Radiotracers in fluorine chemistry. Part IX fluorination of chlorofluoroethanes by chromia catalysts treated with hydrogen fluoride or hydrogen [18F]-fluoride

Radiotracers in fluorine chemistry. Part IX fluorination of chlorofluoroethanes by chromia catalysts treated with hydrogen fluoride or hydrogen [18F]-fluoride

Jourrlal of Fhorirw Chemistry, 27 (1985) 2 13-229 213 Received June 24, 1984; accepted October 17, 1984 RADIOTRACERS IN FLUORINE FLUORINATION O...

705KB Sizes 1 Downloads 142 Views

Jourrlal of Fhorirw Chemistry, 27 (1985) 2 13-229

213

Received June 24, 1984; accepted October 17, 1984

RADIOTRACERS

IN

FLUORINE

FLUORINATION

OF

CHLOROFLUOROETHANES

TREATED

JERZY

WITH

HYDROGEN

KIJOWSKI,

Chemistry Scotland

FLUORIDE

GEOFFREY

Department,

PART

CHEMISTRY. BY OR

WEBB

and

The University

IX

CHROMIA

CATALYSTS

HYDROGEN

["F]-FLUORIDE

JOHN

WINFIELD.

M.

of Glasgow,

Glasgow

G12

SQQ

(U.K.)

SUMMARY

Passage

of dichlorotetrafluoroethane

1,2,2-trifluoroethane viously

treated

fluorinated activity the

at

with

and

hydrogen

chlorinated

in the products

catalyst

containing

species. and

fluoride

at

reactions halogen

or l,'l,P-trichloro-

K over

623

chromia

K, leads

Incorporation

products,

when

indicating

The

Cl-for-F

>/ 623

is observed

pretreatment,

of F-for-Cl

isomers

temperatures

hydrogen

the

of

[

18

involvement

observed

exchange

are

catalysts,

to the

fluorine-18

F)-fluoride

radio-

at

in

fluorine-

in terms

occurring

of

is used

of a surface

described

reactions

pre-

formation

of series

the

catalyst

surface.

INTRODUCTION

Although

the

fluorination

the presence

of aluminium

has

widespread

received

identification

and

trifluoride, study

reactions

[2,4] cesses,

in

which

terms

of

__)

CC12FCC1F2

+

On the occur

CC12FCC1F2

CC1F2CF3

0022-1139/85/$3.30

under

equations

_+

chromia, the

main

conditions

basis

of

for

by

the

+

have

been

efficient

and

fluoride

in

catalysts on product

preparation

product

conditions

(disproportionation) 1 and

hydrogen

or alumina-based emphases

observed

heterogeneous

reorganisation

for example,

2CC1F2CC1F2

[l-5],

on designing

chloropentafluoroethane. the

of chlorofluoroethanes

of

distributions

have

been

discussed

isomerisation

pro-

2:

CC1F2CF3

CC1F2CC1F2

(I) +

CC12FCF3

(2)

0 ElsevierSequoia/Printed in The Netherlands

214

An ionic

mechanism

r21,

and for

lyst

has been

is

nickel-doped

the

obtained

containing

interactions

direct

which

evidence

species

in

for

the

for

occur

the

by means of

doubt

on the

fluorination

of

of

actions

and an alternative

account

of

(t,

exchange for

the

scheme

is

cata-

surface.

fluorinated

little

However,

a surface

study.

reorganisation

reaction

process

We

fluorine-

dichlorotetrafluoroethane

110 min)

q

concerte?d

work has been

of

over

a fluorine-18

importance

this

Cl-F

at a catalyst

involvement

and 1 ,I ,2-trichloro-1,2,2-trifluoroethane lysts

the

AlF and chromia, a redox function 3 on the basis of an e.p.r. study [6].

suggested

known about

have

has been postulated

isomers

chromia

cata-

Our results

cast

and isomerisation

re-

A preliminary

proposed.

c-.

given

L/J.

EXPERIMENTAL The pelleted,

chromia

catalysts

used

in

this

work were commercial

samples.

