoropropyl)-l,2,4-

triazole. Assuming the crude product was at least 90 per

cent pure, a yield of 0.437 g. or 57 per cent was realized.

Synthesis of 3-perfluoroDropyl-5-perfluoroethyl-

1,2,4-triazole. Using the procedure described above,

0.49 g. of 5-perfluoropropyltetrazole was vaporized into

the hot tube in approximately one hour. The furnace was

maintained at 291-2960, the tetrazole reservoir at 152-1570

and the perfluoropropionitrile flow rate was 5.5 g. per

hour. The solid product was washed from the exit tube with

ethyl ether and the ether evaporated to give a crude yield

of 0.39 g., m.p. 75-800.










The crude product was sublimed twice at reduced

pressure to give 0.31 g. (42 per cent yield) of pure

3-perfluoropropyl-5-perfluoroethyl-l,2,4-triazole, m.p.

79.8-81.0. The melting point and infrared spectra of this

triazole were identical with those of an authentic sample

produced from 2-perfluoropropyl-5-perfluoroethyl-l,3,4-

oxadiazole using the procedure described by Brown and

Cheng.19

Synthesis of 3-oerfluoropropyl-5-perfluoromethyl-

1,2,4-triazole. Using the procedure described for 3,5-

bis(perfluoropropyl)-1,2,4-triazole, 0.59 g. of 5-perfluoro-

methyltetrazole was vaporized into the hot tube in 65

minutes. The furnace was maintained at 3000, the tetrazole

reservoir at 1490 and the perfluorobutyronitrile flow rate

was 7.5 g. per hour. The amber-colored, liquid product was

washed from the exit tube with ethyl ether and the ether

evaporated to give a crude yield of 0.55 g. No suitable

method of purification of the crude product was found,

therefore the product was identified and the yield estimated

by gas-liquid chromatographic analysis. The retention times

of 3-perfluoropropyl-5-perfluoromethyl-l,2,4-triazole was

12.32 minutes using the Perkin-Elmer Column "C" at 1550 with

helium as a carrier gas flowing at 0.19 ml. per second and

was established with an authentic sample prepared from

2-perfluoropropyl-5-perfluoromethyl-1,3,4-oxadiazole using









the procedure described by Brown and Cheng.19 From this

data, a yield of 0.25 g. or 35 per cent was calculated. The

pure triazole melts at 63-650 and is hygroscopic.

Synthesis of 5,5-bis(perfluoromethyl)-1,2,4-triazole.

Using the procedure described for 3,5-bis(perfluoropropyl)-

1,2,4-triazole, 1.00 g. of 5-perfluoromethyltetrazole was

vaporized into the hot tube in 67 minutes. The furnace was

maintained at 2990, the tetrazole reservoir at 150 and the

perfluoroacetonitrile flow rate was 6.5 g. per hour. The

amber-colored, liquid product was washed from the exit line

with ethyl ether and the ether evaporated to give a crude

yield of 0.78 g. No suitable method of purification of the

crude product was found, therefore the product was identified

and the yield estimated by gas-liquid chromatographic

analysis. The retention time of 3,5-bis(perfluoromethyl)-

1,2,4-triazole was 15.83 minutes using the Perkin-Elmer

Column "C" at 140 with helium as a carrier gas flowing at

0.20 ml. per second and was established with an authentic

sample prepared from 3,5-bis(perfluoromethyl)-1,3,4-
19
oxadiazole. From this data, a yield of 0.45 g. or 30 per

cent was calculated.

Synthesis of 3,4,5-tris(perfluoroalkyl)pyrazoles by reaction

of perfluorobutyne-2 with 5-perfluoroalkyltetrazoles

The flow reactor diagrammed in Figure 8, was used in

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ECJ3RF1RK_9COCJI INGEST_TIME 2017-07-19T20:28:28Z PACKAGE UF00097956_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

SOME REACTIONS OF THE 5-PERFLUOROALKYLTETRAZOLES By ROBERT JAMES KASSAL A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITi' OF FLORIDA December, 1963

PAGE 3

ACKNOWLEDGnENTS The author wishes to extend his thanks to his research director, Dr. H. C. Brown, for his guidance throughout this investigation. He also wishes to thank the other members of his committee. Dr. G. B. Butler, Dr. V. M. Jones, Dr. J. D. Vinefordner and Dr. V. V. Wilmot, for their aid and encouragement. The author is indebted to Dr. W. S. Brey for N.M.H. measurements and their interpretation which were used to identify several new compounds. The success of this work was in no small way due to the patience and encoiiragement of the author's wife, Barbara, and his children, Christopher, Kenneth and Cynthia. ii

PAGE 4

TABLE OP CONTENTS Page ACKNOWLEDGMENTS ii LIST OF TABLES vii LIST OF FIGURES viii PRELIMINAEI REMARKS ix Chapter lo INTRODUCTION 1 Statement of the Problem 7 II. DISCUSSION 9 Preparation and General Properties of 5-Perfluoroalkyltetrazoles 9 Reactions of the 5-Perfluoroalkyltetrazoles Below the Thermal Ring-Opening Temperature 12 Synthesis of 2,5-bis(perfluoroalkyl)1,3,^-oxadiazoles 12 Synthesis of 1 ,3bis (3-perfluoroalkyll,3>^-oxadiazolyl-2)perfluoropropanes. 15 Synthesis of 5,5'-bis(perfluoroalkyl)2,2'-bi-l,3,^-oxadiazoles 18 Reaction of 5-perfluoropropyltetrazole with phosgene 21 Reaction of 5-perfluoropropyltetrazole with perfluorobutyronitrile: synthesis of 3 , 5-bis(perf luoropr opyl ) -1 , 2 , ^triazole 26 iii

PAGE 5

Page Hoactions of 5-Perf.l-;oroalkylte-t;ra:3oles at T'2:.'P2raturo3 Above the Theriaal Rl-ngOpening Temperature 50 Reactions v;ith perfluoroalkylnitriles : synthesis of 3»5-his(perfluoroalkyl)1,2,^-triazoles 30 Reactions with perfluorobutyne-2: synthesis of 3 , ^ , 3tris (perf luoroalkyl)pyrazoles 32 Reactions of the perfluorobutyronitrilimine fragment under mild conditions 37 Thermolysis of 5-perfluoropropyltetrazole with no acceptor molecule present. . . 38 Infrared Spectra and Acid Strengths of the Perfluoroalkyl Substituted Pyrazoles, Triazoles and Tetrazoles ^5 Infrared spectra 45 Acid strengths 53 III. EXPERIMENTAL 55 Source of Materials . 55 Method of Molecular Weight and pKa Determination of the 5-Perfluoroalkyltetrazoles and 3 1 4 1 5tri s (Perf luoroalkyl )pyrazoles . 56 Preparation of 5-Perfluoroalkyltetrazoles. . 56 5-Perfluoromethyltetrazole 56 5-Perfluoropropyltetrazole 57 1 , 3bi fi ( 5-Tetrazolyl )perf luoropropane . . 60 Preparation of the Disodium Salt of Bit( trazole 61 Hydrolytic Stability of 5-Pei'fluoropropyltetrazole 62 iv

PAGE 6

Page Oxidative Stability of 5-Perfluoropropyltetrazole 63 Synthesis of 2,5-bis(Perfluoroalkyl)-l,3,^oxadiazoles by Reaction of 5-Perfluoroalkyltetrazoles with Perfluoroacyl Chlorides 63 Synthesis of 2,5-bis(perfluoropropyl)1 , 3 ,^-oxadiazole 63 Synthesis of 2-perfluoropropyl-5-perfluoroiiiethyl-l,5,^-oxadiazole 6^ Synthesis of 2,5-bis(perfluoromethyl)1 , 5 »^-oxadiazole 65 Synthesis of l,5-bis(5-Perfluoroalkyl-l,3,^oxadiazolyl-2)perfluoropropanes 66 Synthesis of 1 ,3bis (5-perfluoropropyl1 , 3 j^-oxadiazolyl-2)perf luoropropane . . 66 Synthesis of 1 , 3his ( 5-perf luoromethyll,3,^-oxadiazolyl-2)perfluoropropane . . 67 Synthesis of 5,5'-bis(Perfluoroalkyl)-2,2'-bi1,3,^-oxadiazoles 58 Synthesis of 5)5'-bis(perfluoropropyl)2,2*-bi-l,3>^-oxadiazole 68 Synthesis of 3 > 5 ' bis (perf luoroethyl )2,2*-bi-l,3,^-oxadiazole 69 Synthesis of 5> 3 ' bis (perf luoromethyl)2,2'-bi-l,3,^-oxadiazole 70 Attempted Isolation of l(2)-Perfluorobutyryl5-perfluoropropyltetrazole . 71 Reaction of 5-Perfluoropropyltetrazole with Phosgene 7^ Reaction of 5-Perfluoropropyltetrazole with Perfluorobutyronitrile Under Mild Conditions: Synthesis of 5»5-his(Perfluoropropyl)-l,2,4-triazole 77 V

PAGE 7

Page Attempted uncatalyzed reaction of 5perfluoropropyltetrazole with perfluorobutyronitrile 77 Boron trifluoride etherate catalyzed reaction of 5-perfluoropropyltetrazole with, perfluorobutyronitrile. . . 78 Gaseous boron trifluoride catalyzed reaction of 5-perfluoropropyltetrazole with perfluorobutyronitrile. . . 79 Reaction of 'p-'peTflxiovo^TOpjltetrazole with perfluorobutyronitrile in the presence of gaseous hydrogen chloride 81 Attempted Reaction of 5-Perfluoropropyltetrazole with Perfluorobutyne-2 in the Presence of Gaseous Hydrogen Chloride. . 84Reactions of 5-Perfluoroalkyltetrazoles Under Thermal Ring Opening Conditions. . 8^ Synthesis of 5,5-bis(perfluoropropyl)1,2,^-triazole in a static system . . 8^ Synthesis of 3,5-bis(perfluoroalkyl)^1,2,-4— triazoles in a flow reactor . . 85 Synthesis of 5,^,5-tris(perfluoroalkyl)pyrazoles by reaction of perfluorobutyne-2 with 5-perfluoroalkyltetrazoles 90 Reactions of perfluorobutyronitrilimine on a cold finger 93 Thermolysis of 5-perfluoropropyltetrazole with no acceptor molecule present 98 Deuteration of the Perfluoroalkyl Substituted Tetrazoles, Triazoles and Pyrazoles 99 IV. SlMr-lARY 101 BIBLIOGRAPHY 10^ BIOGRAPHICAL SKETCH 107 vi

PAGE 8

LIST OF TABLES Table Page lo 2,5-bis(Perriuoroalkyl)-l,3,^-oxadiazoles ... 13 2 o 1 , 3-bis ( 5-Perf luor oalkyl-1 , 5 , ^-oxadiazol7l-2) perfluoropropanes 1? 3 . 5,5' -bis (Perf luor oalkyl ) -2 , 2 ' -bi-1 , 3 ,^oxadiazoles , 19 ^. The Effects of Hydrogen Chloride on the Reaction of 5-Perfluoropropyltetrazole with Perfluorobutyronitrile 29 5. Synthesis of 5,5-bis(Perfluoroalkyl)-l,2,^triazoles in a Flow Reactor . 32 6. N.M.R. Spectral Data of 3 , ^ « 5tri s (Perf luoroalkyl)pyrazoles 36 7. N.M.R. Spectral Data of 2-H-Perfluorobutene-l . A-0 8o pKa Values of Acidic Azoles 5^ vii

PAGE 9

LIST OF FIGURES Figure Page 1. Infrared spectra of: (A) 3>5' "bis (perfluoromethyl)-2,2'-'bi-l,3,^-oxadiazoreT (B) 5,5'bis(perfluoroethyl)-2,2'-bi-l,3,^oxadiazole; (C) 5>5'-^bis(perfluoropropyl)2,2'-bi-l,3,^-oxadiazole 20 2. Infrared spectra of: (A) N,N'-bis(perfluorobutyrimidoyl chloride )urea; (3) 2-(perfluorobutyrimidoyl chloride)amino-5-perfluoropropyl-1 ,5,^-oxadiazole; and (C) N-(perfluorobutyrimidoyl chloride)perfluorobutyramide 22 5. Infrared spectra of: (A) 2-H-perfluorobutene-1 ; and (B) 3>6-bis(perfluoropropyl)l,2-dihydro-l,2,^,5-tetrazine 42 ^. Infrared spectra of: (A) 5-perfluoromethyltetrazole; (3) 5-perfluoropropyltetrazole; and (C) 1 , 3hi s ( 5-te trazolyl )perf luoropropane 47 5. Infrared spectra of: (A) 3>5-his(perfluoropropyl)-l,2,A— triazole; (3) 5-perfluoropropyl-5-perfluoroethyl-l,2,^-triazole; and (C) 3-perfluoropropyl-5-perfluoroEiethyl1,2,^-triazole 48 5. Infrared spectra of: (A) 3-perfluoropropyl^,5-his(perfluoromethyl)pyrazole; and (B) 3 , ^ , 5-tris(perf luoromethyl )pyrazole 49 7o Apparatus for attempted isolation of 1(2)perf luorobutyryl-5-perf luoropropyl tetrazole . 72 8. Flow reactor for 3 > 5"bis (perf luoroalkyl )1,2,4-triazole synthesis 86 9o Apparatus for preparation and reaction of perfluorobutyronitrilimine 9^ viii

PAGE 10

PRELIMINARY REMARKS All temperatures reported are uncorrected and in degrees Centigrade. Nuclear magnetic resonance measurements were obtained using the Varian high-resolution spectrometer and are expressed in parts per million (p.p.m.)» Infrared spectra were obtained using the Perkin-Elmer Infracord recording spectrophotometer. Ultraviolet spectra were obtained using the Beckman Model DK-2 recording spectrophotometer. Chromatographic analyses were obtained using a Perkin-Elmer Model 15^ Vapor Fractometer. Elemental analyses were determined by the Schwarzkopf Microanalytical Laboratory, 56-19 37th Avenue, Voodside 77, New York, The term R^, appears frequently in this report and is used to denote a perfluoroalkyl radical. ix

PAGE 11

CHAPTER I INTRODUCTION The tetrazoles are a class of organic heterocycles characterized by the presence of a five membered unsaturated ring consisting of one carbon and four nitrogen atoms. The parent compound, tetrazole, like the 5-substituted tetrazoles, exists in tautomeric fonns I and II, and while the N=N .N — N H-^ I H-C I '^N NH N — N H I II ring is sometimes referred to as "1-H tetrazole" the hydrogen is apparently equally bound to all the nitrogen atoms of the ring. This structure is apparent from the facts that methylation of its salts with methyl iodide 2 produces both 1and 2-methyl derivatives while methylation of the free tetrazole with diazomethane produces the 1-methyl derivative^ and acylation gives the 2-acyl tetrazoles. A more correct way of depicting the ring would be as follows: 1

PAGE 12

^•"^ R — c: H ^NThis structure would explain the complex N— H absorption in its infrared spectra. It would appear to the uninitiated that the tetrazoles would be relatively unstable due to the high nitrogen catenation. This expected instability is shown when the tetrazole ring is linked to a chain of nitrogen atoms; tetrazolyl-5 azide explodes in aqueous solutions of concen5 trations greater than about 2 per cent at 0°. However, the majority of tetrazoles are reasonably stable and a few, such as 5-(benzoylamino) tetrazole, are stable at temperatures approaching 500°. Tetrazole salts in general have been found to be more thermally stable than the free tetrazole. The tetrazole ring normally survives such reactions as oxidation, reduction, decarboxylation and hydrolysis of its substituents. Only a few tetrazoles have achieved practical importance. The heavy metal salts of a number of the less stable tetrazoles have been described in patents as being useful components in priming and initiating compositions for explosives. 1,5-Pentamethylene tetrazole (Metrazole) is extensively used as a general cardiac and respiratory stimulant.

