Electrochemical fluorination of aliphatic secondary amines

Article abstract of DOI:10.1016/S0022-1139(00)00267-0

Electrochemical fluorination (ECF) has been examined for six aliphatic secondary amines: N,N-di-ethylamine, N-ethyl,N-n-propylamine, N,N-di-n-propylamine, N-ethyl,N-iso-propylamine, N,N-di-iso-propylamine, and N-methyl,N-n-butylamine. It was found from these amines that not only the corresponding F-(N-fluoro-N,N-dialkylamines) but also F-imines having the same number of the carbon atoms were formed in low yields. The suppression of the C-N bond cleavage (blocking effect) which is expected to occur during fluorination due to the presence of bulky N-alkyl group was not observed as a result of the ECF of these aliphatic secondary amines. It was also found that the change of the initial solute concentration of N,N-di-n-propylamine did not affect on the product yields, which is usually observed for cyclic secondary amines. Several F-(N-fluoro-N,N-dialkylamines) were treated with triphenylphosphine for conversion into the corresponding F-imines. An imine bond was generated during this defluorination exclusively at the site of the alkyl group with a longer chain length when there were two different alkyl groups present in F-(N-fluoro-dialkylamines).

Full text of DOI:10.1016/S0022-1139(00)00267-0

Journal of Fluorine Chemistry 106 (2000) 35±42
Electrochemical ¯uorination of aliphatic secondary amines
Takashi Abe*, Eiji Hayashi, Hajime Baba
Fluorine Chemistry Laboratory, Department of Chemistry, National Industrial Research Institute of Nagoya, Hirate-cho 1-1,
Kita-ku, Nagoya 462-8510, Japan
Received 24 November 1999; received in revised form 7 March 2000; accepted 24 March 2000
Abstract
Electrochemical ¯uorination (ECF) has been examined for six aliphatic secondary amines: N,N-di-ethylamine, N-ethyl,N-n-propylamine,
N,N-di-n-propylamine, N-ethyl,N-iso-propylamine, N,N-di-iso-propylamine, and N-methyl,N-n-butylamine. It was found from these amines
that not only the corresponding F-(N-¯uoro-N,N-dialkylamines) but also F-imines having the same number of the carbon atoms were
formed in low yields. The suppression of the C±N bond cleavage (blocking effect) which is expected to occur during ¯uorination due to the
presence of bulky N-alkyl group was not observed as a result of the ECF of these aliphatic secondary amines. It was also found that the
change of the initial solute concentration of N,N-di-n-propylamine did not affect on the product yields, which is usually observed for cyclic
secondary amines. Several F-(N-¯uoro-N,N-dialkylamines) were treated with triphenylphosphine for conversion into the corresponding F-
imines. An imine bond was generated during this de¯uorination exclusively at the site of the alkyl group with a longer chain length when
there were two different alkyl groups present in F-(N-¯uoro-dialkylamines). # 2000 Elsevier Science S.A. All rights reserved.
Keywords: Electrochemical ¯uorination; Aliphatic secondary amines; F-(N-¯uoro-dialkylamines); F-imines; Triphenylphosphine; De¯uorination
1. Introduction
n-butylamine (6): (C2H5)2NH (1), C2H5(n-C3H7)NH (2),
C2H5(iso-C3H7)NH (3), (n-C3H7)2NH (4), (iso-C3H7)2NH
(5), CH3(n-C4H9)NH (6).
Many papers have dealt with the electrochemical ¯uor-
ination (ECF) of tertiary amines [1,2], because F-tert-
amines which are obtained as the main ¯uorination product
are used in applications, such as arti®cial blood, thermal
shock test liquids for semi-conductors and VPS liquids [3].
Per¯uorocarbons including F-tert-amines and F-ethers have
recently been a focus of attention as a new application for
FBS [4]. In contrast to the good ECF yield of F-tert-amines
by the ¯uorination of tertiary amines, poor yields of corre-
sponding F-(N-¯uoro-N,N-dialkylamines) have generally
been obtained in the case of secondary amines. Thus little
data is available on the ECF of aliphatic secondary amines
[5±7]. However, the ECF of cyclic secondary amines has
been well investigated, because this is the optimal method
for the preparation of F-(N-¯uoro-cyclic-alkylamines) [8±
10]. As a part of a series of investigations on the preparation
of F-amines [10], we have conducted the ¯uorination of
aliphatic secondary amines, and present the results herein of
the ¯uorination of N,N-di-ethylamine (1), N-ethyl,N-n-pro-
pylamine (2), N-ethyl,N-iso-propylamine (3), N,N-di-n-pro-
pylamine (4), N,N-di-iso-propylamine (5) and N-methyl,N-
2. Results and discussion
2.1. Fluorination of aliphatic secondary amines (1±6)
The results of the ¯uorination are summarized in Table 1.
It was found that not only F-(N-¯uoro-N,N-dialkylamines)
(A) but also F-imines (B) having the same number of the
carbon atoms as the former were formed in low yields
together with large quantities of cleaved products by ECF
of aliphatic secondary amines. Among the per¯uorocarbons
which are major products as a result of the cleave of the C±N
bond, a small amount of F-(N,N-di¯uoroalkylamine) with a
formula of RFNF2 was also formed (Scheme 1).
The combined yields of A and B were in the range of 3±
16%. The yields (A and B) and the ratio of A:B depended
greatly on the type of alkyl groups in the aliphatic secondary
amines. It is known that aliphatic F-imines (B) can be
prepared by various reactions involving, for example, pyr-
olysis of F-tert-amines [11], the reaction of nitrosyl¯uoride
and F-ole®ns [12], pyrolysis of alkali metal salts of F-
nitrogen-containing propionic acids [13], and the photolytic
* Corresponding author. Tel.: ‡81-52-911-2111; fax: ‡81-52-916-2802.
E-mail address: abe@nirin.go.jp (T. Abe).
0022-1139/00/$ ± see front matter # 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 0 0 ) 0 0 2 6 7 - 0