Experiments were performed both with fresh catalyst, surface 2 -1 fluoriarea66.5mg , and with samples that had been used for catalytic 2 -1 nation, surface areas 9.8 - 11.0 m g . Chlorine and fluorine contents of the

latter

were in

The surface

area

the

of

range

the

0.20

fresh

- 0.64

catalyst

1 ,I ,2-trichloro-1,2,2_trifluorethane fluorine

contents

Hydrogen

were 0.78

fluoride

tetrafluoroethane were used ratio trace

of

the

Scottish

dried

was prepared

product

further

in --

by an exchange 20 $_i)

chlorine

and

99.8%),

dichloro-

and trichlorotrifluoroethane

(I.C.I.

P.L.C.

I9 F n.m.r.

that

spectra

: l,O,

nominally

indicated

and that

)

the mole

CC12FCC1F2 contained

a

CC13CF3.

Universities’

irradiated

and its

I.C. I. P.L.C.,

: CC12FCF3 was 12.8

isomer

Fluorine-18 the

Their

- 28.0 w/w L respectively. 2 -1 m g after use in

22.0

w/w % respectively.

or

(Fluorochem)

CC1F2CC1F2

to

experiments and 8.4

(Matheson

as received.

and 24.1 fell

gave vacua.

reaction

by the

sequence

Research Cs”F(aq) above

6Li(n,a)3H,

Reactor,

East

[EI] which

373 K.

between

160(3H,

Kilbride.

was evaporated

to

Hydrogen t8F]-fluoride

HF(ca, -

3-5

cm3 liquid)

n)18F

using

Work-up

of

dryness

and

the

was prepared

and Cs 18F(ca.

lg;

at

293 - 523 K in a Monel metal pressure vessel (95 cm-3) attached to 18 a Monel metal vacuum system. F count rates were determined using a NaI well

scintillation

1.56

x 0.78

contained Specific -1 mm01 and all

in count

.

counter

ins

diameter.

F.E.P.

rates

of

rates

and Nuclear

Aliquots

counting

Radiochemical count

(Ekco

H”F purity

tubes

of

H18F

fitted

determined

with

for

well

dimensions

by predried

Monel valves

by this

was established

were corrected

Enterprises), complexed

CsF were

(Whitey).

means were > 100 count by half-life

background

and decay.

determination

s

-1

215

Reactions between HF and C2C12F4 or CC12FCClF2 were carried out in a Monel metal (o.d. = 0.25 ins)- 'Pyrex' flow system, Monel needle valves (Whitey or Nupro) and P.T.F.E,/glass stop-cocks (Rotaflow)being used as appropriate,

The reaction vessel was a vertical Monel tube, o.d.

O-5

q

ins, containing a heated zone, 7 ins, within which the catalyst was supported on a copper gauze.

The temperature of the heated zone was determined

by a centrally mounted thermocouple and gas flow was from top to bottom, The flow system had separate chlorofluorocarbonand HF feed and exit lines and a by-pass line for the reactor.

Both reactor and product exit lines

were pumped for several hours to minimise the build-up of less volatile, highly chlorinated products. Catalyst samples were given several pre-treatmentswith HF vapour at 623 K before use in a series of experiments.

Before each experiment the

apparatus was flushed with dry nitrogen while the catalyst was heated to HF vapour, dispensed from a measured volume of liquid HF in a

623 K.

Monel vessel at 333 K, was passed over the heated catalyst, its flow rate HF not retained

being controlled by a differential-pressureflow-meter.

by the catalyst was passed through a Kel-F trap, in which aqueous HF was collected, the remaining effluent was absorbed in NaF - soda lime towers. The reactor was flushed with

dry N2 for IO-30 min while the catalyst was

brought to the desired temperature for reaction.

A chlorofluorocarbon

vapour stream, generated by passing helium at a constant flow rate through a weighed quantity of C2C12F4 held at 258 K or CC12FCC1F2 held at 298 K, was allowed to react with the fluorinated catalyst.

Reaction products,

after removal of trace HF by a small NaF scrubber, were monitored by regular, 5-15 min, sampling of the gas stream using a PE.Fll chromatograph (3 m, 0.125 ins ad. column packed with O.P.N. on Porasil C, thermal conductivity detector).