PAGE 13

The first tetrazole was prepared by Bladin in 1885 by diazotization of w -N-pbenyl-C-cyanof ormamidrazone. NC-C I NH + HNO, NO— C / N=N NH-NH ^6% "^N — N-CgHc A similar method employing nitrous acid and aryl hydrazidines has become important in the synthesis of 5-aryl tetrazoles. The reaction is presumed to proceed via an intermediate imidazide which rearranges to give the tetrazole. Ar-C \ NH NH-NH2 + HNO2 NH 11 Ar — C-N, ^Ar— C / N=N ^^_ NH The reaction of hydrazoic acid or its salts with organic nitriles produces 40-90 per cent yields of 5-aryl or alkyl tetrazoles and is also believed to proceed through an intermediate imidazide. RON + HNRC .^ NH \ N, RC / N = N ^ N•NH The syntheses of two 5-perfluoroalkyl-substituted tetrazoles have been reported using the azide-nitrile 2 12 reaction. '

PAGE 14

In 1929, Stolle -^ reported that 5-aminotetrazole formed the exocyclic N-acetyl derivative (III) when gentlj warmed with acetic anhydride. With vigorous heating of the reaction mixture, nitrogen was eliminated and the l,3»^-oxadiazole (IV) formed. im2-c / N: ^ :1T N — NH iGE^OO)^0 -* CH,-C-N-C H ^ N = N N — NH III reflux (CH^CO)oO 8 hrs :> ^ ^0 CH-,-C-N-C C-CH-. + No " N N IV Recently the synthesis of l,3»^-oxadiazoles hy the acylation of 5-aryl and alkyl tetrazoles has been examined by Huisgen and his coworkers. ' * "^ They suggest the 1^ following mechanism for the reaction. The majority of

PAGE 15

R-C I N=1T R'COCl' N NH © I R-C=N-N=C-R' -» R-C I ^ N — N-C-R' VI -7. R-C=N-N-C-R' -N, 1 2 II R-C C-R" R-C -^N-N^ C-R' available data indicates that VI forms directly from the acylated tetrazole hut kinetic data leave open the possibility that the slow step is the transformation of V to VI. In either case, VI is assumed to be the immediate precursor of the oxadiazole. Huisgen also investigated an analogous reaction with imidoyl chlorides to give 5,A-,5-trisubstituted-l ,2,^triazoles and proposed a mechanism analogous to that in the tetrazole-acylchloride reaction. With only a few exceptions, these syntheses of oxadiazoles and triazoles produced 75-100 per cent yields.

PAGE 16

E-C I + "^ C-R' -» R-C / ^ N =:N N — R" N II N—C-R' -N. R" R-C C-R' R" ©0 II R-C=1T-1T-C-R' t R" © © I R-C=N-N=C-R' t R" N II R-tJ-NrN-C-R' In the past several years, there has been considerable interest in this laboratory in heterocycles containing perfluoroalkyl substituents. One class of heterocycles , the 2,4,6tri's (perfluoroalkyl)-s-triazines exhibited excellent thermal stability and synthesis of fluorocarbon polymers 17 containing £-triazine rings in the backbone was studied. The polymers reported suffered from the inherent drawback of nonlinearity. 2 , 3bis (Perfluoroalkyl ) -1 , 3 « ^-oxadiazoles and 3»5bis (perfluoroalkyl )-l, 2 ,^-triazoles were previously synthesized in this laboratory. The oxadiazoles were prepared by cycle dehydration of a bis (perfluoroacyl)hydra18 zine and the triazoles by dehydration of an N-acyl

PAGE 17

hydrazidine formed by the reaction of a 2,5-bis(perfluoro19 alkyl)-l ,3»^-oxadiazole with ammonia. These heterocycles also possessed good thermal stability and when incorporated into a polymer structure probably would not lead to branching or crosslinking. In view of the 75-100 per cent yields of the acylation reactions of the 5-substituted tetrazoles, a similar method employing ditetrazoles and diacid chlorides was considered as a synthetic route to fluorocarbon polymers containing l,5»^-oxadiazole rings in the backbone. However, the information available on the reactions of the 5-perfluoroalkyltetrazoles was insufficient to determine their suitability for use in polymerization reactions. It was considered advisable, therefore, to investigate initially those reactions of the 5-perfluoroalkyltetrazoles which might furnish a route to other simple heterocyclic systems. Information gained might then be translated to polymer producing reactions. Statement of the Problem The primary purpose of this problem was to investigate some of the ring-opening reactions of the 5-perf luoroalkyltetrazoles as a route to the formation of new heterocycles and to note the effect of the perfluoroalkyl group on the

PAGE 18

8 reactions o Since some of these reactions were carried out at temperatures above the thermal ring-opening temperature of the pure 5-perfluoroalkyltetrazole, a further objective was to show the mechanism of the thermal ring-opening reaction and to trap and identify the intermediate by reacting it with other materials to give more stable, knovm species. As the investigation proceeded, several new heterocycles were prepared. A further objective of this investigation therefore was to compare their spectra and to compare the acidic properties of those heterocycles with hydrogen attached to the ring nitrogen and to show the effect of the perfluoroalkyl group on the acid strength.

PAGE 19

CHAPTER II DISCUSSION Preparation and General Properties of 3-Perfluoroalkyltetrazoles The perfluoroalkyi nitriles are readily attacked by nucleophiles such as ammonia, hydrazine and hydrogen sulfide. These reactions produce the stable perfluoroalkyi amidines, hydrazidines and thioamides, respectively, by the following general mechanism: e .N .NH Rj,-C=N > Rp-C > Rp-C^ where B NH2-, ^2^3'' ^^" * A similar reaction with azide ion in anhydrous media results in formation of the imidazide (VII) which rearranges spontaneously to the tetrazole. R c=^ ^ V\ ^ V\ ( ®)l VII

PAGE 20

10 2 Prior to the appearance of Norris' paper describing the synthesis of 5-perfluoromethyltetrazole in anhydrous acetonitrile, 5-perfluoropropyltetrazole and l,3-his(5tetrazolyl)perfluoropropane were prepared in the present study by reaction of the appropriate nitrile with sodium azide in a methylcellosolve-acetic acid mixture using a method similar to that described by Mihina and Herbst for the preparation of 5-alkyl and 5-aryl substituted tetrazoles. The methylcellosolve, dried by azeotroping with benzene, contained traces of water which hydrolyzed part of the perfluoroalkyl nitrile to the amide. The amide was readily removed from 1 ,3~ bis (3-tetrazolyl)perfluoropropane by recrystallization of the product from perfluorobutyric acid. Perfluorobutyramide could not be readily removed from 5-pei'fluoropropyltetrazole, however, and purification of this tetrazole required preparation of its silver salt, from which the amide could be removed by washing with acetone. After regeneration and distillation, 5-perfluoropropyltetrazole of high purity was obtained and slowly crystallized on standing at room temperature. Norris' method, adapted to the preparation of 5-perfluoropropyltetrazole, gave a product of slightly lower purity which did not crystallize at room temperature; however, the method is less laborious than the longer route involving preparation of the silver salt.

PAGE 21

11 The 5-perfluoroalkyl substituted tetrazoles are colorless, odorless strong acids vrLth the characteristic sour taste of an acido In contact with the skin, they produced a chemical burn similar to that produced by glacial acetic acid. Pure l,3-bis(5-tetrazolyl)perfluoropropane slowly decomposed at 170*» while 5-perfluoromethyltetrazole and 5-perfluoropropyltetrazole decomposed at about 180°. As has been observed with 5-aryltetrazoles, ^ the sodium salts of 5-perfluoropropyl and 5-perfluoromethyl tetrazole were more stable thermally than the free tetrazole and decomposed slowly at 2^5" and 258°, respectively. The latter salt detonated at $00°. Silver 5-perfluoropropyltetrazole undergoes a slow decomposition at 290° which becomes very rapid, but not explosive, at 350°. Both silver 5-perfluoropropyltetrazole and disilver 1 ,3-bis(5-tetrazolyl)perfluoropropane were found to be stable to bright sunlight. Mihina and Herbst reported several silver tetrazoles were light sensitive but also noted that they occluded silver nitrate. This sensitivity may be due to traces of occluded silver nitrate, for if reduction of silver ion is to occur, a corresponding oxidation of the tetrazole must also take place and tetrazoles are generally characterized by excellent oxidative stability.

PAGE 22

12 The 5-perfluoroallcylte-trazoles are stable to both hot aqueous acids and bases. The former is not surprising since the tetrazoles axe strong acids themselves o In alkaline solutions, the tetrazole anion forms; since this 2 anion is a nucleophile itself, it is not readily attacked by another nucleophile o 5-Perfluoropropyltetrazole was found to be stable to both alkaline and acidic potassium permanganate at 100®. The tetrazole ring in general is stable to oxidizing agents. If the ring is substituted v;ith the p-aminophenyl or the thiol group, these substituents may be removed by oxidation without ring rupture. Since the perfluoroalkyl group is not easily oxidized, no reaction of 5-perfluoropropyl tetrazole occurred with potassium permanganate, even at 100°. The infrared spectra of the 5-perfluoroalkyltetrazoles are shown in Figure ^. Reactions of the 3-Perfluoroalkyltetrazoles Below the Thermal Ring-Opening Temperature Synthesis of 2,5-bis(perfluoroalkyl)-l ,3,^-oxadiazoles The synthesis of 2, 5-bis (perfluoroalkyl )-l, 3,^oxadiazoles was conveniently accomplished by reaction of a 5-perfluoroalkyltetrazole with a perfluoroacyl chloride. The reactions were carried out in sealed glass tubes since

PAGE 23

13 the temperatures employed were above the boiling points of the perfluoroacyl chlorides. In small batches, the crude products were readily purified by separating volatile and non-volatile components on the vacuum system. Table 1 shows the 2,5-bisCperfluoroalkyl)-l»3,A-oxadiazoles prepared by this reaction. TABLE 1 2 , 5-b i s ( PERFLUOROALKYL ) -1 , $ , ^-OXADI AZOLES %

PAGE 24

14 An attempt was made to isolate the intermediate 2-perfluorobutyryl-5-perfluoropropyltetrazole, yN=N 3 7 <^ II N — N-C-C,Fy by acylating the silver salt of the tetrazole at 0-25*=> in tetrahydrofuran. Even under these mild conditions, the acylated tetrazole was found to be unstable and rearranged spontaneously with nitrogen elimination to give 3>5-bis(perfluoropropyl)-l,5,4--oxadiazole in 95 per cent yield. The instability of the 2-perfluoroacyl-5-perfluoropropyltetrazoles was in sharp contrast to that of the 2-acyl-5-aryltetrazoles which require a temperature of 110-14-0*» for rearrangement to the oxadiazole. The latter proceeded via 4 24 the following mechanism: ' AT-O 1 » ^N — N— C-R o; Ar-C -N=:N -C-R © .. .. II Ar-C = N-N=C-R Ar-C ^C-R

PAGE 25

15 However, since the high electronegativity of the perfluoroacyl group markedly weakened the tetrazole ring, the ring opening and rearrangement to the oxadiazole probably proceeded as follows: ^^-^^ N^N N NH 01 u /^"^'f -HCl V\ N = N N\ IK. pj N Rp,-C 2 e -N. R -c ;c-R'r, Since nitrogen leaves with its electrons from an electropositive carbon, its leaving may be assisted as diagrammed above. Synthesis of 1 ,3bis (3-perfluoroalkyl-l,3,^-oxadiazolyl-2)perfluoropropanes The reaction discussed above for the synthesis of 3,5-bi_s(perfluoroalkyl)-l ,5,4— oxadiazoles was extended to the synthesis of 1 ,3bis (perfluoroalkyl-l ,3,^-oxadiazolyl-2)perfluoropropanes. Two routes were available for this

PAGE 26

16 synttiesis, the use of a monotetrazole with a diacyl chloride (A) and a difunctional tetrazole with a monoacyl chloride (B)« l,3-his(5-Perfluoroprop7l-l,3,4— oxadiazolyl-2)perCA) 2 Rj.-C^ / ^ N = N II II N NH + C1-C-(CF2)3-C-C1 2 HCl + 2 N. N-IT ^ N-N N=N. N = 1T (B) I c-CCF2)3-C I HN N'^ ^N NH ^' CI fluoropropane was prepared by methods A and B in sealed tubes. In theory, both reactions should occur with the same ease but in practice, method B was preferred since it resulted both in higher yields and a more easily purified product. There were several reasons for this. First, the monotetrazoles were slightly hygroscopic while l,3-bis(5tetrazolyl)perfluoropropane was not. Also, using method A, the perfluoroglutaryl chloride was exposed to the atmosphere during weighing, resulting in sdme hydrolysis, while in

PAGE 27

17 method B the perfluorobutyryl chloride was measured in the vacuum system and condensed into the reaction tube with no contact with the atmosphere. Further, since it is usually advantageous in the reaction of a monofunctional compound with a difunctional compound to use a slight excess of the monofunctional material to promote higher yields, the excess perfluoroacyl chloride employed in method B was simply removed under reduced pressure while the excess 5-perfluoroalkyltetrazole in method A required more elaborate techniques. Table 2 shows the 1 , 3bis ( $-perf luoroalkyl-l,3>^-oxadiazolyl-2)perfluoropropanes prepared by methods A and B. TABLE 2 1 , 3-bis ( 5-PERFLUOHOALKYL-l , 3 , A-OXADI AZOLYL-2 ) PERFLUOROPHOPANES ^ N-N ^N-N h

PAGE 28

18 The infrared spectra of the 1 ,3"bis (3-perfluoroalkyl-l,5,^-oxadiazolyl-2)perfluoropropaiies showed the same weak ahsorption at 6.55 and 6*^0 ;i, assigned to cyclic C=N stretching that was found in the 3,5-his(perfluoroalkyl)18 1 , 3 ,^-oxadiazoles . Synthesis of 5,5'-bis(perfluoroalkyl)-2,2'-bi-l ,5,^oxadiazoles The synthesis of the 3,3' bis (perf luoroalkyl ) -2 , 2 ' bi-1,3,4— oxadiazoles was accomplished by two synthetic routes analogous to A and B shown on page 16. The yields .1T = IT / I II II (C) 2 Rj,-C I + Cl-C-C-Cl N — NH -2 HCl -2 N^ 0. ^0 R„-C ^C-C ^C-R™ F <^ ^ ^ ^ ^ C,) 2H,-C^ ^Na^|('-)>-c((-)| Na^ ^ ^01 N— 'N*^ ^N^^N from method C were much lower than D for the same reasons given on page 16 for method Ao The reaction occurred by the same mechanism described for the synthesis of the 3,5-bis(perf luoroalkyl )-l, 3, ^-oxadiazoles described on page 150 Sodium

PAGE 29

19 chloride rather than hydrogen chloride was eliminated in method D. Table 5 shows the 3 > ^ ' bis (perf luoroalkyl ) -2 ,2 ' bi-1 ,5,^-oxadiazoles which were prepared. These compounds showed several interesting differences from the compounds containing isolated oxadiazole rings. The bioxadiazoles were very high melting by comparison and showed surprising resistance to hydrolysis by hot aqueous alkali while the 1 , 3-bis(perf luoroalkyl-1 , 3,4— oxadiazolyl-2)perf luoropropanes were destroyed in hot water. The infrared spectra of the bioxadiazoles, shown in Figure 1, were also different. The band at 6.35 /i did not appear in the spectra of the bioxadiazoles while another considerably stronger band appeared at 5.81 p. and was probably due to the extended conjugated system, C=N-1T=C-C=N-IT=C. TABLE 5 5,5' -b i s C PSRFLUQROALKYL ) -2 , 2 ' -bi-1 , 5 ,4-OXADlAZOLES 0. ^F~^K ^C-^K /C-^F ^ N-N ^ N-N Nitrogen Evolution Bioxadiazole a Per Cent Per Cent ^ F Method of Theory of Theory N.P. max Log 6 ' CTfip C ^ 55 163.5-16^.5° 165.0-155.8° 234-. 4.09 182.0-182.8° 232.5 4-. 10 193.1-193.9*' 250.5 ^.06 In i-propyl alcohol solution. C,F^-

PAGE 30

20 4000 3000 100 2000 1500 CM' 1000 900 800 700 / \ / \ ' W // W // ' S— » S— N 0.0 1 1 I i i r 7 8 9, 10 , 11 WAVELENGTH (MICRONS) 4000 3000 2000 100 "" ^ ' ' ' ' ^60 < ^40 Z < ^20 1500 , I CM' 1000 900 800 700 C,fc-C C — C 0-0,?, ^ * W // W // ^ ^ N— » H— S -1 1 ' 6 I 1 I ; 1 ; 1 I 7 8 9 10 11 WAVELENGTH (MICRONS) 0.0 4000 3000 2000 100 I ''''' " 'I ' 1500 CM' 1000 900 800 700 / \ / \ C,F_-C C — C 0-0,7, N— S w // n—v null Kel f oil 7 8 9 10 , 11 WAVELENGTH (MICRONS) 0.0 O 3'° > .4Z n .5™ .6 .7 Fig. 1. -Infrared spectra of: (A) 3,5'-bis(perfluoromethyl)-2,2'-.bi-l,3,^-oxadiazole; (B) 3.3^is(perfluoroethyl)-2,2'-bi-l,3,4-oxadiazole; (C) 3,3' -bisCperf luoropr opyl ) -2 , 2 ' -bi-l , 5 ,^-oxadiazole .