36
T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
Table 1
Results of the fluorination of secondary amines
Run
Sample g (mol)
Current
passed (Ah)
Fluorinated
product (g)
Fluoroamines and fluoroiminesa (yield %)
1
2
C2H5(C2H5)NH (1) 24.3 (0.443)
n-C3H7(C2H5)NH (2) 18.2 (0.386)
222
164
8.3b
5.0
C2F5N=CF(CF3) (30) (3.1), (C2F5)2NF (31) (6.2)
n-C3F7NF2 (15) (2.6), C2F5N=CF(C2F5) (22)
(2.4), n-C3F7(C2F5)NF (23) (2.7)
3
4
5
iso-C3H7(C2H5)NH (3)18.0 (0.386)
n-C3H7(n-C3H7)NH (4) 25.4 (0.252)
iso-C3H7(iso-C3H7)NH (5) 27.0 (0.386)
161
207
221
6.0
iso-C3F7NF2 (16) (2.3), iso-C3F7N=CF(CF3) (20)
(3.3), iso-C3F7(C2F5)NF (24) (2.4)
18.9 (0.9)c
21.0
15 (1.3), n-C3F7N=CF(C2F5) (13) (1.8), (n-
C3F7)2NF (12) (7.8)
16 (4.9), iso-C3F7N=CF(CF3) (17) (1.4), iso-
C3F7N=CF C2F5) (20)‡iso-C3F7N=C(CF3)2 (21)
(3.4), 17 (0.9), iso-C3F7(n-C3F7)NF (19) (6.0),
(iso-C3F7)2NF (18) (1.5)
6
n-C4H9(CH3)NH (6) 21.8 (0.386)
196
29.6
n-C4F9NF2 (32) (2.7), CF3N=CF(n-C3F7) (26)
(1.4), n-C4F9(CF3)NF (25) (1.2)
a Arranged in order of the elution time on GC.
b Products obtained in a cold trap.
c Products obtained as cell drainings.
reaction of RFNCl(CF2CFCl)Cl [14]. Recently, a convenient
preparative method which consists of the reaction of F-tert-
amines with SbF5 has been described by Petrov and Des-
Marteau [15].
Looking at the behavior of the carbon±nitrogen multiple
bond during ECF, it is known that the nitrile group of
acetonitrile can survive during ECF under a controlled
potential [16]. Thus, CF3CN is known to be formed at
low yields together with such cleaved products as
C2F5NF2, C2F6, C2F5H, NF3 by ECF of acetonitrile. How-
ever, the formation of F-imines (B) by ECF has never been
reported.
The reaction path for the formation of F-imines can be
explained by the ±HF reaction at the ®nal stage of the
¯uorination as is shown in a simpli®ed form in Scheme 2.
It is generally accepted that ECF starts from the C±H bond of
the carbon where the electron density is higher [17]. Thus,
the alkyl group of ammonium salts (7) of secondary amines
Scheme 1.
Scheme 2.

