Free radical and isomerisation processes during the electrochemical fluorination of n-butyryl chloride, i-butyryl chloride and pivaloyl chloride in anhydrous hydrogen fluoride

Article abstract of DOI:10.1016/j.jfluchem.2006.05.018

Electrochemical perfluorination (ECPF) of the title compounds containing primary, secondary and tertiary carbon atoms was carried out in anhydrous hydrofluoric acid (AHF). Detailed analysis of major and minor products suggest that carbon chain isomerisation involving cyclo-propane intermediate is more prevalent during ECPF of i-butyryl chloride when compared to n-butyryl chloride. Simple statistical probability involving free radical intermediates also support this observation. ECPF involving cyclo-propane intermediate is even more prevalent in pivaloyl chloride containing three methyl substituents. In this case, perfluorinated cyclo-propane intermediates were also observed in the product sample. Distribution of minor perfluorinated and partially fluorinated products also suggest the predominant role of normal free radical pathway involving single-electron transfer.

Full text of DOI:10.1016/j.jfluchem.2006.05.018

Journal of Fluorine Chemistry 127 (2006) 1111–1118
www.elsevier.com/locate/fluor
Free radical and isomerisation processes during the electrochemical
fluorination of n-butyryl chloride, i-butyryl chloride and pivaloyl
chloride in anhydrous hydrogen fluoride
D. Velayutham a, K. Jayaraman a, K. Kulangiappar a, N. Ilayaraja a, Y. Ram Babu b,
P. Santhan Rao b, S. Narayana Reddy b, K. Victor Babu c, M. Noel a,
*
a Central Electrochemical Research Institute, Karaikudi 630006, Tamil Nadu, India
b Indian Institute of Chemical Technology, Uppal Road, Hyderabad 500007, Andhra Pradesh, India
c Central Leather Research Institute, Adyar, Chennai 600020, Tamil Nadu, India
Received 19 April 2006; accepted 24 May 2006
Available online 2 June 2006
Abstract
Electrochemical perfluorination (ECPF) of the title compounds containing primary, secondary and tertiary carbon atoms was carried out in
anhydrous hydrofluoric acid (AHF). Detailed analysis of major and minor products suggest that carbon chain isomerisation involving cyclo-
propane intermediate is more prevalent during ECPF of i-butyryl chloride when compared to n-butyryl chloride. Simple statistical probability
involving free radical intermediates also support this observation. ECPF involving cyclo-propane intermediate is even more prevalent in pivaloyl
chloride containing three methyl substituents. In this case, perfluorinated cyclo-propane intermediates were also observed in the product sample.
Distribution of minor perfluorinated and partially fluorinated products also suggest the predominant role of normal free radical pathway involving
single-electron transfer.
# 2006 Elsevier B.V. All rights reserved.
Keywords: Electrochemical fluorination; n-Butyryl chloride; i-Butyryl chloride; Pivaloyl chloride; cyclo-Propane intermediate
1. Introduction
The present state of understanding of the mechanism of
electrochemical fluorination [1] involves high valent NiF3 film
mediated free radical process as shown in Scheme 2.
Electrochemical perfluorination (ECPF) process consists of
essentially a number of electron transfer and associated
chemical steps. For example, conversion of i-butyryl fluoride
to perfluoro-i-butyryl fluoride involves removal of 14 electrons
and 7 protons from the molecule and replacement of 7 fluoride
ions (Scheme 1).
Each hydrogen is replaced by a similar sequential process
and the free radical intermediate formed during the overall step
(Eq. (1) in Scheme 2) can undergo additional chemical
transformations such as cyclisation and isomerisation men-
tioned above. Systematic analysis of a variety of small and large
reactants using mainly NMR data of partially as well as
perfluorinated samples were found to be in agreement with free
radical pathway [2–7]. There is also reasonable consensus
towards this pathway in recent times [8–16].