PAGE 31

21 Tlie bioxadiazoles also showed strong ultraviolet absorption in the 2$0-234 m;i region (see Table 3) while the 2,5-bis(perfluoroalkyl)-l,5,^-oxadiazoles showed no absorption maxima in the 220-5^0 m;i region.^® The absorption in this region is attributed to the increased conjugation. Reaction of 5-perfluoropropyltetra2ole w ith phosgene Two moles of 5-perfluoropropyltetrazole were found to react with one mole of phosgene in a closed system to give a white solid product tentatively assigned the structure CI \ C^F„-C C-NH-N C-C,Fo ^ N-N VIII on the basis of it's infrared spectra (see Fig. 2B) which showed a broad absorption at 5.1-5.^^ due to N-H stretching and a sharp band at 6,10 ji due to exocyclic C=N stretching. The cyclic C=N absorption of the oxadiazole ring appeared as a weak band at 6.3^>i. The band at 6.82^ was due to N-H deformation and shifts to a higher wavelength when the compound was deuterated. The chlorine was surprisingly stable and did not react with water but did react with dilute aqueous base. Under the reaction conditions, part of VIII reacted further with the evolved hydrogen chloride to give

PAGE 32

4000 3000 100 2000 1500 CM' 1000 900 L_ 800 700 Cl Cl 0,P.-C.H-H-O-N-N.C-C,F_ Mull Xel P oil 0.0 -.2cB O > .4Z n .5"< .6 .7 1.0 1.5 oo 4000 3000 100-^ 2000 1500 1 1 WAVELENGTH (MICRONS) CM' 15 1000 900 — I , I 800 700 N— N Mull K«l F oil 0.0 -< 1 17 8 9, 10 ,11 WAVELENGTH MICRONS 1 4000 3000

PAGE 33

23 CI CI C , P„C =N-NH-C-NH-N= C-C jFr; IX • The infrared absorption of IX (see Pig. 2A) was very similar to that of VIII. The bands due to N-H stretching were more defined and occurred at 2,9^, 3 •20 and 3.^1 U. The N-H deformation band occurred at 5.69 ^. The bands assigned to N— H stretching and deformation shifted to higher wavelengths when the compound was deuterated. The C=0 and C=N stretching absorptions occurred at 5.70 and 6.11^, respectively. The experimental molecular weight and elementary analysis of IX were in good agreement with the proposed structure. To prove that IX was derived from VIII, a sample of VIII isolated from the reaction mixture was heated in the presence of gaseous hydrogen chloride in a sealed tube and thereby converted to IX. Since the reaction of a perfluoroalkyl substituted 1 ,3-^-oxadiazole with hydrogen chloride has not been reported, a sample of 2 , 5bi s (perf luor opropyl ) -1 , 3 , ^-oxadiazole was heated at 210225® in "the presence of a large excess of hydrogen chloride and produced a white solid with an infrared spectra similar to that of IXo This product was assigned structure X resulting from the ring opening addition of hydrogen chloride to the oxadiazole. Only low yields of X have been obtained which may be due to reversibility of the reaction. In a

PAGE 34

2H.0. CI ^ \ « I n CjFr^-C C-C^Frp + HCl 210^ ^ C^Fr7C=N-NH-C-C,Fr7 X preliminary experiment, a low yield of X was also produced from the reaction of N,N'-ibis(perfluorobutyryl)h,ydrazine with phosphorous pentachloride. The chlorine atom of X was also not readily attacked by water at room temperature. The infrared spectra of X (see Fig. 2C) showed bands due to N-H stretching at 5*08 and $.26 _;i and the N-H deformation band at 6.73 >i» The absorption due to C=0 stretch occurred at 5«61 ^ which was lower than found in the perfluoroalkylamides. This shift was due to the electronegative imidoyl chloride substituent. The C=N absorption of X occurred at 6.01^. In a further experiment designed to establish the role of hydrogen chloride in the reaction of phosgene with 5-perfluoropropyltetrazole, equivalent amounts of phosgene and tetrazole were heated in a sealed tube in the presence of four equivalents of hydrogen chloride. The infrared spectra of the crude product showed it to be nearly pure IX from which pure IX was recovered in 55 per cent yield, No VIII was recovered. Additional work is required on the reaction of the perfluoroalkyltetrazoles with phosgene to determine whether the route of the reaction is represented by E or F below.

PAGE 35

25 IT=N N=N. E) R5,-C I + Cl-C-Cl + I C-Rj, _2HG1^N— NH HN N'^ N=N N = N ^ % I II I ^ I ^N N-C-N — N -2IT2 +HC1 CI CI ^0 01 Rj,-C=N-]m-C-NH-N=C-Rp ^hcT" ^"^ ^C-NH-N=C-Rp XII XI ^N = N .N^N F) Rj-C I . Cl-C-Cl ^:^jcl^ R^-C 1 _ ^N — NH N — N-C-Cl \ R^-c ;c-ci XIII / R^CN.H XII < XI <— ^— R„-C ^C-N — N '^ HCl +HC1 '^N-N^

PAGE 36

26 If the 2-perfluoroalk7l-5-chloro-l,3,^-oxadiazole (XIII) can be synthesized, the chlorine should be reactive, as in the chlorotriazines or 2-chloropyridine for example, and may then be reacted with the tetrazole following route F. If successful, this reaction would be good evidence that the reaction of phosgene with 5-perfluoropropyltetrazole occurs by route F which is the route to be preferred since it eliminates the need of the multicentered reaction in the first step of E. The structure of XI may be definitely established if the chlorooxadiazole XIII will react with a perfluoroacyl hydrazide. The resulting hydrazide substituted oxadiazole on reaction with phosphorous pentachloride in an open system should give XI. Reaction of ^-perfluoropropyltetrazole with perfluorobutyronitrile: synthesis of 5,5-bis(perfluoropropyl)-l ,2,4triazole The reaction of 5-perfluoropropyltetrazole with perfluorobutyronitrile was attempted in a sealed tube at temperatures of 100-135** but was not successful. In view of the pronounced ring destabilizing effect shovm by the electronegative perfluorobutyryl group, it appeared that a strong Lewis acid might complex with the electrons of the tetrazole ring and cause it to rupture at relatively low temperatures. The dipolar fragment produced could then add to the perfluorobutyronitrile.

PAGE 37

27 /. N — NH CjF^CN H Boron trifluoride-etherate was found to produce a low yield (11.6 per cent) of $,5-bis(perfluoropropyl) -1,2,4triazole when heated at autogeneous pressure in a sealed tube with one equivalent of 5-perfluoropropyltetrazole and five equivalents of perfluorobutyronitrile. However, based on the nitrogen evolution, only one tetrazole molecule out of four v/hich underwent ring cleavage added to the nitrile to produce a molecule of triazole. Gaseous boron trifluoride formed a stable 1/1 adduct with 5-perfluoropropyltetrazole at autogeneous pressure in a sealed tube. VHien heated at 155** in a sealed tube, the adduct broke liberating free 5-perfluoropropyltetrazole and boron trifluoride. No nitrogen v/as evolved. At low pressure, the adduct gave off boron trifluoride leaving the 5-perfluoropropyltetrazole unchanged o It is apparent that boron trifluoride exerted no pronounced influence on the thermal stability of 5-perfluoropropyltetrazole.

PAGE 38

28 Two reactions were carried out in which, one equivalent of 5-perfluoroprop7ltetrazole was heated with two equivalents of perfluorobutyronitrile in the presence of 10 and 1 mole per cent boron trifluoride. At 10 per cent boron trifluoride, 3>5-bis(perfluoropropyl)-l,2,^-triazole was obtained in 10 per cent yield while at 1 per cent boron trifluoride a 5 per cent yield was obtained. Anhydrous hydrogen chloride was also investigated as a catalyst for promoting a reaction between perfluorobutyronitrile and 5-perfluoropropyltetrazole. After prolonged heating at 125** in a sealed tube with a large excess of hydrogen chloride, the tetrazole was recovered unchanged. However, in the presence of three equivalents of perfluorobutyronitrile the yields of 5,5-bis(perfluoropropyl)-l,2,4-triazole increased with increasing amounts of hydrogen chloride (see Table A-). The reaction probably proceeded through the unstable perfluorobutyrimidoyl chloride (XIV) and followed a path analogous to that proposed for the oxadiazole formation.

PAGE 39

29 TABLE 4 THE EFFECTS OF HYDROGEN CHLORIDE ON THE REACTION OF 5-PERFLUOROPROPYLTETRAZOLE WITH PERFLUOROBUTYRONITRILE NOo

PAGE 40

50 25 perfluoroacetonitrile, -^ A series of A— aryl substituted 1,2,^-triazoles was prepared from a similar reaction employing the more stable N-aryl benzimidoyl chlorides. In order to eliminate any possibility that the effect of the hydrogen chloride is one of activating the tetrazole (rather than the nitrile) and that it is an activated tetrazole molecule which reacts with the carbon-nitrogen triple bond, reaction 5 in Table ^ was repeated replacing the perfluorobutyronitrile with an equivalent amount of perfluorobutyne-2. No nitrogen was evolved and the tetrazole was recovered unchanged. If the hydrogen chloride served to some way activate the tetrazole, the tetrazole would have reacted with the butyne as it did with the nitrile. The fact that this reaction did not take place is further evidence that the hydrogen chloride acts upon the nitrile as shown. Reactions of 3-Perfluoroalkyltetrazoles at Temperatures Above the Thermal Ring-Opening Temperature Reactions with perfluoroalkylnitriles ; synthesis of 3,3bi_s(perfluoroalkyl)-l,2,^-triazoles The thermolysis of 5-perfluoropropyltetrazole in the presence of excess perfluorobutyronitrile at autogeneous pressure in a sealed tube produced low yields (16 per cent)

PAGE 41

31 of 5j5-bisCperfluoropropyl)-l,2,^-triazole, The product resulted from the addition of the nitrile to the dipolar nitrilimine fragment (XV) which was produced by thermolysis of the tetrazole. A similar structure has been proposed "^N — NH XV 1 C,F„CN H N ~ N-N ^ for the fragment resulting from the thermolysis .of 2,5-diphenyltetrazole. In the sealed tube reaction, the tetrazole decomposed in the liquid phase while the nitrile was in the gas phase and the reaction occurred only at the interface or with the nitrile dissolved in the liquid tetrazole,, In order to increase the yield of this reaction, the tetrazole was heated in the vapor phase in the presence of the nitrile using the flow reactor shown in Figure 8. The yield of 3,5-bis(perfluoropropyl)-l,2,'4— triazole was increased by this method to 57 per cent. Table 5 shows the results of extending this reaction to the synthesis of triazoles with other perfluoroalkyl substituents.

PAGE 42

32 TABLE 5 SYNTHESIS OP 3 , 5-bis (PERPLUOROALKYL) -1,2, ^-TRIAZOLES IN A PLOW REACTOR Ej,C I .E.j,CN^ Rj,-0 0-E'p ^

PAGE 43

53 for the formation of the triazoles except the dipolar nitrilimine fragment added to a carbon-carbon triple bond rather than to the carbon-nitrogen triple bond, . In contrast to the triazoles, the pyrazoles were easily purified by standard techniques o ^N — N

PAGE 44

3^ At the initiation of this work, it was not known if the 5-perfluoroalkyltetrazoles would undergo thermal ring cleavage with elimination of the 2,3or the 3>^nitrogen atoms. The fragment from either mode of cleavage could add to a nitrile to produce a triazole but the addition product with perfluorobutyne-2 would result in an imidazole (XVII) from 2,3-nitrogen elimination and a pyrazole (XVI) from 3 » ^-elimination as shown below.

PAGE 45

RpC-N-NH ®t e Rp-C=N-NH XV N— NH 'CF ,C5CCF-. R'j.CN XVI 35 Rp-C N=N N NH 3,^ elimination 2,3 elimination N—N R^-C \ H N I Rp-C / N H Ne <^ N © R'pCN .CF^C=CCP, XVII

PAGE 46

35 N.M.R. spectra were obtained on the 5-perfluoroalkyltetrazole-perfluorobutyne-2 reaction products and as seen from Table 6 definitely establish the pyrazole structure. TABLE 6 N.M.R. SPECTRAL DATA OF $ , 4 , 5tri s (PERFLUOROALKYL) PYRAZOLES CF-, (a) C CF,-CFo-CFo-C ^C-CF, (b) 3 2 2 \\ / 3 N— NH (c) (e) (d) Group

PAGE 47

37 The proton nuclear magnetic resonance spectrum of 3-perf luoropropyl-^ , 5-bis (perf luoromethyl )pyrazole was obtained and indicated that the hydrogen was on a nitrogen. The equivalency of the 3 and 5 methyl' groups in 5>^>5tris (perf luoromethyl )pyrazole established the mobility of the H atom, e.g . , \ y \ ? CP,-C C-CF, ±=^ CF,-C C-CF, ^ N-N N-N H H The definite establishment of the pyrazole structure conclusively demonstrated that the 5-perfluoroalkyltetrazole rings opened with elimination of the 5>^ nitrogen atoms to give the nitrilimine XV. Reactions of the perfluorobutyronitrilimine fragment under mild conditions In order to further establish the existence of the perfluorobutyronitrilimine fragment in the vapor phase thermolysis reactions of 5-perfluoropropyltetrazole, the tetrazole was distilled at low pressure into a short, tubular furnace and the thermolysis products condensed on a liquid nitrogen cooled cold finger (see Fig. 9)o The material with which the nitrilimine fragment was to react was also condensed on the cold finger. The reactions took

PAGE 48

38 place at an undetermined temperature by allowing the cold finger to warm slowly to room temperature* . 5,5-bis(Perfluoropropyl)-l ,2,^-triazole resulted from the reaction of perfluorobutyronitrilimine with perfluorohutyronitrile and 3-perfluoropropyl-5-perfluoromethyl-l,2,^-triazole from reaction with perfluoroacetonitrile. The reaction of perfluorobutyne-2 with perfluorobutyronitrilimine produced $-perfluoropropyl-4,^bis (perfluoromethyl)pyrazole and reaction of the fragment with water gave perfluorobutyrhydrazide. The formation of the heterocycles from this reaction has already been discussed. The hydrazide is produced as follows: ^ ^ OH C,Fr;C=N-NH + H2O — *C,Fr,C=N-NH2 ^^ C,Fr;C-NH-NH2 From this series of experiments it was concluded that perfluorobutyronitrilimine is the reactive species in triazole and pyrazole synthesis described above. Further, perfluorobutyronitrilimine has a finite lifetime and is sufficiently reactive to attack perfluoroalkylnitriles, perfluorobutyne-2 or water at or below room temperature* Thermolysis of 3-perfluoropropyltetrazole with no acceptor molecule present The literature contains few reports of the decomposition, due to purely thermal effects, of 5-aryl-substituted 2S tetrazoles and none of 5-alkyl tetrazoles. Pinner

PAGE 49

39 observed the formation of 5,5-diplienyl-l,2»^-triazole and 5,6-diplienyl-l,2,^,5-tetrazine when 5-phenyltetrazole was heated above its melting point. In a more thorough study, Huisgen^ observed, in addition, the formation of 2,^,6triphenyl-s-triazine , ^-amino-3 , 5-diphenyl-l ,2,^-triazole and 3,6-diphenyl-l,2-dihydro-l,2,4,5-tetrazine<. These thermolyses were carried out at about 200° and the products indicated that the tetrazole ring opened both by loss of hydrazoic acid to produce benzonitrile and by loss of the 5,^ nitrogen atoms to produce the phenylnitrilimine fragment. When 5-perfluoropropyltetrazole was passed at low pressure through a hot tube heated to about ^00« , and the thermolysis products quickly removed from the heated zone, the reaction followed a course different from that reported for the decomposition of the 5-aryltetrazoles at atmospheric pressure. No formation of 5,5-bisCperfluoropropyl)-l ,2,^triazole was observed and the lack of red color in the crude products indicated the absence of $,6-bis(perfluoropropyl)1,2,^,5-tetrazine. The infrared spectra of the crude gaseous products showed the presence of only trace amounts of perfluorobutyronitrile. A group of bands occurred at ^.85, 8.02 and ^3,6-bis(Perfluoroalkyl)-l,2,A-,5-tetraz3.nes, presently under investigation by other workers in this laboratory, have an intense red color.