T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
37
Scheme 3.
will be ¯uorinated from the carbon which is farthest from the
nitrogen atom. As the reaction proceeds, partially ¯uori-
nated ammonium salts (8) of secondary amines which
survived without exhibiting C±N bond cleavage will form
a weak ammonium salt due to the decreasing basicity caused
by the inductive effect of ¯uorine. Partially ¯uorinated
secondary amines (9 and 10) which exist as a very weak
cationic species in AHF progresses to A. However, in some
parts of 9 and 10, the ¯uorine a of the nitrogen is eliminated
as HF resulting in ¯uorinated imines (11 and B). F-(N-
¯uoro-N,N-dilakylamines) (A) are also formed from these
imines as ¯uorination products. It is known that F-(2-
azapropene) easily takes on HF resulting in F-(N,N-
dimethylamine) with a formula of (CF3)2NH [18].
Therefore, the addition of HF to ¯uorinated imines (11
and B) to form partially ¯uorinated secondary amines (9 and
10) in AHF is considered to occur as a reversible inter-
conversion between them although the formation of F-(N,N-
dialkylamines) (9) was not con®rmed among the ECF
products in our investigation.
The results of the ¯uorination of N,N-di-n-propylamine
(4) are shown in Scheme 3 as a typical example of the ECF
of aliphatic secondary amines. F-(N-¯uoro,N,N-di-n-propy-
lamine) (12) and a F-imine, n-C3F7N=CF(C2F5) (13), which
retained the original framework of 4 but with a newly
formed ±N=CF± bond, were obtained as the corresponding
¯uorination products in low yields together with cleaved
products such as C3F8 (14) (major product) and n-C3F7NF2
(15).
The effect of the solute concentration of 4 on the yields of
¯uorination products was examined, because it is known that
ECF at a higher solute concentration is one of the most
effective ways for raising yields in the ¯uorination of tertiary
amines [19±21] and secondary cyclic amines [9]. The results
are shown in Table 2. However, it was found that changing
the solute concentration of 4 within the range of 2.5±8.9%
under comparable conditions did not improve the yield of
12.
In the case of the ECF of N,N-di-iso-propylamine (5), the
isomerization of the iso-propyl group to n-propyl group
occurred at high levels which resulted in the formation of
a complex mixture of products (Scheme 4). In addition, a
F-tert-amine, F-(N-ethyl,N-methyl,N-iso-propylamine) was
also obtained in small yields (Yˆ1.5%).
A small quantity of an isomerized product, F-(N-¯uoro,N-
iso-propyl,N-n-propylamine) (19) was also obtained
(Yˆ1.5%) together with the desired F-(N-¯uoro-di-iso-pro-
pylamine) (18) (6.0%). As a result of the isomerization of an
iso-propyl group, a F-imines mixture was formed, and iso-
C3F7N=C(CF3)2 (21) could not be isolated from the mixture
by means of GC. It was obtained only as an inseparable
mixture with its isomeric F-imine, iso-C3F7N=CF(C2F5)
(20). Therefore, the composition and determination of the
former F-imine (21) was conducted by comparing the
sample with an authenticated sample which was prepared
by reacting compound 18 with triphenylphosphine accord-
ing to a method described in [22] (also see the experimental
section (Scheme 3)). The formation of compound 21 is not
unprecedented. It has been prepared by the photolytic
reaction of per¯uoropropene with nitrosyl¯uoride by And-
reads [12].
It is known that 2-¯uoro and/or a 2-methyl substituent in a
pyridine ring inhibit the cleavage of the C±N bonds during
ECF, which results in the enhancement of the ECF yield of
pyridines with a substituent at the 2-position compared with
that of pyridine [22,23]. The reason for this reduction of the
Table 2
Effect of the solute concentration of amine 4 on the composition and the yield of products
Run
Sample
Current
Fluorinated
product (g)
Productsa (yield %)
concentration (wt.%)
passed (Ah)
1
2.5b
104
7.8
C3F8 (14) (16.8), n-c3F7NF2 (15) (0.3),
n-C3F7N=CF(C2F5) (13) (1.3), (n-C3F7)2NF (12) (5.8)
14 (20.1), 15 (1.3), 13 (1.8), 12 (7.8)
2
3
5.2c
8.9e
207
319
18.9 (0.9)d
28.6 (1.3)
14 (18.2), 15 (1.3), 13 (2.0), 12 (6.7)
a Arranged in order of the elution time on GC.
b 4 (12.8 g) was used.
c 4 (25.4 g) was used. The datum is duplicated with that given in Table 1.
d Products obtained as cell drainings.
e 4 (41.2 g) was fed. After 202 A h had been passed, 120 ml of AHF was added.