In addition to this desired process, additional competitive
processes involving C–C bond cleavage, isomerisation and
cyclisation process may also occur simultaneously. Hence,
comprehensive understanding of the detailed mechanistic
pathways has eluded researchers in this field over five decades
despite widespread commercial exploitation of the process.
Carbon chain isomerism is a common phenomena fre-
quently encountered in electrochemical fluorination. Formation
of a cyclo-propane intermediate through 1,3 coupling of carbon
free radicals formed on NiF3 surface due to preferable
orientation was proposed to quantitatively understand the
isomerisation process as shown in Scheme 3 [17].
* Corresponding author. Tel.: +91 4565 227772;
fax: +91 4565 227779/227713.
The cyclo-propane intermediate containing three C–C bonds
a, b and g can now undergo cleavage and fluorination on all the
E-mail address: yemenoel@yahoo.co.in (M. Noel).
0022-1139/$ – see front matter # 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2006.05.018

1112
D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
(iii) Does the cyclization model work for reactant molecules
containing trialkyl group as well?
Scheme 1.
The electrochemical fluorination of primary (n-butyryl
chloride), secondary (i-butyryl chloride) and tertiary (pivaloyl
chloride) carboxylic acid chlorides were taken up to address
some of the issues stated above.
The ECPF of n-butyryl chloride (NBC) and i-butyryl
chloride (IBC) was compared earlier by Gambaretto et al. [6].
The report indicated relatively higher conversion of i-butyryl
fluoride to perfluoro-n-butyryl fluoride when compared to
conversion of n-butyryl fluoride to perfluoro-i-butyryl fluoride.
In the present work, a more detailed investigation into these two
processes including partially fluorinated and cleavage products
was undertaken. To our knowledge, the ECPF of pivaloyl
chloride has not been studied in detail so far.
Scheme 2.
three bond positions. Cleavage at g position above will give the
fluorinated product with original carbon skeleton. Further
cleavage at a or b positions would lead to formation of a new
isomer (linear). Thus, there is an overall 2:1 chance of
isomerisation for each pair of H atom removal from the –CH3
group leading to n and iso product respectively. The formation
of n-perfluoro alkyl derivative from the i-alkyl reactant in
Scheme 3 would be much higher if the further replacement of
two more pairs of H atoms also occur through this cyclo-
propane intermediate. Statistically this ratio would be as high as
26:1 (see Scheme 7 reported by Ignat’ev et al. [17]).
The above model explains the mechanism of carbon chains
isomerism during electrochemical fluorination at least qualita-
tively. Ignat’ev et al. found some quantitative correlation as
well for the formation of higher amounts of perfluoro-n-
butanesulfonyl fluoride during the electrochemical fluorination
of i-butanesulfonyl fluoride and the formation of perfluoro-n-
propanoyl fluoride during the ECPF of cyclo-propanoyl
fluoride. However, ECPF of i-propanesulfonyl fluoride did
not show agreement with this mechanism. Many other
isomerisation processes reported earlier also are at variance
with the predictions of cyclo-propane intermediate model [17].
It is apparent that single-free radical pathway (Scheme 1) and
diradical/cyclo-propane pathway (Scheme 2) operate simulta-
neously during ECPF process.
2. Results and discussion
2.1. Formation of perfluoro carboxylic acid and their
product distribution
During the ECPF of IBC, NBC and pivaloyl chloride (PC)
no perfluorinated product was recovered from the bottom of
the electrochemical reactor. The vapor phase products were
collected in the cold trap at À70 8C and the weight of total
products obtained was taken at the same temperature. Typical
experimental results obtained during the ECPF of the above
three compounds are shown in Table 1. The current efficiency
values are only indicative as these are obtained by assuming
that no cleavage products were present in the product
mixture.