Members of the series C2C16_nFn, n

q

2-6, were quanti-

tatively determined but it was not possible to differentiate between isomeric 19 F n.m.r. spectroforms. Identificationof isomers was accomplished by scopy

of

were not

larger

product

always

achieved,

Constant

samples. particularly

for

chlorofluorocarbon reactions

flow

involving

rates

CC12FCC1F2g

623 K, and heating the exit line and NaF trap to 373 K made no improvement. The problem exit are

was circumvented

as GC samples

were

to

taken.

some extent

by determining

Mean values with their

flow standard

rates

at

the

deviations

quoted. Fluorine-18 tracer studies were carried out in a similar manner with

the addition of an F.E.P. counting tube at 90o to the reactor at its upper end.

The reactor

assembly could be removed from the flow system without

216 loss

of

integrity,

pellets

could

Separate

in

tion

of

I8 F count

H”F

uptakes

rate

of

or

for

counting n.m.r.

the

mid-point

of

catalyst

cells.

The composition

of

large

‘*F

losses,

in

either

vacua

to

be

a low

-in

small

situ -3

volume by 19F

was determined

unless

to

was varia-

corresponded

were collected

in --

were>90%

the

20 pellets)

fraction

presumed

temperature

and that

CC13CC1F2 was a large

were > 90% except

balances

three

chlorofluorocarbon.

and weighed

transfer each

top

.

products

Mass balances

were

catalyst,

(ca. -1

the

point

reactor

bed

after

products,

Radiochemical

the

and counted,

train,

spectroscopy.

sampling

mmol (g catalyst)

chlorofluorocarbon

> 623 K where

the

that

the

- 1.0

volatile

component. at

the

fractionation

highly

after

over

0.9

[18F]-labelled temperature

the

established

than at

18

F count rates of II3 flows of H F and the

that

be determined

experiments

IO - 30 K higher

to

order

in

loss

some experiments of

labile

H18F from

encountered.

RESULTS Reactions

of

dichlorotetrafluoroethanes

Flowing take

of

HF over

HF by the

pelleted

H1’F of

sample’s

history,

for

those

obtained

less

than

the

range

ethane, with

0.3

HF-treated

CF3 (115) (113) also

ratio

using

of

fresh

catalyst

(112a)

rate,

623 K results HF uptake,

depends

previously

Table

flow

minor

on the

however

conditions

products.

products

in

the up-

catalyst samples

uptake

1.

catalysts

determined

fluorinated

is

are

always

in

Dichlorotetrafluoro-

: CC12FCF3 (114a)

as major

are

the

catalyst,

under

and C2C12F4 isomers

at

of

mmol HF (g catalyst)-‘,

fresh

12.8

q

at

: 1.0,

698 K to

The compounds and a trace

of

react

give

CC1F2

CC12FCC1F2

C2F6 (116)

is

observed. A typical is

min-‘1

of

product shown in

C2C12F4 is

molecules

and the

any given

time

the

reacting

2

count

those

CC1F2CC1F2 (114)

and CC13CClF2

tography,

is

specific

example

HF- and Hq8F- treated

catalyst

The magnitude

measured

- 1.2

mole

chromia

catalyst.

by using

with

q

flow (2)

Fvx -’

is

time, -1 It

given

- (number

1 (a ).

constant

Fvx

rate. is

distribution Figure

the

and is

time,

number of

the

determined

that

the

feed

quotient

of

the

molecules

then

follows

effluent)

that

the

by gas chromarate

volume

number of

(F molecule

number of

contacting

, where v cm3 is the catalyst

by equation

in

with

Assuming

feed

the

catalyst at 3 -1 and x cm mm

substrate

molecules

(3)

(3)

treatment

1

t

*

used

H18F

1 - 4 and

f 3

859 564

30

f 2

f 3

1077

25

+ 2

+ 3

793 349

f 3

836

30

25

30

25

+ 2

409

a catalyst

sample

in situ.

fresh catalyst; nos. 5 - 6

CC12FCClF2.

by counting

reactions with

Determined

7 - 8

4.0

11

Nos.

3.5

4.0

9

IO

3.5

3.5

3.5

6

8

4.0

5

7

_+ 5

2712

30

3.5

4 25

f 3

729

20

3.5

?: 5

f 3

3

942 2169

25

-1

30

s

3.5

count

min

3.5

rate+

and

taken

up

mm01

catalyst.