PAGE 50

^0 9o62-9.67 ^ (doublet). Tiiese bands may be due to tetra27 fluoroallene but due to tbe small amount present, this component was not separated. Another band in the infrared spectra of the crude products occurred at ^.70 ji. This band v;as also due to a minor component which was not separated, but it is interesting to note that it is in the diazoalkane region and may be due to 1-diazo-l-H-perfluorobutane. The major product of the thermolysis was identified as 2-H-perfluorobutene-l, and was obtained in 45 per cent yieldo The N.M.R. spectral data shown in Table 7 were used to determine the structure of the olefin. TABLE 7 N.M.Ro SPECTRAL DATA OF 2-H-PERFLUOROBUTENE-l CF,-CF2-CH»CF2 Group p. p.m.* Area F -4.5

PAGE 51

^1 The infrared spectra (see Pig* 5A) of the olefin showed a strong band at 5.66^ due to C=C stretching. A small amount of yellow 3,6-bisCperfluoropropyl)l,2-dihydro-l,2,4-,5-tetrazine was recovered from the nonvolatile products of the thermolysis. The dihydrotetrazine was identified by its infrared (see Fig. 3B) and ultraviolet spectra which are very similar to 3,6-bis(l-H-perfluoroethyl)-l,2-dihydro-l,2,^,5-tetrazine prepared by Carboni and Lindsey.^® 3,6-bis(Perfluoropropyl)-l,2dihydro-l,2,^,5-tetrazine was easily oxidized to 3,6-bis(perfluoropropyl)-l,2,^,5-tetrazine by nitric acid. The reaction probably proceeds as follows: H H ^3^7\ I 390-39? ^S^7^^^-^^ N — ^ XV \ 1-3 hydride shift c,F^cmT2 H CF,-CF2-CH-CF2 < CF^-CP2-CF2-C ; carbene insertion between << and^ carbons

PAGE 52

42 4000 3000 100 i "" '' " ' l ' " 2000 1500 CM' 1000 I 900 800 I 700 I 0.0 .2cp O > .4Z O .5™ .6 .7 1.0 1.5 7 8 9, 10 , 11 12 13 14 15 WAVELENGTH (MICRONS) 4000 3000 100 I n .. I.... I . 2000 1500 CM' 1000 900 I , I 800 700 -0.0 / \ N— B Mull K«l ? oil 7 8 9, 10 , 11 WAVELENGTH (MICRONS) Fig. 3, -Infrared spectra of: (A) 2-H-perfluorobutene-1; and (B) 3 , 6bi s ( perf luor opropyl ) -1 , 2~dihydro1,2,4,5-tetrazine.

PAGE 53

^3 The thermolysis of ^-perfluoropropyltetrazole initially forms the nitrilimine fragment, XV, as shovm by the addition reactions of this product with perfluoroalkyl nitriles and perfluorobutyne-2. This behavior is analogous to the reported first step of the decomposition of 5-phenyltetrazole. The observed difference in products must occur after the nitrilimine is formed and there appear to be two major factors involved. Pirst, 5-phenyltetrazole was pyrolyzed at atmospheric pressure while 5-perfluoropropyltetrazole was pyrolyzed at low pressures. Therefore, the phenylnitrilimine fragments, once formed, were in closer proximity and had the greater chance of combining to give the observed s^m-tetrazines and N-amino-3,5-diphenyl-l,2,^triazole (from rearrangement of 3,5-diphenyl-l,2-dihydrotetrazine under the influence of heat). The second difference may be seen from a consideration of the relative stabilities of the phenyl and perfluorobutyronitrilimine fragments. Phenylnitrilimine can be stabilized by the following 1^ resonance forms:

PAGE 54

(^2Vc=N-NH ^ r ^3"^^=^ e I N-NH etc <2>-C^N-©0 •NH e \ r-\ ® \ ^a 0© O-^" .NH © © •NH / — V ^ ^ <;>J>-c=N-: "©1 Q=C=N\ © •NH etc Perfluorobutyronitrilimine can have only 5 canonical forms as follows: © C,Fr;C=N-NH XY C5FoC-N=NH XVI 0© C,F„-C=N=NH ©© C,Fy-CsN-NH I © © C^Fr,-C=N-NH 1) ( Thus tlie phenyl group contributes to the stabilization of phenylnitrilimine while the electronegative perfluoropropyl group serves to activate perfluorobutyronitrilimine through XV. Therefore, in perfluorobutyronitrilimine a hydride ion can then easily transfer to the carbon to give 1-diazo-l-Hperfluorobutane which will eliminate nitrogen at the reaction

PAGE 55

^5 temperature to give the carbene. Alternatively, a concerted 1-5 hydrogen shift with simultaneous nitrogen elimination may produce the carbene directly from XVI. The carbene then inserts between the e^ and ^ carbon atoms to give the observed product, 2-H-perfluorobutene-l. Similar c< -^ insertion reactions have been reported in the decomposition 29 of a number of diazoalkanes. Infrared Spectra and Acid Strengths of the Perfluoroalkyl Substituted Pyrazoles, Triazoles and Tetrazoles Infrared spectra The perfluoroalkyl substituted pyrazoles, triazoles and tetrazoles are acids and can exist in the tautomeric forms H /^\ ^^\ /^^ . H H Rp-C C-R'^ ^z; ^F"\ J^~^'^ ^ N-N N-N '^ H H

PAGE 56

^6 The N-H stretching absorption of these three classes of perfluoroalkyl substituted heterocycles were similar. From Figures ^, 5> an
PAGE 57

4000 3000 100 r-^ ''! 2000 1500 CM' 1000 900 L_ 800 700 II==R -1 — I — I — I — I — \ — I — I — I — I — I — r6 7 8 9, 10 , 11 12 WAVELENGTH MICRONS 0.0 .6 .7 1.0 5 14 15 4000 3000 2000 100 I "" ' "" ' " ' ' I ' ' ' ' I ' ' 1500 CM' 1000 900 I I I 800 700 -^ -0.0 ^=' 7 8 9, 10 11 WAVELENGTH (MICRONS) 12 13 14 15 100 4000 3000 liniliii I I 2000 1500 , I CM' 1000 900 \ 1 I 800 700 I HR — N R = N , I 0.0 6 7 8 9, 10 11 WAVELENGTH (MICRONS) 12 13 14 15 47 Fig. 4. -Infrared spectra of: (A) 5-perfluoromethyltetrazole; (B) 5-perfluoropropyltetrazole; and (C) l,3-bis(5-tetrazolyl)perfluoropropane.

PAGE 58

4000 3000 100 2000 1500 CM' 48 1000 900 L_ 800 700 7 8 9 10 11 WAVELENGTH (MICRONS) 0.0 4000 3000 100 I n ,, I.,., I 2000 1500 CM' 1000 900 _L _L 800 700 L_ R /\ ' ^ w // ^ ^ N— N Mull Kel F all 7 8 9, 10 . 11 WAVELENGTH (MICRONS) 0.0 .2o3 o > .4Z n .5"' .6 .7 1.0 1.5 15 4000 100 3000 2000 1500 CM' 1000 900 800 700 I 7 8 9, 10 11 WAVELENGTH (MICRONS) 0.0 Fig, 5. -Infrared spectra of: (A) 3t3"bis (perfluoro-propyl)-l,2,4-triazole; (B) 5-perfliioropropyl-5-perfluoroetliyl-1,2,4— triazole; and (C) 5-perfluoroprop7l-5perf luoromethyl-l , 2 , 4-triazole .

PAGE 59

^9 4000 100 T 1 ; 1 ; ' r 7 8 9, 10 , 11 WAVELENGTH MICRONS Fig. 6. -Infrared spectra of: (A) J-perfluoropropyl4-,5-bis(perfluoromet]iyl)pyrazole; and (B) 3f^»5tris"rperfluoromethyl)pyrazole.

PAGE 60

50 between tlie two tetrazole rings and could not exchange readilyo Deuterated forms of tlie other heterocycles showed rapid exchange of hydrogen for deuterium. The replacement was noticeable spectroscopically after a few minutes exposure to the atmosphere. The cyclic C=N absorption of 5-perfluoropropyltetrazole (see Fig. ^B) appeared as a weak shoulder at 6.62 u. The N— H deformation absorption occurred at 6.75^. A band at 7»09 U in sodium 5-perfluoromethyltetrazole has 51 been attributed to 17=1^ absorption.'^ A similar band in 5-perfluoropropyltetrazole occurred at 7ol7^ and was probably due to the N=N group. In 5-perfluoropropyltetrazole-d the bands at 6.62 and 7.17^ retained their positions and relative intensities while the 6.75 /i band shifted into the C-F region, 7.8-9.0 u. The silver salt showed a cyclic C=1T absorption at 6.75^. Free 5-pei'i'luoromethyltetrazole showed a relatively intense band at 6.59 li (see Fig. ^A) . This has been 51 attributed to a C-GF^ stretching frequency.-^ The band at 7.16 u was due to N=N absorption and the N-H deformation band was at 6.75^. The deformation frequency appeared at 7.75 )i ill 5-perfluoromethyltetrazole-d. No band assignable to cyclic C=]!T stretching absorption occurred but in light of its position and intensity in 5-perfluoropropyltetrazole, it may be obscured by the 0— CF, absorption.

PAGE 61

51 The infrared spectra of l,3-bis(5-tetrazolyl)perfluoropropane was somewhat different from the 5-perfluoroalkyltetrazoles. Six distinct peaks were observed in the N-H stretching region while the monotetrazoles had only three. l,3-bis(5-Tetrazolyl)perfluoropropane appeared to have two bands due to N-H deformation at 6.89 and 7.09/. These disappeared on deuteration while a shoulder at 7.15/ remained as a weak peak. Consistent with the previous assignments, the shoulder at 7.15/ was assigned to N=N absorption. 1 , $-bis( 5-Tetrazolyl )perf luor opr opane-d2 showed two new bands at 6.59 and 7.70/. The latter is due to N-D deformation and the former, surprisingly, appeared in the cyclic C=N region. No bands in the 6.56.6 _;a region appeared in the spectra of 1 , 3bis C 5-tetra-' zolyl)perfluoropropane, even in concentrated mulls. The broad weak absorption at 5.75/ was due to the so-called immonium element. '''^ Since it did not occur in the monotetrazoles, it probably results from intramolecular association. The infrared spectra of the 3>5-bis(perfluoroalkyl)-l,2,A-triazoles (see Fig, 5) iiave already been discussed''"^ and only a few additional comments are in order. The infrared spectra of the deuterated triazoles indicated the bands in the 7.13-7.15 >i region were due to N-H

PAGE 62

52 deformation while the bands in the 6.86-6.90 u region were 19 due to the cyclic C=N element. As pointed out by Brown ^ this was an unusually long wavelength for C=IT stretching. The spectra of 3-perfluoropropyl-5-perfluoromethyl-l,2,^triazole (see Fig. 5C) shows a band at 6.50_^ not found in the other triazoles. This, as in 5-perfluoromethyltetrazole, was assigned to C-CF^, stretching. The infrared spectra of 3-perfluoropropyl-4,5bis (perfluoromethyl)pyrazole and 3 < ^ > ^tris (perf luor omethyl)pyrazole (see Fig. 6) were similar and will be discussed together. Bands at 6.61 and 6.58 u respectively were in the region previously assigned to C— CF^, stretching and were probably due to the C-CF, stretch in the pyrazoles. The N-H deformation absorptions occurred at 7.03 and 6.99^ respectively and the remaining two bands, 6.31 and 6.72^ in 3-perfluoropropyl-^, 3bis (perf luoromethyl)pyrazole , and 6.2^ and 6.78 u in 3 > ^ 1 3tris (perf luoromethyl )pyrazole were due to the ring unsaturation. These bands arose from C=C and C=N stretching and in view of the positions of C=N absorption noted in the other acidic azoles discussed above, it appeared reasonable that the bands at the longer wave lengths arose from C=1T and those at the shorter wave length from C=C stretching. However, in 3,5-dimethylpyrazole bands at 6.03 and 6.44^ were assigned to C=C stretching

PAGE 63

55 and a band at 6.28 u was assigned to C=>N stretching. Acid strengths The insertion of s£ hybridized nitrogen atoms into the pyrrole ring shows a progressive acid strengthening effect (see Table 8). This effect has been attributed to both the inductive attraction of electron density from the tt cloud by the nitrogen and its resonance stabilization of the anion formed by loss of the acidic hydrogen. '^ The /% /\ yyK -c c^ -C C -c cN— NH ^ N-N N— N © © effect of each S2^ nitrogen is approximately equal to that of a conjugated nitro group. The replacement of hydrogen bound to a carbon by a perfluoroalkyl group produced a similar and only slightly less pronounced effect on the acid strength. Prom Table 8, it is seen that introduction of an s£ nitrogen decreases the pKa by ^-5 units and the average effect per perfluoroalkyl group is in the 5-3 • 5 pKa units range. This illustrates the extreme electronegativity of the perfluoroalkyl group and the transmission of its effect to the

PAGE 64

5^

PAGE 65

CHAPTER III EXPERIMENTAL Source of Materials Perfluoroacetic, perfluoropropionic and perfluoro"butyric acids were purchased from Columbia Organic Chemicals Company, Columbia, South Carolina. Perfluoroglutaronitrile and perfluoroglutaryl chloride were supplied by Hooker Chemical Company, Niagara Falls, New York. Perfluorobutyne-2 was purchased from Peninsular ChemResearch, Gainesville, Florida, The sodium azide (Eastman purified grade) was used as received. Perfluoroacetonitrile, perfluoropropionitrile and perfluorobutyronitrile were prepared by dehydration of the respective perfluoroalkylamides as described by Swarts^ and Gilman and Jones. The Fisher anhydrous reagent grade acetonitrile used was further dried by refluxing it at least three hours over graniilar barium oxide and distilling it out of the barium oxide mixture. The apparatus was vented through a dry ice-cooled trap to protect against contamination of the acetonitrile by atmospheric moisture. 55

PAGE 66

56 Method of Molecular Veigh.t and pKa Determination of the 5-Perfluoroalkyltetrazoles and 3^^^5~ tris(Perfluoroalkyl)pyrazoles The molecular weights of the tetrazoles and pyrazoles were determined by titrating a weighed sample of the compound with OolOOO F NaOH. The samples were dissolved in methanolwater solution with the exception of 5-perfluoromethyltetrazole and l,3-bis(5-'tetrazolyl)perfluoropropane, which are water soluble and were titrated in water solution. During the titration, the pH was determined with a Beckman Model G pH meter. The pKa values of the compounds were read from the pH meter at the half-neutralized point where pH equals pKa. Preparation of ^-Perfluoroalkyltetrazoles 5-Perfluoromethyltetrazole 5-Perfluoromethyltetrazole was prepared by treating perfluoroacetonitrile with sodium azide in anhydrous 2 acetonitrile as described by Norris. The reaction was slow using acetonitrile dried with barium oxide but was effectively catalyzed by a trace of glacial acetic acid<, The tetrazole was obtained in 5^ per cent yield, b.p. 82-83** (5.1 mm,). Molecular weight calculated for C2HFjN^: 138. Pound by titration against 0.1000 N NaOH: 138; pKa, 1.70.