For one compound namely IBC, the experiments were
carried out at current densities ranging from 1.2 to 2.4 A/dm2
(Table 1, experiment nos. 1–4). Relative decrease in total
weight of products formed and hence current efficiency was
noticed when the current density increased beyond 2 A/dm2
(Table 1, experiment no. 4). Increasing the cell temperature to
10 from 5 8C (experiment no. 5) also did not have significant
effect on total yield. In general, overall current efficiency of 40–
46% could be achieved under optimum conditions. Further
improvement in current efficiency up to 51% could be achieved
with effective cooling of electrochemical reactor at À10 8C
(Table 1, experiment no. 6).
Further experimental investigations focusing on the relative
predominance of these two pathways would indeed be
desirable. Some of the issues that need to be addressed include
the following:
(i) Is the diradical/cyclo-propane model applicable for
conversion of n-alkyl group to i-alkyl group as well? Is
there any preferred direction from branched alkyl chain to
linear alkyl chain or vice versa?
The ECPF of NBC (Table 1, experiment no. 7) and PC
(experiment no. 8) were also carried out under such optimum
conditions. The perfluorinated products could once again be
collected only in the cold trap. As is common in ECPF
processes, more cleavage and volatility loss is noticed while
(ii) What are the effects of substituents on the three carbons
forming the cyclo-propane intermediate?
Scheme 3.

D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
1113
TFA
Table 1
Results of ECPF of i-butyryl chloride (IBC), n-butyryl chloride (NBC) and pivaloyl chloride (PC)
Experiment
no.
Reactants
Current
density
(A/dm2)
Total amount
of reactant
added (g)
Total charge
passed (A h)
Perfluorinated
product
Current
efficiency
(%)
Selectivity (mol%)a
obtained (g)
PFBA (i)
PFBA (n)
PFPA
1
2
3
4
5
6
7
8
IBC
IBC
IBC
IBC
IBCb
NBC
PC
1.2
66.7
106
237
376
364
606
380
370
215.6
384
57.6
102.4
93.6
94.4
100.8
80
42.2
47.3
44.8
27.1
46.1
37.6
37.7
51.2
46.3
42.7
41.8
41.5
45.7
6.9
55.8
55.4
50.6
54.7
42.8
92.0
12
2.1
1.9
1.7
1.4
3.1
1.0
21.0
0.8
<0.5
<0.5
5.7
1.6
2.02
2.43
1.6
103
167.7
112.1
104
2.3
8.3
2.02
1.6
<0.1
<0.1
<0.1
76.9
110.1
44.8
113
7.6
IBCc
1.6
54.4
44.8
Note: Current efficiency was calculated based on the perfluorinated products obtained at the cold trap maintained at À70 8C. ECPF reactor and condenser were
maintained at 5 and À30 8C, respectively, except for experiment nos. 5 and 6. Average cell voltage = 5.3 Æ 0.2 V. PFBA, perfluorobutyric acid; PFPA,
perfluoropropionic acid; TFA, trifluoroacetic acid. Volume of the electrolyte: 200 ml; active anode area: 2.47 dm2.
a
Based on 19F NMR data.
b
ECPF cell temperature maintained at 10 8C.
ECPF cell temperature maintained at À10 8C.
c
moving from primary to tertiary carboxylic acid chlorides
(compare experiment nos. 5, 7 and 8 in Table 1).
orientation of 1–3 carbon atoms at the NiF3 surface would
be higher. In linear chain compounds this probability is indeed
less and hence lowers isomerism.
The total product from the trap was treated with alkali and
then neutralised as described in Section 3. Though many other
compounds were also present in trace quantities (Sections 3.2
and 3.3), the main constituents were found to be perfluoro
carboxylic acids namely perfluoro-i-butyric acid (PFIBA),
perfluoro-n-butyric acid (PFNBA), perfluoropropionic acid
(PFPA) and perfluoroacetic acid (PFAA). The product
distribution of these four acids in the eight experiments are
summarised in Table 2.