399

0.45

0.31

1 - 6

reactions with

* 2

+ 3

5 2

* 2

f 2

C2C12F4;

322

629

241

0.33

505

1.06 0.83

+ 2

407

0.25

f 5

.+ 2

232 k 2

2825

425

+ 5

? 3

-1

0.30

1.19

0.97

777

s

2067

count

0.94

-1

rate

nos.

t after reaction

Count

1.02

g

by catalyst

H18F

9 - 11 used Nos.

Count

rates

Initial

count

flow time

K

catalyst

at 623

and

2

cm3 liq.

Treatment

H18F

of catalysts

1

Experiment*

H"F

TABLE

F retained

7 - 11

50.6

73.2

37.0

69.0

63.6

48.6

56.7

58.3

95.3

82.4

%

by catalyst

18

218

LO

30

I i

6C

I

\

u

‘?,.~,.__.,./.-.“-‘\.ll~ /

-----w

-50

-v 8

-40 N -

“-‘-*b._.._.~\.,,5

w30 %

./

-20

1.

0

.-.-.-.-.-.-.-.,-.-.112a -.__-.-.-, -.--_.--._.113 150 50 100

LL/L

0 10

L-

50

1 Tz--

--?50

Fig. l(a) Product distribution with time in the reaction C2C12F4 + HF-treated catalyst at 698 K.

(b) z(Z) molecule

vs. time (min) for the same reaction,

Numbers of effluent molecules are calculated from G.C. traces, making allowance for the difference in catalyst and sampling loop volumes. Summations of G.C. data for

two experiments, Table 2, indicate mass

balances are > 95X, that the yields of CClFZCF3 are 36 - 34L, and that the ratio CClFZCF3 : C2C12F4 in the product is 0.58.

Plots of r(Z) vs. time,

Figure l(b), are linear in the later stages of the reaction but deviate initially.

The observed behaviour indicates that the reaction is zero order

with respect to gaseous C2C12F4 at least in its later stages. 18 The behaviour of C2C12F4 when flowed over H F-treated catalyst is similar to that described above except that its degree of conversion to fluorinated and chlorinated products was lower due to the necessarily shorter reaction times.

Radiochemical data for the reactions are given in Tables 1

and 3 and the identified products in Table 4. In each case retention of 18 F activity by the catalyst after C2C12F4 flow is significant, for example 18 after reactions at 698 - 703 K, the proportion of F activity remaining is in the range 59-48%.

Obviously not all of the fluoride species on the

catalyst is catalytically active.

219

-

Y

-

3

t

*

5.75

9.00

5

6

Based

on initial

30 35

0.66 0.45 0.70

rates

30

0.68

1.

30

0.59

of the catalyst

in

30 30

0.57

Table

1)

rate

1280 3938 387

la1 + I

698 698 698 703

f 3

i: 2

f 7

c 4

+ 5

692

-1

2052

s

623

count

173 K

in

8.7

product t 7;

22.4 2.8

f 1

97

12 + 1

la.7

f 2

332

21.6

* 1

11.3

total

of "F

Incorporation

45

113 K count s -1

at

of product

collected

Count

623

K

Temperature

(Table

time

min

Flow

catalysts

mm01

CC12FCF3

[18F]-labelled

to those

count

a.50

4

correspond

a.69

3

Numbers

7.27

7.54

2

mm01

+

with

CClF2CClF2

of C2C12F4

1

Experiment*

Reactions

TABLE

221

222 Although incorporation of I8F activity in the products is observed in all cases, the degree of incorporationobserved depends on the reaction temperature and on the history of the catalyst sample.

Reactions carried out

at 698 K with previously unused catalyst (reactions3 and 4 in Table 3) 18 F incorporation,the products (Table 4) being result in approximately 20% CClF2CClF2, CC12FCF3, CClF CF3, some CC12FCC1F2, and a trace of CC13CF3. At 623 K incorporation of '*F in the products, CC1F2CC1F2, CC12FCF3, and CC1F2CF3, is only 9 - 11% (reactions 1 and 2, Tables 3 and 4).