PAGE 67

57 5-Perfluoropropyltetrazole A one liter 3-iieck flask was equipped with, a gas inlet tube, magnetic stirrer and dry ice-cooled reflux condenser. The flask was charged with sodium azide, $3.75 g. (0.52 mole), dry methyl cellosolve, 120 ml, and glacial acetic acid, 150 ml. in that order. The flask was immersed in an ice bath and the sodium azide slurry was stirred vigorously. Perfluorobutyronitrile, 116 g, (0.595 mole), was added slowly beneath the surface of the liquid by means of the gas inlet tube at a rate adjusted to give a slow reflux of perfluorobutyronitrile o When addition of the nitrile was complete, the flask was heated at 70" for one hour. After cooling, the reaction mixture was poured slowly into 200 ml. of 20 per cent hydrochloric acid and stirred. The tetrazole separated as a slightly viscous oil. The aqueous layer was removed and the heavier oil layer was washed ten times with dilute hydrochloric acid, then once with water and dried under reduced pressure to give 137 go of crude 5-perfluoropropyltetrazole. The crude tetrazole was dissolved in I3OO ml. of acetone and added with stirring to a solution of 98 g. of silver nitrate in I3OO ml. of distilled water. The voluminous silver 5-perfluoropropyltetrazole formed rapidly and was removed by filtration, washed with water and acetone and dried at reduced pressure. This procedure removed a carbonylcontaining contaminant which cannot be removed by distillation.

PAGE 68

58 Free 5-perfluoropropyltetrazole was regenerated from the silver salt by addition of gaseous hydrogen chloride to a slurry of the salt in one liter of anhydrous diethyl ether. (Complete conversion of silver 5-perfluoropropyltetrazole to the free tetrazole and silver chloride is easily detected by a rapid settling of the precipitated silver chloride, when stirring is discontinued, to give a clear solution of tetrazole in ether. The silver tetrazole settles slowly and any cloudiness in the solution is indicative of incomplete conversion to the free tetrazole.) The ether solution of 5-perfluoropropyltetrazole was filtered to remove the silver chloride, concentrated under reduced pressure, then transferred to a distillation flask. A few grams of PpOc were added and the tetrazole was distilled from '^2^3' ^^^^^ 96.2 g. , 77.6 per cent, b.p. 60° (0.02 mmo), m.p. 31-5^°. Anal. Calcd. for C^HF^N^: C, 20.17; H, 0.41; F, 55.89; N, 23.55; mol. wt., 238. Found: C, 20.42; H, 0.55; F, 56.00; N, 23.75; mol. wt., 241 (by titration against OdOOO N NaOH) ; pKa, 1.73. The synthesis described above was carried out before Norris's paper on the synthesis of 5-perfluoromethyltetrazole appeared. Since Norris's method is the more convenient of the two methods, a slightly modified version was used to repre^are 5-perfluoropropyl tetrazole.

PAGE 69

59 Sodium azide, 3O0O g. (0.4-52 mole), was placed in a 3-neck flask equipped with, a magnetic stirrer, a dry icecooled reflux condenser, a dry nitrogen sweep to exclude air and a gas entry tube. Anhydrous acetonitrile, 300 ml., and glacial acetic acid, 2 ml., were added. Perfluorohutyronitrile, 95«0 g. (0.487 mole), was admitted under the surface of the liquid. The reaction was exothermic and the perfluorobutyronitrile addition rate was adjusted to maintain a reaction temperature of 55-60°. Toward the end of the addition, heat was applied to maintain the reaction temperature. Heating was continued for thirty minutes after the nitrile addition was complete. The solution was filtered while hot, concentrated to 250 mlo by boiling under slightly reduced pressure and then cooledo The first crop of white, deliquescent needles of sodium 5-perfluoropropyltetrazole was removed by filtration. Further concentration of the acetonitrile solution produced a second crop of crystals. The salt was dried under reduced pressure at 60° for 24hours and at 85° for 24hours. The total yield of sodium 5-perfluoropropyltetrazole was 105.5 g. j 85.5 per cent. The free tetrazole was generated by slurrying the sodium salt in anhydrous ether and bubbling in gaseous hydrogen chloride. The sodium chloride formed was removed by filtration and the ether solution was concentrated under reduced pressure. The 5-perfluoropropyltetrazole was

PAGE 70

60 distilled at 87-88" (5.5 nm.); yield, 85.0 g. , 75.^ per cent. TMs product did not solidify and was apparently not as pure as the tetrazole purified by means of the silver salt. l,5-bis(5-Tetrazolyl)perfluoropropane Sodium azide, 8.68 g. (0.155 mole), glacial acetic acid, 7.95 ml., and methylcellosolve, 50 ml., were placed in a 100 mlo flask equipped with a dry ice-cooled reflux condenser and nitrogen sweep. The contents of the flask were frozen; the flask was pumped free of air and perfluoroglutaronitrile, 10.0 g. (0.0^95 mole), was condensed in through the vacuum system. The reactants were warmed to 0** and stirred for thirty minutes. The mixture was then heated at 70° for fifteen minutes, refluxed for fifteen minutes, and cooled to room temperature. Acetone, 25 ml., and concentrated hydrochloric acid, 8 ml., were added to the reaction mixture and the solids formed were removed by filtration. An additional 50 ml. of acetone and 7 ml. of concentrated hydrochloric acid were added and the mixture was refiltered. The solvent was removed under reduced pressure and the crude product was redissolved in acetone and filtered. Most of the solvent was again removed under reduced pressure and the solid 1 , 5bis ( 5-tetrazolyl )perfluoropropane was precipitated by the addition of 60 ml. of benzene. The yield of crude product was 11.58 g. (81 per

PAGE 71

) 61 cent), m.p. 156-157.5°. The product was crystallized twice from perfluorobutyric acid to give a pure, white solid, m.po 159.8-160.5°. 1 , 5bis ( 5-Tetrazolyl )perf luoropropane decomposed slowly at 170*. Anal . Calcd. for C^E^^^^qI C, 20.83; H, 0.67; F, 39.58; N, 38.88; molo wt. , 288. Found: 0, 21.03; H, 0c8^; F, 36.69; N, 39o28; mol. wt. , 287 (hy titration against 0.1000 N NaOH) ; pKa, 1.70. Preparation of the Disodium Salt of Bitetrazole The procedure for the synthesis of the disodium salt 56 of bitetrazole was reported by Friederich-'^ in a patent and is given here since the patent may not be available. Sodium cyanide, 50 g. (Io04 mole), sodium azide, 65 S« (1.00 mole), and distilled water, 600 ml., were added to a two liter 3-n-eck flask fitted with a water-cooled reflux condenser, a stirrer, a 500 ml, addition funnel and a thermometer. The mixture was stirred vigorously and 55 g. of manganese dioxide was added with continued stirring. A solution consisting of sulfuric acid, 100 g. , glacial acetic acid, 80 g. , and distilled water, 200 g. , was prepared and added to the flask over a one hour period during which the temperature was maintained at 10-20° . The mixture was slowly heated to 100° over four

PAGE 72

62 hours and maintained at IOC for three hours and then cooled to room temperature. The black manganese bitetrazole was removed by filtration and washed with water. The bitetrazole was stirred with 55 g. of NapCO, dissolved in I50 ml. of HpO. Sufficient water was added to make a thin paste and the mixture was filtered hot through a fine filter to remove manganese carbonate. The aqueous solution was cooled to 0° for 24 hours and the first crop of hydrated disodium bitetrazole was removed by filtration. Three additional crops were obtained by further concentration and filtration. The four crops were combined and crystallized three times from water. Dehydration was effected at 80** under reduced pressure to yield 51.7 g. (56.8 per cent) of disodium bitetrazole. Hydrolytic Stability of 5-Perfluoropropyltetrazole 5-Perfluoropropyltetrazole, 0«50 g. (0.0021 mole), and 6 ml. 1 N NaOH solution (0.0063 eq.) were placed in a 10 mlo flask equipped with a reflux condenser. A gas buret was connected to the reflux condenser. The solution was refluxed five hours. No gas was evolved. 5-Perfluoropropyltetrazole, 0.4-2 g. (84 per cent), was recovered by acidification with concentrated hydrochloric acid. The procedure was repeated using 6 ml. of 1 N

PAGE 73

63 tiydrocliloric acid and O.5O g. of 5-perfluoropropyltetrazole. No gas was evolved. Tlie recovered tetrazole weighed 0.^7 g. (94 per cent). Oxidative Stability of 3-Perfluoropropyltetrazole Approximately 0.2 ml, of 5-perfluoropropyl tetrazole was dissolved in 1 N sodium hydroxide and a few drops of 4 per cent potassium permanganate were added. No discoloration of the permanganate occurred at room temperature nor within ten minutes at 100**, Approximately 0<,2 mlo of 5-perfluoropropyltetrazole was shaken with a potassium permanganate solution acidified with nitric acid. No reaction occurred at room temperature nor after 30 minutes at 100**, Synthesis of 2,5-his(Perfluoroalkyl)-l,5,^-oxadiazoles by Reaction of 3-Perfluoroalkyltetrazoles with Perfluoroacyl Chlorides Synthesis of 2,5-bis(perfluoropropyl)-l,5»^-oxadiazole 5-Perfluoropropyltetrazole, O.5I g. (0,0021 mole), was weighed in a previously constricted heavy-wall glass tube. The tube was connected to the vacuum system and pumped free of air. Perfluorobutyryl chloride, O.5O g.

PAGE 74

6^ (0.0022 mole), was condensed into the tube. The tube was sealed and heated at 90° for six hours and at 110° for 11.5 hours. The tube was cooled in liquid nitrogen and opened to the vacuum system; nitrogen, 0.00105 mole (50% of theory) , was recovered. The nitrogen was removed and the reaction tube was resealed and heated 21 hours at 125°. The tube was reopened and the volatile portion of the reaction mixture was transferred to the vacuum system. A rough separation in the system gave Oo35 S» (2^.9 per cent) of nearly pure 2,5-bis(perfluoropropyl)-l ,3»^-oxadiazole. Synthesis of 2-perfluoropropyl-5-perfluoromethyl--l,3^^oxadiazole 5-Perfluoropropyltetrazole, 2.00 g. (0.0084 mole), was weighed in a previously constricted heavy-wall glass tube. The tube was connected to the vacuum system and pumped free of airo Perfluoroacetyl chloride, 1,J>6 g. (0.010$ mole), was condensed into the tube; the tube was sealed and heated ten hours at 130° and 50 minutes at 150°. The tube was cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.24 g. (0.0084 mole), was re..^vered. The volatile oxadiazole was transferred through the vacuum system to another tube to separate it from the nonvolatile products. Hydrogen chloride, produced as a by-product, and residual perfluoroacetyl chloride were

PAGE 75

65 removed to the vacuum system, leaving 2.03 g. (8? per cent yield) of 2-perfluoropropyl-5-perfluoromettLyl-l,5,^oxadiazole, b.po 98°. The infrared spectra of the oxadiazole showed the weak, characteristic oxadiazole hands at 6, 30 and 6.58 u. Synthesis of 2 , 5-bis (perf luoromethyl )-l , 3 , ^-oxadiazole 5-Perfluoromethyltetrazole, lo^7 g. (0.010? mole), was weighed in a previously constricted heavy-wall glass tubeo The tube was connected to the vacuum system, pumped free of air, and perfluoroacetyl chloride, 1.69 g. (0.0128 mole), was condensed in the tube. The tube was sealed and heated at 95*' for 50 minutes and at 105** for 50 minutes. The temperature was then raised to 200° and maintained for 50 minutes. The tube was cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.010? mole, was recovered. The hydrogen chloride and residual perfluoroacetyl chloride were removed through the vacuum system. A small amount of a nonvolatile product was separated by transferring the oxadiazole to another tube by means of the vacuum system. The yield of 2, 5-bis(perfluoromethyl) -1,5, ^-oxadiazole was 2.0? go (95 per cent), b.p. 65°. The reported b.p. is 65**.

PAGE 76

66 Synthesis of 1^3bis (3-Perfluoroalkyl-1^3,4oxadiazolyl-2)perfluoropropanes Synthesis of 1 , 3"bi s ( 3-perf luoropropyl-1 , 3 1 ^-oxadiazolyl~2) perfluoropropane From ^-perfluoropropyltetrazole and perfluoroglutaryl chloride . 5-Perf luoropropyltetrazole, 3.06 g. (0.0129 mole), and perfluoroglutaryl chloride, 1.78 g. (0.006^3 mole), were weighed in a previously constricted heavy-wall glass tube. The tube was sealed and heated for 18 hours at 125-130**, then cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.292 g. (0.010^ mole, 81% of theory), was recovered. The crude liquid product was washed with a dilute potassium carbonate solution to dissolve any residual tetrazole or acid chloride. The desired product precipitated as a white solid and was rapidly filtered, washed with water and dried at reduced pressure. Sublimation at reduced pressure produced 1.16 g. (30% of theory) of 1,3bis( 5-perf luoropropyl-1 , 3 ,4-oxadiazolyl-2)perf luoropropane , m.p. 37.0-37.8°. From l,3-bis(5-tetrazoyl)perfluoropropane and perfluorobutyryl chloride . l,3-bis(5-Tetrazoyl)perfluoropropane, 5.00 g. (0.0173 mole), was weighed in a previously constricted heavy-wall glass tube. The tube was connected

PAGE 77

67 to the vacuum system and pumped free of air. Perfluorobutyryl cHloride, 80O8 go (0.05^5 mole), and 25 ml. of anhydrous methylene chloride were condensed into the tube.. The tube was sealed and heated at 125° for 6.5 hours with continuous agitation, then cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.818 g. (0,0292 mole, 85% of theory), was recovered. The methylene chloride was removed under reduced pressure, leaving 10.06 g. (95.^ per cent yield) of pale yellow 1 , 3-bis ( 5-perf luor opropyl-1 , 3 > ^oxadiazolyl-2)perfluoropropane, m.p. after crystallization from hexane, 36o2-37.5*'» Anal . Calcd. for C^5F2oV2°» ^^-^^^ ^' ^^'^^' N, 8.97. Found: C, 25.30; F, 60.73; N, 8.95. Synthesis of l,3-his(5-perfluoromethyl-l,3,^-oxadiazolyl-2)perfluoropropane l,3-bis(5-Tetrazolyl)perfluoropropane, 5.00 g. (0.0173 mole), was weighed in a heavy-wall glass tube. The tube was attached to the vacuum system and held under reduced pressure for 2.5 hours to remove traces of moisture. Anhydrous methylene chloride, 25 ml., and trifluoroacetyl chloride, 5.80 g. (0.0^^ mole), were condensed into the tube through the vacuum system. The reaction tube was sealed and heated with agitation at 125-135** for ten hours, then cooled in liquid nitrogen and opened to the vacuum system. Nitrogen evolution was found to be quantitative.