Second even if the cyclo-propane intermediate is formed
there is still a 2:1 possibility of formation of n-butyl derivative
when compared to i-butyl compound during each pair of
fluorine addition as in Scheme 4 for example, the cleavage of
cyclo-propane intermediates at a and g position would still lead
to n-butyric acid intermediates, b cleavage alone would lead to
i-butyric acid intermediate. Hence, even if the process proceeds
through cyclo-propane intermediate, the mole ratio of PFNBA
to PFIBAwould be only 26:1 (see Scheme 7 in [17]). In general,
wherever cyclo-propane intermediate pathway is followed,
there is higher probability of formation of linear chain from
secondary and tertiary carboxylic acids.
These data clearly indicate that PFNBA is always formed in
higher quantities during the ECPF of IBC (Table 1, experiment
nos. 1–5) when the ECPF is carried out at a cell temperature of
5 8C. The product distribution between PFIBA and PFNBA is
found to be around 45:55 (experiment nos. 1–4). Increase in cell
temperature leads to slightly higher levels of cleavage products
(experiment no. 5). A higher selectivity of PF i-butyric acid
(54.4%) was obtained when the reactor temperature was
maintained at À10 8C. Even under this condition around 45%
of perfluoro-n-butyric acid was still obtained. Raising or
lowering of the ECPF temperature thus decreases or increases
perfluoro-i-butyric acid yield only through increase and
decrease of cleavage product formation. Formation of
significantly higher quantity of PFNBA in all these experiments
(experiment nos. 1–6) indeed clearly suggests the diradical/
cyclo-propane intermediate route of fluorination steps
(Scheme 3) at least partially.
In pivaloyl chloride, all the three methyl groups of the
compound lie in the same plane and hence there are much
higher possibility for diradical pathway. Sterric factors would
also lead to cleavage products. Due to this combined effect only
small quantity of perfluoropivaloic acid could be identified in
the NMR spectra (see Section 3.3). PFIBA and PFNBA are
found to be the major products (experiment no. 7, Table 1). The
overall yield of perfluorinated products was also lower for this
compound.
2.2. Minor constituents of fluorinated products obtained
during the ECPF of IBC
1
Quite interestingly ECPF of NBC leads to 92% yield of
PFNBA. The isomer is noticed only to the extent of 6.9%
(experiment no. 6, Table 2). This result is also in agreement
with the earlier observation by Gambaretto et al. [6]. Two
reasons may be attributed for this observation. In general,
radical process involving one NiF3 site and the carbon center of
the organic molecule (Scheme 2) has much higher probability
than the four centered interaction required for a diradical
mechanism (Scheme 3). When the fluorination involves
secondary or tertiary carbon atoms, the probability for
Analysis of the minor components using 19F as well as H
NMR indicated that practically all the minor products obtained
from ECPF of n-butyryl chloride are invariably observed in the
ECPF products obtained from IBC as well. Hence more
detailed analysis of minor products obtained during ECPF was
confined to products obtained from IBC alone. The complete
NMR data covering 22 compounds obtained from the product
mixture during ECPF of IBC are listed in Table 2. These
compounds are further classified under the four following
categories:

1114
D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
Table 2 (Continued )
Table 2
19F and 1H NMR data of polyfluorinated compounds obtained during the ECPF
No.
Structure
Chemical shift,
J (Hz)
of i-butyryl chloride and n-butyryl chloride
d (ppm)
No.
Structure
Chemical shift,
J (Hz)
a: À185.0
2J Ha Fa = 42.4
4J Ha Fd = 6.3
d (ppm)
c: À179.5
21
22
d: À74.4
a: 2.46 sept
b: 1.07 d
3J Ha Hb = 6.9
[18]
b: 5.7 d, sept, d
1
2
3
a: À186.0
b: À75.6
[21]
a: À160.4
3J Hb Fa = 22.2
b: 1.68 d
a: À73.0
3J Hc Fb = 21.2
b: À180.2
c: 1.71 dm
(a) reactant IBC and its partially and perfluorinated products
without cleavage (1–6);
(b) products based on partially and perfluorinated rearranged
product, namely NBC (8–11);
(c) products based on cleavage of IBC or NBC (12–20);
(d) products from intermolecular methyl transfer (21–22).
a: À194
2J HF = 55
b: À134 d
c: 6.52 td
3J HF = 11.5
4
b: À64.3
3J HF = 8
5
6
a: 4.81 sept
a: À179.66 sept
b: À74.38 d
3J FF = 7.8
[17]
The NMR data for each subgroup are summarised in Table 2.