The

behaviour of extensively fluorinated catalysts (e.g. reactions 5 and 6 in Tables 3 and 4) is less reproducible. The yield of CC1F2CF3 in 18 reaction 5 is very small although F incorporationin the products is 18 significant, but in reaction 6 F incorporationis barely detectable and no CClF CF was observed. Although it was not possible to determine 182 3 specific F count rates of the individual products, their relative order may be estimated from those reactions in which two product fractions were counted and analysed by n.m.r. spectroscopy.

For example the data

obtained from reaction 4 indicate the order of specific count rates CClF2CF3 > CC1F2CC1F2, the derived values, assuming that C2C13F3 isomers are inactive and that specific count rates of C2C12F4 isomers are equal, -1 -1 are 436 and 1277 count s mm01 respectively. Although the derived count rates depend upon the assumptions made, the relative order does not.

The data are not consistent with all components having equal

specific count rates.

Reactions of 1,1,2-trichloro - 1,2,2-trifluoroethaneswith HF- and H'*F- treated catalysts The reaction between CC12FCC1F2 and HF-treated, extensively fluorinated catalyst at 703 K results in a mixture of CC13CC1F2 (112a), Ccl2 FCC1F2 (113), C2C12F4 isomers (114 + 114a) and CC1F2CF3 (1151, Table 2. The latter is a minor component and was not observed as a product in reactions involving HF-treated, fresh catalyst at 623 K, Table 5.

The

G.C. mass balance for the 703 K reaction was satisfactory but those at 623 K were not, however overall ratios of products were determined, Table 5.

Formation of more highly chlorinated products is a feature of

these reactions, and the distribution of products sampled at various times during a reaction, Figure 2, indicates that CC13CC1F2 is 'held-up' significantly in the reactor and exit line.

This may mean that a 'build-

up' of chlorine-containingspecies occurs on the catalyst during the early stages of its use.

223 TABLE 5

Reactions of CC12FCClF2 with HF-treated catalyst at 623 K*

wt.

of

Flow time

Av. flow rate

Overall product ratio determined from G.C. sampling

g

min

cm3min-T

112a

4.4191

135

31.8 + 0.8

2.4

2.6

1.0

4.4165

180

24.2 f 1.5

2.2

2.7

1.0

4.4165

165

26.7 + 1.7

2.5

3.1

1.0

Catalyst

*

Fresh catalyst treated by flowing HF

for

:

20-30

:

113

min

at

623

114

K.

112a,

CC13CC1F2; 113, CC12FCC1F2; 114, CC1F2CC1F2.

'8F data and product yields for reactions of CC12FCC1F2 with H18Ftreated, fresh or previously fluorinated catalysts are given in Tables 6 18 and 7 respectively,and the appropriate H F uptakes are contained in Table 1.

Reactions at 623 K using fresh catalyst (nos. 7 and 8) lead to

CC13CC1F2, CC12FCC1F2, CC1F2CC1F2, and minor quantities of CC12FCF3. Incorporationof 18F in the products is 24 - 28%.

One of the fluorinated

catalysts used (reaction 8) shows similar behaviour, although product yields are lower 18

and

a small

quantity

of

CC13CF3

is

produced.

Specific

count rates determined for the products from this reaction, assuming -1 CC13CC1F2 to be inactive, are 161 and 2139 count s mm01-I for C2C13F3 F

and

C2C12F4

isomers respectively.

The other fluorinated catalyst used

was less effective, and no reaction was observed below 703

K.

this temperature the extent of reaction is small (reactions 10

Even at and

11).

6

*

corresDond to

14.7

11

Numbers

14.1

in

8.20

9

10

7.16

8

Table

35

35

25

30

40

9.32

7

1.

time

min

Flow

['RF]-1abelled

those

with

CC12FCClF2

of CC12FCClF2

mm01

Experiment*

Reactions

TABLE

t Eased

703

703

623

623

623

K

on initial

Temperature

catalysts

rates

-1

rates

f 3

* 3

+ 4

f 2

+_ 3

count

895

1119

1392

570

1095

count s

233 K

collected

Count

173 K -1

of the

62

45

481

93

196

catalyst

+ 1

f 1

+ 2

+_ 1

+_ 2

count s

at

of products

(Table

20.7

16.5

32.9

27.8

23.8

I).

products t %

of 18F in

Incorporation

*

114a CC12FCF3.

count rates see Table 6.