PAGE 78

68 All of the volatile materials were removed from tlie tube under reduced pressure to leave 7.^2 g. of white solid contaminated with a trace of yellow liquid. The solid product was sublimed under reduced pressure three times to give pure, white 1 , $bis ( 5-perf luoromethyl-l , 3 >^oxadiazolyl-2)perfluoropropane melting at 44,1-4^,8**. Synthesis of 5,5'-bis(Perfluoroalkyl)-2,2'bi-1 ^3i^-oxadiazoles Synthesis of 5,5'-bis(perfluoropropyl)-2,2'-bi-l,3,^oxadiazole From 3-perfluoropropyltetrazole and oxalyl chloride o 5-Perfluoropropyltetrazole, 2.50 go (0.0105 mole), was weighed in a heavy-wall glass tubeo The tube was attached to the vacuum system, pumped free of air, and 2 mlo of dry methylene chloride and 0.667 g. (0.0055 mole) of oxalyl chloride were condensed in through the vacuum system. The tube was sealed and heated at 100<» with agitation for 5.5 hours and then cooled in liquid nitrogen and opened to the vacuum system. Nitrogen evolution was found to be 90 per cent of the theoretical amount. The volatile materials were removed from the tube under reduced pressure and the remaining solid product crystallized twice from toluene and sublimed under reduced pressure to give lo57 S« (55 per cent

PAGE 79

69 yield) of 5,5'-bis(perfluoropropyl)-2,2'-bi-l,3,^oxadiazole melting at 165. 5-164»5°. From disodium "bitetrazole and perfluorobutyryl chloride . Dry disodium bitetrazole, 5.30 g. (0.0181 mole), was weighed into a previously constricted heavy-wall glass tube, and the tube was pumped free of air. Methylene chloride, 15 ml., and perfluorobutyryl chloride, 11.1 g. (0.0^78 mole), were condensed in the tube which was sealed and heated at 95** for four hours and at 125" ^or three hours with constant agitation. On opening the tube, approximately the theoretical amount of nitrogen was found to have formed. The crude product was recrystallized twice from toluene. Sublimation at reduced pressure gave 6.51 S» (75. Aper cent yield) of 5,5'-bis(perfluoropropyl)-2,2'bi-l,5,^-oxadiazole, m.p. 165.0-165.8**. Anal. Galcd. for C^o^lA°2^' 25.33; F, 56.12; N, 11.80; 0, 6.75o Found: C, 25.55; F, 56.00; N, 11.70. Synthesis of 5,5' -bis(perfluoroethyl)-2,2'-bi-l,5,^oxadiazole Dry disodium bitetrazole, 5.00 g. (0.0275 mole), was weighed into a previously constricted heavy-wall tube. The tube was attached to the vacuum system, pumped free of air and 25 ml. of dry methylene chloride and 12.0 g. (0.066 mole) of perfluoropropionyl chloride were condensed in the

PAGE 80

70 tube. Ttie tube was sealed, heated 15 hours at 100® and four hours at 125° with agitation and then cooled in liquid nitrogeno On opening the tube to the vacuum system, nitrogen was found to have evolved quantitatively. The reaction mixture was concentrated by removing part of the solvent under reduced pressure, then washed with water to dissolve the sodiiim chloride; the crude product was separated by filtration and washed with water. A crude yield of 9*62 g. (95o^ per cent) of 5>5' bis (perf luoroethyl)-2,2'-bi-l,5,A-oxadiazole, m.p. 182-18^. 8«> was obtained. The product was crystallized twice from toluene, m.p. 182-182.8°. Synthesis of 5 > 5 ' bis (perf luoromethyl ) -2 ,2 ' -bi-1 , 3 >^oxadiazole Dry disodium bitetrazole, 6.00 g. (0.053 mole), was weighed into a previously constricted heavy-wall glass tube. The tube was attached to the vacuum system, pumped free of air, and 25 ml. of methylene chloride and 90I g. (0,069 mole) of perfluoroacetyl chloride were condensed in the tube. The tube was sealed and heated at 100-105° for 1? hours and at 128° for three hours with constant agitation. A yield of 8^ per cent of the theoretical amount of nitrogen was obtained on opening the tube. The reaction mixture was mixed with water, the solid product removed by filtration

PAGE 81

71 and washed three times with water to remove the sodium chloride. A crude yield of 6.81 go (75.^ per cent) of 5,5'his(perfluoromethyl)-2,2'-bi-l,5,^-oxadiazole was obtained. After two crystallizations from toluene, the product melted at 195.5-19^.6''. Attempted Isolation of l(2)-Perfluorobutyryl~5perfluoropropyltetrazole The apparatus shown in Figure 7 was constructed for this reaction. The N-acetylated tetrazole, if formed, was expected to be easily hydrolyzed; therefore, this apparatus was designed to allow the reaction to be carried out under strictly anhydrous conditions, including filtration of the reaction mixture and solvent removal from the final product, without exposure of the product to the atmosphere. Preparation of silver 5-perfluoropropyl tetrazole . 5-Perfluoropropyltetrazole, 10.0 g. (0.0^1 mole), was dissolved in 20 mlo of acetonitrile and 7.15 g. (0.058^ mole) of silver nitrate was added to the solution. The mixture was warmed with stirring to bring it into solution. On cooling, silver 5-perfluoropropyl tetrazole precipitated and was removed by filtration. The filtrate was concentrated and a second crop of the salt was removed. The total yield of the silver salt was 15.2 g. or 91 per cent, based on the amount of tetrazole used.

PAGE 82

72 stopcock ^ receiving tube ir S 2V^0 me dium sintered glass filter \l k S 2V^0 "stopcock reaction tube ^J magnetic stirring bar Fig. 7. -Apparatus for attempted isolation of 1(2)perfluorobutyryl-5-perfluoropropyltetrazoleo

PAGE 83

73 Acylation of silver ^-perfluoropropyltetrazole with perfluorobutyryl chloride « Silver 5-perfluoropropyltetrazole, 2,50 g. (0.0072 mole), was placed in the reaction tube of the apparatus shown in Figure ?» The apparatus was attached to the vacuum system and the salt thoroughly dried under reduced pressure. The tube was then cooled with liquid nitrogen and 10 ml. of tetrahydrofuran, freshly distilled from barium oxide, was condensed into the tube after passing through a phosphorous pentoxide drying tube. The solvent was degassed and 3.5^ S» (0.0143 mole) of perfluorobutyryl chloride (100 per cent excess) was condensed into the tube. The mixture was allowed to warm slowly to room temperature with continuous stirring. The apparatus was then inverted and the pressure in the receiving tube slowly reduced to promote filtration of the product from the precipitated silver chloride. Nitrogen was generated in the reaction indicating the acylated tetrazole rearranged to 2,5-bis(perfluoropropyl)-l,3,^-oxadiazole under the reaction conditions. Solvent and volatile products, 13.31 g. » were removed from the receiving tube through the vacuum system and analyzed by gas-liquid chromatography using a 1/4 inch column two meters in length containing a silicone oil liquid phase on a diactomaceous support (Perkin-Elmer Column C). Helium, at 0.14 ml. per second, was used as a carrier gas

PAGE 84

7^ with a column temperature of 117". 2,3"bi3 (Perfluoropropyl)1,5,^-oxadiazole was retained 6.13 minutes and the tetrahydrofuran 10.88 minutes after injection. The determination of the yield 2 , 5his (perf luor opropyl ) -1 , 3 « A— oxadiazol e was based upon the integrated areas of a chromatogram of a known solution of 2,5-his(perfluoropropyl)-l,3,^-oxadiazole in tetrahydrofuran produced under the same conditions. This method indicated a 2.78 go or 9^.5 per cent yield of 2,5bis (perf luoropropyl ) -1 , 3 » 4-oxadiazole Reaction of 3-Perfluoropropyltetrazole with Phosgene Reaction with no added hydrogen chloride . 5-Perfluoropropyltetrazole, 5.00 g. (0.021 mole), was weighed in a previously constricted heavy-wall tube. The tube was attached to the vacuum system, pumped free of air, and 10 mlo of dry methylene chloride and 1.0^ g. (0.0105 mole) of phosgene were condensed in the tube. The tube was sealed and heated seven hours at 135"' » cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.0197 mole (93.^% of theory), was recovered. The tube contents were allowed to warm to room temperature and the solvent and volatile products were removed at reduced pressure, leaving the crude solid products in the tube. The crude solids were dissolved in anhydrous ether

PAGE 85

75 and separated in an alumina column with ether as an eluent. Crude N^N'-bisCperfluorobutyrimidoyl chloride) urea, 0.85 S« » m.po 98-103°, was the first product eluted. A white solid, probably 2-(perfluorobutyrimidoyl chloride)amino-5perfluoropropyl-l,3,^-oxadiazole) 1,^9 S, m.P» 86-88<» , was obtained after continued elution with ether and subsequently with methanol. The crude N,N'-bisCperfluorobutyrimidoyl chloride)urea was further purified by passage through a 22 inch alumina packed column with ether as solvent and eluent. After evaporation of the ether, the product was sublimed twice at reduced pressure to give 0.51 g. of pure N,N' -bis(perf luorobutyrimidoyl chloride) urea, m.p. 105-107''. Anal . Galcd. for G^E2^l2^m.'^n.^' C, 20.81; H, 0.39; CI, 15.68; F, 51.25; N, 10.79; 0, 5.08; mol. wt. , 519<. Found: C, 21.18; H, 0.87; 01, 13.85; F, 53.7^; N, 10.93; mol. wt., 527 (by freezing point depression in dioxane). The product assumed to be 2-(perfluorobutyrimidoyl chloride)amino-5-perfluoropropyl-l,3,^-oxadiazole, 0.2 g. , was placed in a heavy-wall tube, air in the tube was removed through the vacuum system and 0,5 g. anhydrous hydrogen chloride admitted with the tube cooled by liquid nitrogen. The tube was sealed and heated at 140* for seven hours, then opened to the vacuum system. The remaining hydrogen chloride was removed at reduced pressure leaving a light colored crude

PAGE 86

76 product in the tube. Vittiout removing th.e tube from the vacuum system, the crude product was heated and caused to sublime to the upper portion of the tube. The infrared spectra of the sublimate showed it to be nearly pure N,1T'bis (perfluorobutyrimidoyl chloride)urea. The yield was not determined. Reaction in the presence of added hydrogen chloride . 5-Perfluoropropyltetrazole, 1.00 g. (.00^2 mole), phosgene, 0.208 g. (0.0021 mole), hydrogen chloride, O.507 g. (0.008^ mole) and 1.5 ml. of anhydrous methylene chloride were sealed in a heavy-wall tube and heated ten hours at 95" and seven hours at 160°. The tube was cooled in liquid nitrogen and opened to the vacuum system and 0.0052 mole of nitrogen was recovered. The tube was warmed to room temperature and the solvent and residual hydrogen chloride removed under reduced pressure. The infrared spectra of the crude product, a pale amber solid mixed with a small amount of oil, showed it to be almost pure N,K'-bis(perfluorobutyrimidoyl chloride) urea. It was purified by eluting with ether from an alumina column to give 0.601 g. (55.2 per cent yield) of pure N,N'-bis(perf luorobutyrimidoyl chloride) urea, m.p. 105-106.5°. Reaction of 2 , 5bis (perf luoropr opyl ) -1 , 3 > ^-oxadiazole with hydrogen chloride . 3 > 5-bis (Perf luoropr opyl) -1,5 >^oxadiazole, 0.50 g. was weighed in a heavy-wall tube, the

PAGE 87

11 tube connected to the vacuum system and pumped free of air. Hydrogen chloride, 0.75 S» was condensed in the tube and the tube sealed and heated at 210° for 2^ hours and 225"» for 20 hours. The tube was cooled in liquid nitrogen and opened to the vacuum system. The residual hydrogen chloride and other volatile products were removed under reduced pressure leaving 0,05 g. of a white solid product, presumably N-(perfluorobutyrimidoyl chloride)perfluorobutyramide in the tube, m.p, 110-112°, The product was found to be stable to water at room temperature but is attacked by aqueous base. Reaction of ^--Perfluoropropyltetrazole with Perfluorobutyronitrile Under Mild Conditions; Synthesis of 3,5bis ( Per f luor opr opyl ) -1 , 2 , ^-triaz ol e Attempted uncatalyzed reaction of 5-perfluoropropyltetrazole with perfluorobutyronitrile 5-Perfluoropropyltetrazole, 1,32 g. (0,00555 mole), was weighed in a previously constricted heavy-wall glass tube which was then attached to the vacuum system and pumped free of air. After 5 ml, of dry methylene chloride and 1,08 g. (0.00555 mole) of perfluorobutyronitrile were added through the vacuum system, the tube was sealed and heated 15.5 hours at 100°, When the reaction tube was

PAGE 88

78 cooled in liquid nitrogen and opened to the vacuum system, no nitrogen was found and the tetrazole and nitrile were recovered unreacted. Boron trifluoride etherate catalyzed reaction of 3-perfluoropropyltetrazole with perfluorobutyronitrile 5-Perfluoropropyltetrazole, 0.56 g. (0.0028 mole), and 0.60 mlo of boron trifluoride etherate were placed in a previously constricted heavy-wall glass tube. The tube was attached to the vacuum system, cooled in liquid nitrogen, pumped free of air and perfluorobutyronitrile, 2.71 g. (0.0139 mole), was condensed into the tube. The tube was sealed and heated five hours at 65°, two hours at 100°, two hours at 150° and then cooled in liquid nitrogen and opened to the vacuum system. Nitrogen, 0.03^5 g. (0.00125 mole), or 4^1per cent of the theoretical amount was recovered. The tube was warmed to room temperature and the remaining boron trifluoride-ether complex was pumped out. The crude product, 1.12 g. , was analyzed by gas-liquid chromatography using the Perkin-Elmer Column "C" at 145° with helium as a carrier gas flowing at 1.15 ml. per second. 5,5-bis(Perfluoropropyl)-l,2,^-triazole in the product was identified by comparison of the chromatogram to one produced by a known sample of the triazole under identical conditions. The retention time of the triazole was 4,05 minutes. The

PAGE 89

79 yield, estimated from the integrated area under th.e curves of the chromatogram, was 0.13 g. or 11.6 per cent. Gaseous boron trifluoride catalyzed reaction of ^-perfluoropropyltetrazole with perfluorobutyronitrile Formation of the boron trifluoride-tetrazole adduct . 5-Perfluoropropyltetrazole, Oo50 g» (0.0021 mole), was weighed in a doubly constricted, heavy-wall glass tube. The tube was connected to the vacuum system and pumped free of air. Boron trifluoride gas, Ool^5 go (0.0021 mole), was measured in the vacuum system at 155 nim. The tube containing the tetrazole was opened to the vacuum system containing the boron trifluoride, No reaction or solution of the boron trifluoride in the tetrazole was apparent within 15 minutes. The boron trifluoride was then condensed into the tube and the tube was sealed. On warming the tube to room temperature, a white, solid, boron trifluoride-tetrazole adduct slowly formed. The tube was heated six hours at 60**. Some melting, but no other apparent reaction had occurred, therefore, heating was continued for 2.5 hours at 90°. The tube was then cooled in liquid nitrogen and opened to the vacuum system; no nitrogen had evolved. The tube was resealed and heated two hours at 135** and then rapidly cooled to liquid nitrogen temperature. Considerable gas condensed on the tube wall. When warmed to room temperature, the product was liquid but slowly

PAGE 90

80 reacted to re-form tlie original solid adduct. The tube was re-cooled in liquid nitrogen and opened to tlie vacuum system, No nitrogen was found. When warmed to room temperature at low pressure, the adduct slowly liberated boron trifluoride, leaving the relatively nonvolatile 5-perfluoropropyltetrazole unchanged. Reaction of ^-perfluoropropyltetrazole with perfluorobutyronitrile in the presence of 10 mole per cent boron trifluoride . 5-Perfluoropropyltetrazole, 0.50 g. (0.0021 mole), was weighed in a heavy-wall glass tube; the tube was connected to the vacuum system and pumped free of air. Boron trifluoride, 0.0143 g. (0.00021 mole), and perfluorobutyronitrile, 0.84 g. (0.0042 mole), were condensed in the tube and the tube was sealed and heated six hours at 100°. The tube was opened to the vacuum system and the residual nitrile pumped off. The crude product was dissolved in ether and extracted with 1 N sodium hydroxide solution to separate the triazole product and remaining tetrazole from any nonacidic products. The alkaline solution was acidified with concentrated hydrochloric acid to give 0.35 g» of a colorless oil which was analyzed by gas-liquid chromatography using the Perkin-Elmer Column "C" at 122° with helium as a carrier gas flowing at 1.36 ml, per second. The major components of the oil were identified as 5,5-bis(perfluoropropyl)-l,2,4-triazole, 23.7 per cent, and 3-perfluoropropyl tetrazole, 63.0 per cent, by comparison of the