These data clearly suggest that both Schemes 2 and 3 are
operative during the ECPF of IBC. Intermediates correspond-
ing to these two pathways in each group are briefly indicated
below.
a: 2.59 t
b: 1.56 qt
c: 0.89 t
3J Hc Hb = 7.6
3J Hb Ha = 2.3
[18]
7
8
9
a: À128.0
b: À120.6
c: 1.78 tt
a: À105.6
c: À63.5
b: 3.17 qt
Among the partially fluorinated IBA derivatives (Table 2)
compounds 4–6 may be attributed to be formed by cyclo-
propane intermediate. Compounds 2 and 3 would have
come only from free radical pathway. Compound 2 is an
interesting intermediate involving first substitution at tertiary
carbon.
3J Hb Fc = 9.7
3J Hb Fa = 5.2
a: À126.8
b: À119.6
c: À127.5
d: 6.38 tt
2J HF = 52.5
10
If partial fluorination along with isomerisation occur during
the ECPF of i-butyryl fluoride through cyclo-propane inter-
mediate, fluorine addition would occur at a and b position of
the resulting n-butyric acid (see Scheme 3, where R = COF).
Compounds 9 and 10 would thus be accounted by cyclo-
propane intermediate route. Compound 8 can however be
formed only through radical pathway without cyclo-propane
intermediate.
a: À126.98 s
b: À119.37 q
c: À80.77 t
[19,20]
3J F b,c = 8.6
11
12
a: À172.5
b: 4.83 d, b
c: 1.79 dd
2J HF = 47.6
3J HF = 4.6
3J HH = 2.3
a: À105.5
3J HF = 2.3
Methyl cleavage during ECPF of IBC leads to fluorine
substituted propionic acids. Five such compounds (12–16) were
identified in the product mixture. Methyl cleavage would
normally accompany fluorine substitution at the cleavage
position. Compound 14 namely CF3CH2COOH appears to be
an exception involving methyl cleavage and H atom absorption.
Compounds like 18, which are indeed low boiling are also
probably trapped in the perfluorinated liquid, leading to distinct
NMR spectral data.
13
14
b: 2.58 t
b: À64.4
3J HF = 10.9
2J Ha Fa = 52.5
a: 3.4 q
a: À215.9
c: À75.4
15
16
b: 6.35 dq
a: À127.3
b: À134.3 d
c: 6.58 tt
J2 Hb Fb = 53
2J Hb Fb = 53
3J Hb Fa = 5.4
The methyl radical formed during cleavage can combine
with intermediate molecules of IBC or NBC leading to the
formation of five carbon chains of the type (21), four carbon of
the type (22) could be identified in the product mixture.
All these data suggest that all commonly observed free
radical substitution processes also simultaneously occur during
the carbon chain isomerisation involving cyclo-propane
intermediates during the ECPF of IBC. Quiet interestingly
no product containing cyclopropyl group was observed among
these constituents (Table 2).
a: À121.77
b: À83.02
[20]
17
18
a: À215.9
2J HF = 41.8
3J HF = 5.8
c: À75.6
b: 6.14 d, sept
a: À75.5
À64.4
[21]
[21]
19
20
CF4

D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
1115
Scheme 4.
2.3. Minor constituents of fluorinated products obtained
during the ECPF of pivaloyl chloride
acid (25), perfluoro-3-methylbutyricacid (26), and perfluoro-n-
pentanoic acid (27) were also obtained as minor products.