114a

0.01

-

-

113a

1.09

0.01

Cf.68 trace

0.85

0.65

0.82

113

trace

0.07

0.03

114a

For

trace trace

trace trace

0.16

0.58

0.36

114

19 F n.m.r. spectroscopy.

0.01

0.04

112a

Collected at 173 K mmol

112a, CC13CC1F2; 113, CC12FCC1F2; 113a, CC13CF3; 114, CC1F2CC1F2;

Product analyses by

0.13 0.09

0.12 0.07

0.13 0.05

0.67 0.05

0.61 0.04

114

Numbers correspond to Tables 1 and 6.

11.48 0.03

0.08

6.20 0.04

-

11

0.06

9

4.85

-

113a

10.99 0.19

0.27

I3

6.37

113

IO

0.22

112a

Collected at 233 K mmol

7

Experiment*

Products from reactions of CC12FCClF2 with [18F]-labelledcatalysts

TABLE 7

113flow terminated

time mln Fig.

2.

Product

distribution

with

CC12FCC1F2 + HF-treated

time

in

catalyst

the at

reaction

623 K.

DISCUSSION

The present out

work differs

previously

[1,2,41

in

over

the

results

in

flowedseparately This

procedure

ducts,

but

species

to

has been

it

enables

inferred in

demonstrates

that

there

occur

during

is

is

reaction

significant

Evidently catalyst

of

the not

other use.

of

kinetics

aluminium

()

carbon

1)

pro-

of

HF

tetrachloride the

use

monitored, In all

retention

example

adsorption

[9],

be directly

reversible.

for

fluorinated

fluorine-containing

trifluoride to

37% Table

reactions,

of

were

as a mixture.

to

Although

adsorption completely

than

a surface

demonstrated.

the presence

it

rather

low conversions

involvement

the

HF allows

catalyst

relatively

the

from

fluorination 18 Fl-labelled

catalyst.

that

synthetic investigations carried It3 HF or H F and the chlorofluorocarbon

chromia

be unambiguously

[

studied

from the

of

the

of and

reactions

18F by the

fluorination

of

chromia,

221 The

behaviour of 1,1,2-trichloro-1,2,2-trifluoroethane and the

dichlorotetrafluoroethanescan

be rationalised on the basis of a series

of related halogen-exchangereactions.

These are summarised in the

Scheme, in which Fa and Cla represent fluorine- and chlorine- containing species present on the chromia surface.

CC13CClF2

CC13CF3

+--c1 a

CC12FCF3

Scheme

The

I8 F

steps

tracer experiments provide direct

CC1F2CC1F2

--+

evidence

for the fluorination

CClF2CF3 and CC12FCClF2 --+

are important reactions under the conditions used. CC12FCClF2 B

CClF2CClF2; both The reaction

CC12FCF3 is of minor importance but does occur.

the limited data obtained for the process CC12FCF3 +

From

CC1F2CF3, it

appears that the asymmetric isomer is fluorinated more readily than is CClF2CClF2. The product distributions observed from both substrates, CC12FCClF2 and C2C12F4 isomers, are consistent with the ease of fluorination of a C-Cl bond increasing in the order -CClF2 < -CC12F.

This is in keeping

with the results obtained using other fluorinatingagents, for example Sb"' , Sb”, chloride, fluoride mixtures [IO].

Exact

descriptions

for

the transition states of the fluorinationscannot be given, but it is likely that prior adsorption of the chlorofluoroethaneis required. Adsorption of CClF2CClF2 and chlorofluoromethaneson lanthanide trifluorides and other trihalides has been suggested from an i.r. study [II]. Although direct evidence for adsorption in the present case has not yet been obtained, circumstantial evidence comes from the reaction of C2C12F4 which approximates to zero-order in gaseous C2C12F4, Figure l(b).

228

The observation of CC12FCC1F2 and CC13CClF2 as products from reactions involving C2C12F4 isomers and CC12FCClF2 indicates that the chlorination reactions CC1F2CClF2 ----+ CC12FCCLF2, CC12FCCLF2 ---+ possibly CC12FCF3 ---+

CC12FCClF2 occur.