PAGE 91

81 chxomatograni to one produced by autlientic samples under the same conditions. The retention time of the triazole was 9o8^ min. , and the tetrazole 32.80 min. after air. The yield of 3>5-his(perfluoropropyl)~l, 2, ^-triazole was calculated to be 9.8 per cent. Reaction of 3-perfluoropropyltetrazole with perfluorobutyronitrile in the presence of one mole per cent boron trifluoride . The procedure described above was repeated using 2.00 g. (0.008^ mole) of ^-perfluoropropyltetrazole, 3.28 g. (0.0168 mole) of perfluorobutyronitrile and 0,00562 g. (8.^ x 10"^ moles) of boron trifluoride. The reaction tube was heated 11 hours at 100®, then opened to the vacuum system and 25 per cent of the theoretical amount of nitrogen was recovered. The crude product, 2.12 g. analyzed as described above, contained a 6.2 per cent yield of 3»5-bis(perfluoropropyl)-l ,2,^triazole. Reaction of 5-perfluoropropyltetrazole with perfluorobutyronitrile in the presence of gaseous hydrogen chloride Heating 5-perfluoropropyltetrazole vjith excess hydrogen chloride . 5-Perfluoropropyltetrazole , 0.50 g. (0.0021 mole), was weighed in a constricted heavy-wall glass tube, the tube was connected to the vacuum system and pumped free of air. Anhydrous hydrogen chloride, 1.89 g, (0.051? mole), was condensed in the tube and the tube was sealed

PAGE 92

82 and heated 14 hours at 125°, The tube was cooled in liquid nitrogen and opened to the vacuum system. No nitrogen was liberated. The tube was warmed to room temperature and pumped free of hydrogen chloride. The residual liquid was identified as 5-perfluoropropyltetrazole by its infrared spectrumo Reaction of 3-perf luoropropyltetrazole with perfluorobutyronitrile in the presence of gaseous hydrogen chloride. Three experiments in which 5-perfluoropropyltetrazole reacted with perfluorobutyronitrile in the presence of gaseous hydrogen chloride were performed under identical conditions with the exception of the amount of hydrogen chloride present. Three constricted heavy-wall glass tubes of approximately 50 ml. capacity were prepared and 0.50 g. (0.0021 mole) of 5-perfluoropropyltetrazole was placed in each tube. Each tube was then connected to the vacuum system, pumped free of air and 1.28 g. (0.0063 mole) of perfluorobutyronitrile was added. The following amounts of gaseous hydrogen chloride were placed in the tubes: tube 1, 0.024 g. (0.00063 mole); tube 2, 0.48 g. (0.0126 mole); tube 3, 0.96 g. (.0252 mole). The tubes were sealed and heated 11 hours at 135** with constant agitation. The tubes were then immersed in liquid nitrogen and opened to the vacuum system. The following amounts of free nitrogen were recovered from the tubes (yields were based on

PAGE 93

85 the evolution of one mole of nitrogen per mole of tetrazole): tube 1, Oo0289 g. (0.00103 mole or ^9.2 per cent); tube 2, 0.0491 g. (0.00175 mole or 85.^ per cent); tube 5, 0.0595 g. (0.00212 mole or 100 per cent). The tubes were pumped free of hydrogen chloride leaving the crude products-mixtures of white solids and colorless liquids-in the tubes. The solid portion of the crude product, mainly 3,5-bis(perfluoropropyl)-l ,2,4triazole, was separated from the liquid by quickly mixing the crude product with methylene chloride and filtering immediately. The residue was dissolved in ether, filtered and the ether allowed to evaporate. From tubes 2 and $ were obtained 0.18 g. and 0«50 g. respectively of almost pure 3 , 5-bis (perf luoropropyl ) -1 , 2 , 4-triazole ; no triazole was separated with this procedure from tube 1. The filtrates from the above filtrations were concentrated and analyzed by gas-liquid chromatography and estimated to contain 0.20, 0.11, and Oo055 g. of 5,5-bis(perfluoropropyl)-l,2,4-triazole from tubes 1, 2, and 3, respectively. The total yields were: tube 1, 0.20 g. or 23.5 per cent; tube 2, 0.29 g. or 35.0 per cent; tube 3, 0.36 g. or 42.2 per cento

PAGE 94

84 Attempted Reaction of 3-Perfluoropropyltetrazole with. Perfluoro"butyne-2 in the Presence of Gaseous Hydro p; en Chloride 5-Perfluoropropyltetrazole, 0.50 g. (0.0021 mole), was weighed in a heavy -wall glass tube, the tube was attached to the vacuum system and pumped free of air, Perfluorobutyne-2, 1,02 g, (0,0063 mole), and hydrogen chloride, 0,96 g, (0,0252 mole), were condensed into the tube. The tube was sealed, heated for 11 hours at 155'' > then cooled in liquid nitrogen and opened to the vacuum system. No nitrogen was found in the tube and the tetrazole was recovered after pumping off the hydrogen chloride and perfluorobutyne-2. Reactions of 5-Perfluoroalkyltetra2oles Under Thermal Ring Opening Conditions Synthesis of 3,5-bis(perfluoropropyl)-l,2,4-triazole in a static system 5-Perfluoropropyltetrazole, 1.00 g. (O.OOA-2 mole), was weighed in a previously constricted heavy-wall glass tube, the tube attached to the vacuum system and pumped free of air. Perfluorobutyronitrile, 1.64 g. (0.0042 mole), was condensed into the tube sind the tube was sealed and heated at 200*> for one hour and 240" for 1.5 hours. The

PAGE 95

85 tube was then cooled in liquid nitrogen, opened to the vacuum system and Ool^7 g. (0.0055 mole) of nitrogen was recovered. The tube was warmed to room temperature and the volatile materials removed under reduced pressure. The dark viscous residue in the tube was dissolved in diethyl ether and removed from the tube. After removal of the solvent, the oil was heated and 0.28 g. of yellow crystals sublimed from it. After resubliming at 80** (75 mm.), the slightly yellow product melted at 85-97° and was identified as 5,5-bi£(perfluoropropyl)-l,2,^-triazole (crude yield 16,5 per cent). The crude product was further purified by dissolving it in aqueous potassium carbonate and extracting the solution with diethyl ether. The alkaline solution was then acidified with hydrochloric acid and the precipitated triazole removed by filtration. After resubliming at atmospheric pressure, the melting point was 103-105**. The 19 reported melting point is 110-111°. Synthesis of 5,5-bis(perfluoroalkyl)-l,2,^-triazoles in a flow reactor The following series of reactions between 5-perfluoroalkyltetrazoles and perfluoroalkylnitriles was cairried out in the flow reactor shown in Figure 8o The nitrogen was dried by passage through a copper helix immersed in a dry ice-acetone bath.

PAGE 96

CO •H m •p (Q (1> rH O ISI rt •H -P I •dCM rH 03 O fH o :3 rH f^ Q) ft CO •H ,Q I lA o
PAGE 97

87 Synthesis of 3,5-his(perfluoroprop7l)-l ,2,^-triazole. « 5-Perfluoropropyltetrazole, 0.45 g, , was placed in the vaporization pot and the pot attached to the inlet of the pyrolysis apparatus. The furnace was preheated to 265** and the hot tube flushed simultaneously with perf luorobutyronitrile and nitrogen to insure complete removal of air and the presence of the nitrile in the hot tube when the tetrazole entered the hot zone. The tetrazole reservoir was then heated to 158° to begin vaporization of the tetrazole and the nitrogen and perfluorobutyronitrile flows were adjusted to 0,04 liter per minute and 4.0 g. per hour, respectively. After 20 minutes the furnace temperature was raised from 255° to 271° to affect more complete decomposition of the tetrazole. At the same time, the reservoir temperature was increased to 155** to increase the vaporization rate of the tetrazole and the nitrile flow rate was increased to 6.5 S» P®3? hour. After 70 minutes of operation of the apparatus, the tetrazole reservoir temperature was raised to IbO* and the furnace temperature maintained at 275-275°. The tetrazole had completely vaporized and the reaction was completed after 110 minutes. The nitrile flow and furnace temperature were maintained for an additional 15 minutes and the nitrogen sweep was maintained for 20 minutes.

PAGE 98

88 Th,e product was removed from the exit tube of the pyrolysis apparatus with diethyl ether and the ether removed by evaporation. Sublimation of the product under reduced pressure produced 0«^76 g. of a white solid, melting at 68-81 **, contaminated with a small amount of liquid. An additional OolO go of solid product was recovered from the dry ice trap to give a total crude yield of 0.486 g. A portion of the crude product was further purified by absorbing the liquid impurity on filter paper, then subliming the dried solid at reduced pressure, to give a product melting at 109-110° whose infrared spectra was identical to that of a known sample of 3 , 5bis (perf luoropropyl )-l ,2 ,4triazole. Assuming the crude product was at least 90 per cent pure, a yield of 0o437 g. or 57 per cent was realized. Synthesis of 3-pei'fluoropropyl-5-perfluoroethyl1 v2,4— triazole . Using the procedure described above, 0.^9 g« of 5-perfluoropropyltetrazole was vaporized into the hot tube in approximately one hour. The furnace was maintained at 291-296°, the tetrazole reservoir at 152-157** and the perfluoropropionitrile flow rate was 5.5 g. per hour. The solid product was washed from the exit tube with ethyl ether and the ether evaporated to give a crude yield of 0.39 g., m.p. 75-80°.

PAGE 99

89 The crude product was sublimed twice at reduced pressure to give O.3I g. (4-2 per cent yield) of pure 3-perfluoropropyl-5-perfluoroetliyl-l,2,4-triazole, m.p. 79.8-81,0®. Tile melting point and infrared spectra of this triazole were identical with those of an authentic sample produced from 2-perfluoropropyl-5-perfluoroethyl-l ,3,4oxadiazole using the procedure described by Brown and 19 Cheng. Synthesis of 3-peJ^fluoropropyl-3-perfluoromethyl1,2,4triazole . Using the procedure described for 3,5bis(perfluoropropyl)-l,2,4-triazole, 0.59 g. of 5-perfluoromethyltetrazole was vaporized into the hot tube in 65 minutes. The furnace was maintained at 300°, the tetrazole reservoir at 14-9° and the perfluorobutyronitrile flow rate was 7.5 g. per hour. The amber-colored, liquid product was washed from the exit tube with ethyl ether and the ether evaporated to give a crude yield of 0.55 g. No suitable method of purification of the crude product was found, therefore the product was identified and the yield estimated by gas-liquid chromatographic analysis. The retention times of 3-perfluoropropyl-5-perfluoromethyl-l, 2, 4— triazole was 12.52 minutes using the Perkin-Elmer Column "C" at 153° with helium as a carrier gas flowing at 0.19 ml. per second and was established with an authentic sample prepared from 2-perfluoropropyl-5-perfluoromethyl-l,3,4-oxadiazole using

PAGE 100

90 19 the procedure described by Brown and Clieng, From this data, a yield of 0.25 g. or 35 per cent was calculated. The pure triazole melts at 63-65° and is hygroscopic. Synthesis of 3 ^ 5bis (perf luor omethyl ) -1 , 2 , 4— triazole . Using the procedure described for 3 i 5~ bis (perf luoropropyl )1,2, ^-triazole, loOO g. of 5-perfluoromethyltetrazole was vaporized into the hot tube in 67 minutes. The furnace was maintained at 299°, the tetrazole reservoir at 150** and the perfluoroacetonitrile flow rate v/as 6.5 g. per hour. The amber-colored, liquid product was washed from the exit line with ethyl ether and the ether evaporated to give a crude yield of 0o78 g. No suitable method of purification of the crude product was found, therefore the product was identified and the yield estimated by gas-liquid chromatographic analysis. The retention time of 3 »5-bis (perf luor omethyl )1,2,4-triazole was 15.83 minutes using the Perkin-Elmer Column "C" at 140° with helium as a carrier gas flowing at 0,20 ml, per second and was established with an authentic sample prepared from 3,5-bis(perfluoromethyl)-l ,3,4— 19 oxadiazole. From this data, a yield of 0,4-5 g. or 30 per cent was calculated. Synthesis of 3,'^,5-tris(perfluoroalkyl)pyrazoles by reaction of perfluorobutyne-2 with 5-perfiuoroalkyltetrazoles The flow reactor diagrammed in Figure 8, vas used in

PAGE 101

91 the pyrazole syntheses. Perfluorobutyne-2 was admitted through the nitrile entry tube and ice was used to cool the exit tube to facilitate condensation of the product. Since the pyrazoles were prepared in larger quantities than the fcriazoles, an ice cooled trap was placed between the exit line and the dry ice trap to collect the products. Synthesis of 3-perfluoropropyl-A-,5-bi£(perfluorome thyl ) pyr az ol e . 5-Perfluoropropyltetrazole, 10.05 g. (0,0420 mole), was placed in the vaporization pot and the pot attached to the inlet of the pyrolysis apparatus. The furnace was preheated to 300** and the hot tube flushed simultaneously with perfluorobutyne-2 and dry nitrogen to insure complete removal of air and the presence of the butyne in the hot tube v;hen the tetrazole entered the reaction zone. After thorough flushing, the butyne and nitrogen flow rates were adjusted to 6,0 g. per hour and 0,05 liter per minute, respectively. The tetrazole reservoir was then heated to 155-161*, These reaction conditions were maintained for the seven hours necessary to completely vaporize the tetrazole, A total of 42 g, (0.259 mole) of perfluorobutyne-2 was used in the reaction. The mole ratio of tetrazole-butyne was lo0-6,08. The crude, amber-colored liquid was drained from the exit line and traps and distilled to give 10,49 g. (6? per

PAGE 102

92 cent yield), of 3-perfluoropropyl-^,5-Ms(perfluoroiiiethyl)pyrazole boiling at 178°. The product was pale red due to a trace of $ , 6bis (perf luoropropyl ) -1 , 2 , ^ , 3-t etrazine and was analyzed by gas-liquid chromatography using the PerkinSlmer Column "C" at 152° with helium as a cairrier gas flowing at 0,52 mlo per second. The retention time of the pyrazole was 10.88 minutes and the pyrazole was found to be 99 per cent chromatographically pure. The trace of red sym . -tetrazine was removed by dissolving the entire sample in water-saturated ether and passing the solution through a column packed with alumina (pH-^ grade). The pale yellow solution which passed through the column was redistilled at 104-106° (45 mm,). Anal. Calcd, for CqEF^^^^: C, 25.81; H, 0,27; F, 66,40; N, 7.53; mol. wt,, 372, Found: C, 26,09; H, 0o38; F, 66,29; N, 7.58; molo wt. , 380 (by titration in methanol -water solution against 0.1000 N NaOH) , pKa, 5.12. Synthesis of 3»^»5-tris(perfluoromethyl)pyrazole. 5-Perfluoromethyltetrazole, 10.00 g, (0.0725 mole), was placed in the vaporization pot of the apparatus described above and the pot attached to the inlet tube. The furnace was preheated to 305° and the perfluorobutyne-2 and dry nitrogen flow rates were adjusted to 6,5 g. per hour and 0,05 liter per minute, respectively. The tetrazole reservoir was then heated to 150-155°. These conditions were maintained

PAGE 103

95 for approximately seven hours; after this time the reservoir was heated to 165° to increase the rate of tetrazole vaporization. A detonation occurred in the hot tuhe, presumably due to an excessive amount of tetrazole present in it. Approximately loO g. of the tetrazole remained in the reservoir. The dark, crude product, 8.75 go, a mixture of solid and liquid, was washed from the exit line with ethyl ether and the ether allowed to evaporate. The crude product was sublimed under reduced pressure and then recrystallized once from hexane and twice from reagent grade petroleum ether to give 5.51 g. of fine white needles, m.p. 69-70*. A volatile contaminant remained after the r eery st all ization and was removed by continued pumping on the sample; during this procedure part of the product sublimed. 5>^,5tris (Perf luoromethyl )pyrazole , ^.72 g. (26.6 per cent yield), m.p. 700 5-72.5" was obtained. Anal . Galcd. for C^EF^'S^' ^» 26.^7; H, 0.57; P, 62.87; N, 10o29; mol. wt., 272c Found: C, 26.25; H, 0.42; N, 10o55; mol. wt. , 27^ (by titration in methanolwater against 0.1000 N NaOH) , pKa, 5.15. Reactions of perfluorobutyronitrilimine on a cold finger Apparatus and procedure . The following series of reactions was carried out in the apparatus shown in Figure 9.