Formation of such compounds would involve many free radical
steps including intra molecular rearrangements.Quite striking
support for ECPF process through cyclo-propane intermediate
comes from the observation of three cyclo-propane derivatives
(29)–(31) in the product mixture. As mentioned earlier (Section
1
The minor compounds characterised by 19F and H NMR
data from the ECPF product of pivaloyl chloride are presented
in Table 3. In addition to perfluoropivaloic acid (24) other
perfluoro carboxylic acids such as perfluoro(2-methyl)butyric
Table 3
19F and 1H NMR data of perfluorinated and polyfluorinated compounds obtained during the ECPF of pivaloyl chloride
No.
Structure
Chemical shift, d (ppm)
J (Hz)
Selectivity (mol%)
23
a: 1.1 s
24
25
a: À69.0
3.1
a: À160.0
b: À68.98
[17]
2J FcA,B = 293
c: À118.58, À119.2
d: À80.8
7.2
a: À124.6
b: À186.0
c: À72.19
26
1.0
a: À125.68
b: À123.4
c: À118.7
d: À80.8
[22]
[23]
12.8
27
28
a, d: À72.2
0.7
b, c: À178.97
a, d: À123.4
b À124.0
2J Fa,b = 282
[23,24]
29
e À72.5
9.8
2J Fa,b = 278.6
30
a, d: À119.83
b, c: À120.56
e: À181.47
[23,24]
3.4
31
À150.0 s
[25]
<0.1

1116
D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
Table 3 (Continued )
No.
Structure
Chemical shift, d (ppm)
J (Hz)
Selectivity (mol%)
4.5
a: À122.07, À122.69
b: À216.22
2J FaA,B = 291
2JHF = 45.4
32
d: À74.15
c: 6.27 d, m
a: À142.4
33
34
b: 1.66d
3J HF = 22.2
1.3
6.0
a: À187.1
b: À72.2
2J FaA,B = 282
c, f: À130.1
d, e: À130.7
[23]
35
À63.4
[23]
0.3
a: À181.3
2J HF = 40
3J HFa = 16
b: À73.3
5.3
c: À217.4
e: À73.8
36
2
d: 6.35 ddm
1.8
4
6
(CF2H)2–CF–COOH
<0.1
7.6
11
17
12
21.0
20
22
CF4
<0.1
1.0
3.1) pivaloyl chloride contains three methyl groups on the same
plane of the tetrahedral molecule. The formation of cyclo-
propane intermediate for such a trialkyl group is at least three
times higher than the dialkyl carbon structure. Hence cyclo-
propane structures involving C5 compound itself (29), C4
compounds with methyl group removal (30), and even cyclo-
propane itself (31) could be identified in the mixture. Formation
of perfluoro cyclo-propane and its derivatives are indicated in
Scheme 5. 19F NMR data shows that compounds such as 4, 6, 9,
11, 14, 17, 18, 19 and 22 are also formed during the ECPF of
pivaloyl chloride. In all the three cases very small quantities of
starting material has been found even after completing the
electrolysis (1, 7, 23).
employed. Products were collected in FEP traps kept at À70 8C
after passage through a condenser kept at À30 8C. Perfluorinated
carboxylic acid fluoride obtained was neutralised using 10%
aqueous sodium hydroxide in a doublewalled glass vessel kept at
À20 8C. Sodium salt of the perfluoro carboxylate was treated
with mineral acid to regenerate perfluoro carboxylic acid.
Electrochemical reactor temperature was varied between À10
and+10 8C. Noproduct couldbeisolatedfrom celldrains. Allthe
products were obtained only at the cold trap kept at À70 8C.
Gas chromatograms were recorded on a Hewlett Packard
6890 plus gas chromatograph. The temperature of the injector,
column (HP 1701, 30 m  0.32 mm  0.25 mm) and the
detector (FID) were held at 130, 90 and 250 8C, respectively.