CC13CClf2, and

The observation of CCl3CF3

as a trace product in some reactions, Table 7, suggests that CCl2FCF3 ----+ CC13CF3 is possible.

Chlorinationis particularly important in

experiments involving CCI2FCClF2 at 625 K.

The fate of chlorine-contain-

ing species formed from the fluorination reactions was not investigated, but it is reasonable to assume that they are present, at least in part, as surface species on chromia and can take part in further reaction. There are at least three different types of fluoride present on chromia as a result of HF treatment 112.1and the 18F labelling of CC12FCClF2 and CClF2CCIF2 observed in the reactions of these compounds could be the 19F exchange on the H'SF-treated chromia,

result either of direct 18F- for

or of sequential chlorination then fluorination reactions.

Although the

former route cannot be excluded completely, we favour the latter as being more in keeping with the overall experimental observations.

When

CC12FCClF2 is flowed over H"F-treated chromia at temperatures below F rncorporationin the products is approximately 623 K the degree of If3 10%.

The mixture obtained is predominantly I1*Fl-CC12FCClf2,but traces

of CC1 CClF2 and CClF2CCLF2 are always present.

There is no evidence

that 'I' F exchange occurs prior to the fluorinationand chlorination 18 reactions. Similarly flowing C2C12F4 isomers over H F-treated chromia at 623 K results in I8F incorporation in the products of approximately IO%, reactions 1 and 2 in Table 3.

The products under these conditions,

Table 4, are [18F]-C2C12F4and a small quantity of ['SF]-CCiF2CF3. CC12FCClF2 is not observed, but under these conditions it would be expected to react readily to give [18Fl-CC1F2CClF2and [18~1-~~1,~~~3,Tables 6 and 7.

In those reactions where the specific count rates of [18F]-

c2Cl6_nFn and 118Fl-C2C15_nFn+1,n = 4 or 3, have been compared, those of the fluorinated products are greater than those of the reactants, C2C12F4 or C2C13F3.

This is to be expected if the labelling route for

the latter involves two events at the chromia surface. Our proposed model involves fluorinationand chlorination reactions, related by the active species for the latter originating from the former. However, there is no reason to believe

that fluorinationand chlorination

reactions are concerted as is a requirement for the dismutat~on reaction postulate [2,4].

Similarly direct isomerization is not a necessary post-

ulate to explain the distribution of products observed.

229 REFERENCES

1

M. Vecchio, (1974)

2

G. Groppelli

L. Kolditz, 434 --

G. Kauschka

(1977)

L. Kolditr,

4

L. Marangoni,

5

L. Marangoni,

U. Calov,

Chem.,

(1981/82)

J. Kijowski,

8

J.E.

G.P. -23

10

11

(1977)

J. Fluorine

Chem.,

Schmidt,

Z. Anorg.

Allg.

Chem.,

and

W.

Schmidt,

Z. Anorg.

and

L. Conte,

J. Fluorine

Chem.,

L. Colombo,

Chem.

Stgsser,

Z. Anorg.

19 _-

C. Gervasutti

, 476

and

Ind.

135. G. Heller

(1981)

G. Webb

and

R.

23.

and

J.M.

Winfield,

J. Fluorine

Chem.,

24 --

133. Scottish

Nucl.

Sci.

Gambaretto,

(1973)

Research

Abs.,

2

Reactor

(1968)

F. Avezzu

and

Centre,

Report

No.

SRRC

30592.

E. Gola,

J.Appl.

Chem.

Biotechnol.,

175.

e.g.

W.A.

Sheppard

W.A.

Benjamin,

M.D.

Taylor

and

Inc.,

and

T.-T.

C.M. 1969,

Sharts,

'Organic

Fluorine

Chemistry,'

pp.76-7.

Cheung,

J. Inorg.

Nucl.

Chem.,

2

(1973)

3499. 12

4

55.

G. Rasia, (1982)

Whitley,

26/68;

W.

G. Kauschka

V. Nitrsche,

Chem.

(1984)

and

C. Gervasutti

64

7

9

434

L. Kolditz, Allg.

Tatlow,

21.

(Milan), 6

J.C.

41.

3

Allg.

and

117.

J. Kijowski,

G. Webb

and

J.M.

Winfield,

unpublished

work.