PAGE 104

9^ cold finger liquid nitrogen nitrile, butyne or water to vacuum pump glass wool 2. ^5/50 furnace jacket furnace 2-1/2" long X 11 mm, packed with 1/16" glass helices S 1V20 10 ml. tetrazole reservoir '^ heating mantle thermocouple well Fig. 9. -Apparatus for preparation and reaction of perfluorobutyronitrilimine .

PAGE 105

95 This apparatus was constructed to enable the reactions of the perfluorobutyronitrilimine fragment, generated by thermolysis of 5-perfluoropropyltetrazole, to be carried out under mild conditions at a point removed from its place of generation. The apparatus incorporates a furnace, capable of reaching 600 «* , so arranged that the emerging thermolysis products will be frozen on a liquid nitrogen cooled surface after traveling no more than 1/2 inch from the exit of the furnace. The apparatus was evacuated, then the material to be reacted with the nitrilimine fragment was evenly condensed on the cold finger in a narrow band by allowing it to deposit slowly while rotating the cold finger in its standard taper joint. The tetrazole was heated and distilled into the hot zone at the lowest pressure obtainable by the vacuum pump. The thermolysis products emerging from the hot zone were condensed evenly on the material on the cold finger vjith which they were to react and were then covered with another layer of that material. This sandwiching of the nitrilimine between two layers of the material with which it was to be reacted produced maximum surface contact of the reactants. The reactions were effected by allowing the cold finger to warm slowly to room temperature. Reaction of perfluorobutyronitrilimine with perfluorobutyronitrile . Perfluorobutyronitrile, 0.4-1 g. (0.0021

PAGE 106

96 mole), was condensed on the cold finger of the apparatus shown in Figure 9. 5-Perfluoropropyltetrazole, 0,50 g, (0,0021 mole), previously placed in the reservoir, was heated to 60° and distilled through the hot zone held at 589°. The thermolysis products condensed on top of the nitrile. Uhen all the tetrazole was distilled from the reservoir, an additional 0,^-1 g, (0.0021 mole) of perfluorobutyronitrile was condensed on top of the thermolysis products. The condensed materials were allowed to warm slowly and stand at room temperature for several minutes. The gaseous portion of the product was then removed to the vacuum system and the crude, relatively nonvolatile product, 0,11 g, , was washed out with ether and examined by gas-liquid chromatography using the Perkin-Elmer Column "C" at 1^7° with helium as a carrier flowing at 1.21 ml. per second, 5»5his(Perfluoropropyl)-l ,2,^-triazole, with a retention time of 5.98 minutes, was found to be present as one of many components of the crude mixture. The thermolysis was repeated at ^55° with 0,70 g. of 5-perfluoropropyltetrazole; no nitrile was used in this experiment. After removal of the gaseous products, the remaining crude oil was examined by gas-liquid chromatography and found to contain no 5 ^ 5bis (perf luoropropyl ) -1 , 2 ,^triazole.

PAGE 107

97 Reaction of perfluorobutyronitrilimine v/itb. perfluoroacetonitrile . The reaction described above with perfluorobutyronitrile was repeated replacing the butyronitrile with 0o20 g. (0.0021 mole) of perfluoroacetonitrile. The hot zone was maintained at W^° during the thermolysis. The crude oil produced was examined by gas-liquid chromatography using the Perkin-Elmer Column "C" at 135** with helium as a carrier gas flowing at lo27 mlp per secondo 3-Perfluoropropyl-5perfluoromethyl-l,2,^-triazole with a retention time of 5.18 minutes, was found to be present as a minor component of the crude mixture. Reaction of perfluorobutyronitrilimine with perfluorobutyne-2 . The reaction described above with perfluorobutyronitrile was repeated replacing the perfluorobutyronitrile with 0.34 g. (0.0021 mole) of perfluorobutyne-2. The hot zone was held at 590-400° during the thermolysis. The crude oil produced was analyzed by gas-liquid chromatography using the Perkin-Slmer Column "C" at 122° with helium as a carrier gas flowing at 3.60 ml. per second, 3-Perfluoropropyl-4,5-bis(perfluoromethyl)pyrazole, with a retention time of 5.55 minutes, was found to be present as a minor component of the crude mixture. Reactio n of perfluorobutyronitrilimine with water . The reaction described above with perfluorobutyronitrile was repeated replacing the nitrile with 0.04 g. (0.0022 mole) of

PAGE 108

98 water. Ttie hot zone was held at 395** during the thermolysis. Analysis of the crude product, 0.18 g, , was obtained using the Perkin-Elmer Column W (which employs Carbowax as a liquid phase on a Teflon support) at 85"* with helium as a carrier gas flowing at 2.57 ml, per second, Perfluorobutyrhydrazide, with a retention time of 4-,$^ minutes, was found to be present in the crude product. Thermolysis of 3-perfluoropropyltetrazole with no acceptor molecule present A total of ^6 g. of 5-perfluoropropyltetrazole, in five separate portions, was pyrolized at 390-595° by distilling it, at the lowest obtainable vacuum pump pressure, from a reservoir at about 100° into a 12 inch heated zone constructed of 12 mm. Pyrex tubing packed with 1/8 inch single turn Pyrex helices. The products were collected in a liquid nitrogen cooled trap. The volatile portion of the products was transferred, via the vacuum system, to another trap and the remaining nonvolatile portion, a small amount of amber-colored solid and oil, was washed out with ether and saved. The volatile material was distilled twice through a 12 inch vacuumjacketed column packed with glass helices and gave 15*9 g. of a colorless gas boiling at 1^,4-1^,8°. Its infrared spectrum (see Figure 3) showed an absorption maximum at 5o66ji due to carbon-carbon double bond stretching but lacked

PAGE 109

99 a maximum at 1$.3^ which usually appears in compounds having a C-yFrpS^oxx^. The gas absorbed bromine from a caLrbon tetrachloride solution in sunlight but not in the dark and readily decolorized potassium permanganate solution. The product was identified as 2-H-perfluorobutene-l. Anal. Calcd. for C^HFr^: C, 26.23; H, 0.55; F, 72.68; mol. wt., 185. Found: C, 26.02; H, 0.7^; F, 72.88; mol. wt. (Dumas), 186. After the ether was removed, the nonvolatile residue from the pyrolysis was crystallized from a mixture of equal parts of chloroform and carbon tetrachloride, then recrystallized three times from chloroform, sublimed at reduced pressure and crystallized from benzene to give 0.31 g. of pure, yellow, 3 , 6bi s ( perf luoropr opyl ) -1 , 2-dihydr o-l ,2,^,5tetrazine, m.po 117-118°. Ultraviolet absorption maximum in isorpopyl alcohol occurred at 232 m^; log = 3.5^. Anal. Calcd. for CqH2F-l^N^: C, 22.86; H, 0,^8; F, 63.33; N, 13.33. Found; 0, 22.90; H, 0.59; F, 63.39; N, 13.19. Deuteration of the Perfluoroalkyl Substituted Tetrazoles, Triazoles sind Pyrazoles The acidic heterocycles were deuterated by allowing the acidic hydrogen to exchsuige with deuterium in DpO

PAGE 110

100 solutions. 5-Pei'fluoromethyltetrazole and l,3"bis (5-"betrazolyl)-perfluoropropane are water soluble and were deuterated by dissolving 0.1 go samples, in 4 inch, test tubes, in 0.5 ml. of D2O. The tubes were degassed and allowed to stand, protected from the atmosphere, for one hour. The DpO was removed under reduced pressure and the infrared spectra of the product taken immediately after removing the tube from the vacuum system. 5-Perfluoropropyltetrazole, 3»5-bis(perfluoropropyl)1,2, 4-triazole , 5-pei*f luor opr opyl-5-perf luor omethyl-1 ,2,4triazole, 3-perfluoropropyl-4, 5bis (perfluoromethyl)pyrazole and 3,4,5-tris(perfluoromethyl)pyrazole are all water insoluble and were deuterated in a mixture of 1.0 ml « of anhydrous dioxane and 0.5 nil. of DpO* using the method described for the water soluble tetrazoles. The infrared spectra of the products were taken immediately after removing the tube from the vacuum system.

PAGE 111

CHAPTSH IV SUMMARY A number of ring-opening reactions of 5-perfluoroalkyltetrazoles were studied. These fell into two classes; ring-opening at relatively low temperatures (below 150**) induced by electronegative substituents on the ring nitrogen; and, thermal ring-opening at 200" or above to produce a perfluoroalkylnitrilimine whose reactions were investigated. Of the former type, lT-perfluorobutyryl-5-perfluoropropyltetrazole was found to spontaneously eliminate nitrogen at room temperature to produce 3,5-bis(perfluoropropyl)l,3,^-oxadia2oleo Perfluoroacylation of free 5-perfluoroalkyltetrazoles occurred at 100-150® to give oxadiazoles in yields up to 96 per cent. l,5-bis(5-Perfluoroalkyl-l,5,4oxadiazolyl-2)perfluoropropanes and 5,5' bis Qperf luoroalkyl)-2,2'-bi-l,3,'4— oxadiazoles were also produced by an extension of the above type of reaction. Perfluorobutyrimidoyl chloride, an unstable compound reversibly produced by addition of hydrogen chloride to perfluorobutyronitrile, was found to undergo a similar reaction with 5-perfluoropropyltetrazole to produce 3,5-bis(perfluoropropyl)-l,2,^-triazole. A mechanism for these types of reactions was proposedo 101

PAGE 112

102 5-Perfluoropropyltetrazole was found to react with phosgene to give stable N,N'bis (perfluorobutyrimidoyl chloride)urea. An intermediate in this reaction, tentatively identified as 2-(perfluorobutyrimidoyl chloride)amino5-perfluoropropyl-l,5>^-oxadiazole, was isolated and converted to the urea derivative in a separate reaction with hydrogen chloride. Perf luorobutyronitrilimine , produced by thermolysis of 5~pei*fluoropropyltetrazole, was removed from its place of generation and reacted at or below room temperature with water, perfluoroalkylnitriles and perfluorobutyne-2o At higher temperatures and lov/ pressure the thermolysis resulted in a 45 per cent yield of 2-H-perfluorobutene-l. This probably resulted from a rearrangement of the nitrilimine to 1-diazo-l-H-perfluorobutane which decomposed to the carbene, CF;,-C?2~C?2~CH. The observed olefin could then be formed by insertion of the carbene. A small amount of 3, 5-bis (perf luoropropyl)-l,2-dihydro-l ,2,4,5tetrazine was formed by dimerization of perf luorobutyronitrilimine under the reaction conditions » When perfluoroalkylnitrilimines were generated in the presence of perfluoroalkylnitriles or perfluorobutyne-2 , they readily added to the carbon-nitrogen or carbon-carbon triple bond to give 5,5-bis(perfluoroalkyl)1,2,4-triazoles and 3,4,5-tris(perfluoroalkyl)pyrazoles

PAGE 113

103 respectively. Ttiis was found to be a good synttietic nethod for preparation of the pyrazoles but a more convenient synthesis exists for preparation of the triazoles. Infrared spectral assignments of the perfluoroalkylsubstituted tetrazoles, triazoles and pyrazoles were made and a comparison of the acidities of these perfluoroalkylsubstituted heterocycles and their parent compounds was made.

PAGE 114

BIBLIOGRAPHY 1. ?. R. Benson, Chem. Revs., fta, 1(19^7). 2. V. P. Norris, J. Org. Chem. , 27, 32^8(1962). 3. R. Stolle and P. Henke-Stark, J. Prakt. Chem. , 12^, 261(1950). ^. R. D. Huisgen, Angew. Chem., 72, 566(1960). 5. K. A. Hoffman and Ho Hock, Ber. , fj^, 29^7(1911). 6. F. R. Benson, op. cit . , p. 7. 7. H. Rathsburg, C.A., 17, 1147(1923) [Brit. Pat. 185,555 (1921)]. ~ 8. V. Brlin, C.A. , 29, ^579(1955) [U.S. Pat. 2,001,299 (1935)]; C.A., 30, 490(1936) [U.S. Pat. 2,021,478 (1935)]. 9o E. V. Herz, C.A. , 27, 1013(1933) [Ger. Pat. 562,511 (1931)]. 10. J. A. Bladin, Bar., IS, 1544(1885). 11. J. S. Mihina and R. M. Herbst, J. Org. Chem., 15, 1082-92(1950). 12. V. G. Pinnegan, R. A. Henry and R. Lofquist, J. Am. Chem. Soc, 80, 3908(1958). 13. R. Stolle, Ber., 62, 1118(1929). 14. R. Huisgen, J. Sauer and M. Seidel , Chem. Ber,, 93 , 2106-2124(1960). 15o J. Sauer, R. Huisgen and H. J. Sturm, Tetrahedron, 11 , 241-51(1960). 16. R. Huisgen, J. Sauer and n. Seidel, Chem. Ber., 95, 2885-91(1960). 104

PAGE 115

105 17. H. C. Brown, J. Poly. Sci. , fj^, 9-22(1960). 18. H. C. Brown, M. T. Clieng, L. J. Parcell and D. Pilipovich, J. Org. Chem., 26, ^407-9(1961). 19. H. C. BrovTn and M. T. Cheng, J. Org. Chem., 27 , 3240(1962). 20. D. R. Rusted, UoS. Pato 2,676,985 (195^). 21o H. C. Brown and Do Pilipovich, J. Am. Chem. Soc, , 82 , 4700(1960). 22. V;. L. Reilly and H. C. Brown, J. Am. Chem. Soc, 78 , 6052(1956). 23. V. Lossen and C. Lossen, Ann. 265, 101(1891). 24. Eo Huisgen, J. Sauer, H. Sturm and J. Markgraf, Chem. Ber., 95, 2112(1960). 25. E. R. Bissel and R. E. Spenger, J. Org, Chem. , 24, 1147(1959). 26. A. Pinner, Ber., 27, 990(1894). 27o Personal communication with Dr. Thomas L, Jacobs, Aug. 21, 1965, University of California, Los Angeles. 28. R. A. Carboni and R. V. Lindsey, Jr. , J. Am. Chem. Soc. , 80, 5795(1958). 29. E. Shechter, Abstr. of Papers, Am. Chem. Soc. Meeting, March, 1961, p. 15-0. 3O0 L. J. Bellamy, Infrared Spectra of Complex Molecules , Methuen and Company Ltd. ," London, 195^, P» 141. 51, A. D. Harris, Ro H. Berber, H. B. Jonassen and G. K. Uertheim, J. Am. Chem. Soc, 85, 2930(1965) o 32. P. Mirone and M. Vampiri, C.Ao, 46, 9423(1952). [Atti accad. nazl. Lincei, Rend., Classe sci» fis. , mat. e nat. 12, 585-7(1952)] « 33o A. Albert, Heterocyclic Chemistry , Alathone Press, London, 1959, PP. 142-143 »

PAGE 116

105 3^. F. Swarts, Bull. Classe. sci. Acad. roy. Belg. , (5) 12, 692(1926). ~~ 35. H. Gilman and R. G. Jones, J. Org. Chem. , 65, 1458(1943). 36. W. Friederich, U.S. Pat. 2,710,297 (1955).

PAGE 117

BIOGRAPHICAL bICSTCH Robert James Kassal was born October 25, 1936, at Berwick, Pennsylvania. In June, 195^, b.e was graduated from Hempstead High School, In June, 1958, he received the degree of Bachelor of Arts from Hofstra College. Prom June, 1958, until August, I960, he was employed as a chemist by American Cyan am id Company. During this period he attended the Polytechnic Institute of Brooklyn night graduate school. In I960, he enrolled in the Graduate School of the University of Florida and worked as a graduate research assistant in the Department of Chemical Engineering. Prom September, I960, until the present time he has pursued his work toward the degree of Doctor of Philosophy. Robert James Kassal is married to the former Barbara Swanson and is the father of three children. He is a member of Gamma Sigma Epsilon. 107

PAGE 118

This dissertation was prepared under th.e direction of the chairman of the candidate's supervisory committee and has been approved by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 21, 1963 Dean, College of Ar-?g ajix^^Tences Dean, Graduate School Supervisory Committee: Chairman ' U.Vn^^ \ / 1). i'iX..(, ' .