Nitrogen was used as a carrier gas.
3. Experimental details
19F and 1H NMR spectra were recorded on a JEOL ECA 500
Spectrometer (500 MHz for 1H and 470 MHz for 19F,
respectively). Acetone-d6 (99.5 atom D, Aldrich) was used
3.1. General
1
as solvent. Chemical shifts for 19F and H NMR spectra were
A 200 ml stainless steel electrolysis cell with alternate nickel
anodes and cathodes (effective anode area = 2.47 dm2) was
reported in relative to CFCl3 and TMS, respectively. Negative
shifts are up field from the reference (CFCl3 and TMS).

D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
1117
Scheme 5.
Purity of anhydrous hydrogen fluoride (M/s TANFAC
Industries Ltd., Cuddalore, Tamil Nadu, India) was more than
99.9%. Reagent grade i-butyryl chloride, n-butyryl chloride and
pivaloyl chloride (M/s Merck Germany) were used.
radical generated in this step can undergo further fluorination,
cleavage, cyclisation or intermolecular interactions leading to
formation of a wide variety of partially fluorinated and
perfluorinated minor products. The second pathway (Scheme 2)
involving diradical process and cyclo-propane intermediate is
prevalent only when branched alkyl groups like i-propyl and t-
butyl groups are present in the reactant molecules. This
pathway leads to formation of n-alkyl products from starting
compounds containing i-propyl group. In the case compounds
containing t-butyl group, cyclo-propane intermediate pathway
is even more prevalent. Perfluoro cyclo-propane and its
derivatives are obtained as minor product in this process.
3.2. GC–MS results
GC–MS of mixture of perfluorinated compounds obtained
from the ECPF of i-butyryl chloride were measured using
Agilent 6890 GC with 5973N mass selective detector (Column
HP 5MS; 30 m  0.25 mm  0.25 mm) at 70 eV. The follow-
ing were the prominent products identified from m/z values.
Results indicate that heptafluorobutyric acid, pentafluor-
opropionic acid and trifluoroacetic acid were the main products.
In addition to the above, m/z peaks due to partially fluorinated
products were also obtained. GC–MS results are reproduced
below:
Acknowledgements
Thanks are due to CSIR, New Delhi for financial support
under CSIR-Task Force Network projects. Thanks are also due
to Ministry of Environment and Forests, Government of India
for their assistance.
(a) MS of perfluorobutyric acid. MS m/z (rel. int.): 197
[M À OH]+ (3.8), 169 [C3F7]+ (39.7), 150 [C3F6]+ (73.5),
131 [C3F4]+ (14.7), 119 [C2F5]+ (83.8), 100 [C2F4]+ (70.6),
69 [CF3]+ (100), 50 [CF2]+ (8.8), 45 [CO2H]+ (91.1) and 31
[CF]+ (14.7).
References
[1] P. Sartori, N. Ignat’ev, J. Fluorine Chem. 87 (1998) 157–162.
[2] V.S. Plashkin, L.N. Pushkina, S.V. Sokolov, Zh. Org. Khim. (Russ.) 10
(1974) 1215–1224 (Chem. Abstr. 18 (1974) 105221y).
[3] T. Abe, K. Kodaira, H. Baba, S. Nagase, J. Fluorine Chem. 12 (1978) 1–25.
[4] F.G. Drakesmith, D.A. Hughes, J. Appl. Electrochem. 9 (1979) 685–697.
[5] T. Abe, H. Baba, E. Hayashi, S. Nagase, J. Fluorine Chem. 23 (1983) 123–
146.
(b) MS of perfluoropropionic acid. MS m/z (rel. int.): 147
[M À OH]+ (7.0), 119 [C2F3]+ (50.4), 100 [C2F4]+ (7.0), 69
[CF3]+ (73.5), 50 [CF2]+ (14.0), 45 [CO2H]+ (100) and 31
[CF]+ (33.6).
(c) MS of trifluoroacetic acid. MS m/z (rel. int.): 97 [M À OH]+
(35.0), 69 [CF3]+ (73.5), 50 [CF2]+ (14.0), 45 [CO2H]+
(100), 31 [CF]+ (33.6) and 28 [CO]+ (28).
(d) MS of heptafluoropropane. MS m/z (rel. int.): 170 [M]+
(1.5), 151 [M À F]+ (2.3), 120 [C2F5H]+ (1.5), 101
[C2F4H]+ (1.5), 82 [C2F3H]+ (1.5), 70 [CF3H]+ (1.5), 69
[CF3]+ (100) and 51 [CF2H]+ (13).
[6] G.P. Gambaretto, M. Napoli, L. Conte, A. Scipioni, R. Armelli, J. Fluorine
Chem. 27 (1985) 149–155.
[7] M. Napoli, L. Conte, G.P. Gambaretto, F.M. Carlini, J. Fluorine Chem. 45
(1989) 213–224.
[8] T. Ono, Yo. Inoue, Yo. Arakawa, Yo. Naito, C. Fukayra, K. Yamanouchi,
K. Yokoyama, J. Fluorine Chem. 43 (1989) 67–85.
[9] T. Abe, E. Hayashi, H. Fukaya, H. Baba, J. Fluorine Chem. 50 (1990) 173–
196.
[10] T. Abe, H. Fukaya, E. Hayashi, Y. Hayakawa, M. Nishida, H. Baba, J.
Fluorine Chem. 66 (1994) 193–202.
4. Conclusions
[11] L. Conte, G.P. Gambaretto, J. Fluorine Chem. 125 (2004) 139–144.
[12] N. Bartlett, R.D. Chambers, A.J. Roche, R.C.H. Spink, L. Chacon, J.M.
Whalen, Chem. Commun. (1996) 1049–1050.
The present investigation suggests that free-radical pathway
involving single-electron transfer is the predominant mechan-
ism (Scheme 1) of electrochemical fluorination. The free
[13] P. Sartori, D. Velayutham, N. Ignat’ev, M. Noel, J. Fluorine Chem. 83
(1997) 1–8.

1118
D. Velayutham et al. / Journal of Fluorine Chemistry 127 (2006) 1111–1118
[14] P. Sartori, D. Velayutham, N. Ignat’ev, M. Noel, J. Fluorine Chem. 87
(1998) 31–36.
[19] C.D. Avignon, P. Calas, A. Commeyras, C. Amatore, J. Fluorine Chem. 57
(1991) 357–379.
[15] N. Ignat’ev, P. Sartori, J. Fluorine Chem. 101 (2000) 203–207.
[16] D. Velayutham, K. Jayaraman, M. Noel, S. Krishnamoorthi, P. Sartori, J.
Fluorine Chem. 115 (2002) 21–26.
[20] B.N. Huang, A. Haas, M. Lieb, J. Fluorine Chem. 36 (1987) 49–62.
[21] H. Gunther (Ed.), NMR Spectroscopy, An Introduction, John Wiley &
Sons, 1980.
[17] N.V. Ignat’ev, U.W. Biermann, U. Heider, A. Kucheryna, S.V. Absen, W.
Habel, P. Sartori, H. Willner, J. Fluorine Chem. 124 (2003) 21–37.
[18] J.G. Grasselli (Ed.), CRC Atlas of Spectral Data and Physical Constants
for Organic Compounds, CRC Press Inc., 1974.
[22] S.S. Malouet, H. Blancou, J. Iteir, A. Commeyras, Synthesis (1991) 647.
[23] T.A. Kestner, J. Fluorine Chem. 36 (1987) 77–82.
[24] T. Abe, M. Tamura, A. Sekiya, J. Fluorine Chem. 126 (2005) 325.
[25] V.W. Gash, D.J. Bauer, J. Org. Chem. 31 (1966) 3602–3607.