38
T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
Scheme 4.
cleavage in the case of 2-methyl substituent is explained by
the blocking effect on the C±N bond scission due to the
presence of methyl group(s) which sterically protects the C±
N bond. Therefore, it was expected that the degree of the
scission of the C±N bond of secondary amines would be
reduced when two bulky groups like di-iso-propylamine (5)
are present. However, the results of ECF of 5 showed that
such an effect was not observed when the F-(N-¯uoro,N,N-
di-iso-propylamine) (18) (Yˆ6.0%) was compared with that
of F-(N-¯uoro-N,N-di-n-propylamine) (12) (Yˆ6.2%)
obtained from 4.
From the ¯uorination of such compounds as 2, 3 and 6
which have different alkyl groups (R1 is not equal to R2 in
Scheme 1), not only corresponding F-(N-¯uoro-N,N-dialky-
lamines) (A) but also F-imines (B) were obtained in similar
low yields. Although the formation of two isomeric F-
imines (B) having the C=N bond at a different position
was expected to be formed from these initial materials, only
one B was formed. For example, from 2, an imine,
C2F5N=CF(C2F5) (22), was formed dominantly rather than
an imine with the molecular formula of n-C3F7N=CF(CF3)
(Scheme 5).
aliphatic secondary amines such as 3 which have a branched
N-iso-propyl group and a N-ethyl group (Scheme 6).
This result shows that an imine bond is apt to be formed at
the site of the secondary rather than the tertiary carbon
during ECF. The mechanism of the formation of F-imines by
the ECF of secondary amines seems to be very complex as
shown in Scheme 2. However, the reversible interconversion
in AHF between ¯uorinated imines (11 and B) and partially
¯uorinated secondary amines (9 and 10) seems to be an
important stage for the determination of the structure of
thermodynamically more stable F-imines as the ®nal pro-
ducts. It was thus surmised that the position of the ±N=C±
bond of the F-imine formed is governed by the alkyl group
of the starting aliphatic secondary amines; an imine bond
forms at the site of the longer alkyl chain length and at the
site of a secondary rather than a tertiary carbon, regiospe-
ci®cally.
2.2. The reaction of F-(N-fluoro-N,N-di-alkylamines) (A)
with triphenylphosphine
Several reactions have been known for compounds having
a N±F bond [24]. For example, the ¯uorine of N±F of F-(N-
¯uoro-N,N-dialkylamines) (A) oxidizes iodide ion [25],
Similarly, only one kind of imine, iso-C3F7N=CF(CF3)
(17), was formed as the sole F-imine at low yield from
Scheme 5.
Scheme 6.

T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
39
Scheme 7.
Scheme 8.
cyclopentadienyl iron [26,27] and dicumene chromium [26].
Furthermore, the ¯uorine of N±F of A reacts with triphe-
nylphosphine yielding F-imines (B) as a result of the
de¯uorination reaction [22].
We have investigated similar reactions using F-amines
(23, 18 and 25), which were separated from the ECF
products of 2, 5 and 6, respectively, with triphenylphosphine
in order to examine the product. The reaction proceeded
smoothly resulting in F-imines at good yield as shown in
Scheme 7.
It was found that only one isomic F-imine (22 and 26) was
formed from 23 and 25 which have two different N-alkyl
groups, regiospeci®cally. These results show that the
de¯uorination of A proceeds in the same way as expected
from the observation that a ±C=N± bond of F-imine is
formed at the site of the longer alkyl group.
It is considered that this reaction proceeds in a similar way
to that reported for the reaction of (Et2N)3P and CF3Br, for
example, [32]. Mechanistically, this reaction is considered to
proceed via the formation of the phosphonium salt 27, which
is in equilibrium with pentavalent intermediate 28, through a
`¯uorophilic' attack of the phosphorous center of (Ph)3P by
N±F of A (Scheme 8). Therefore, the de¯uorination step
from this intermediate (27 and 28) is considered to be the
key for the determination of the structure of B in this
reaction.
a Heise Burdon tube gauge was used for handling the volatile
compounds in the reaction of F-(N-¯uoro-di-alkylamines)
with triphenylphosphine and the determination of the vapor
pressure curve of (C2F5)2NF. Analytical GLC work was
carried out with a Shimadzu GC-2C gas chromatograph
using stainless steel columns (3 mm diameter) packed with
30% 1,6-bis(1H, 1H, 12H-F-decyloxy)hexane on Chromo-
sorb PAW/60±80 mesh (6.0 m) and a Shimadzu GC-14B gas
chromatograph using stainless steel columns (3 mm dia-
meter) packed with 25% Fomblin YR on Chromosorb PAW
(6.4 m). A Shimadzu GC-1C gas chromatograph using
stainless steel columns (10 mm diameter) packed with
30% 1,6-bis(1H, 1H, 12H-F-dodecyloxy)hexane on Chro-
mosorb PAW/60±80 mesh (6.0 m) was used for semi-pre-
parative work. The carrier gas was helium in all cases.
Infrared spectra were measured on a Hitachi EPI-G3 spec-
trometer using a 6 cm gas cell with KBr windows. 19F NMR
spectra were measured on a Hitachi 20B spectrometer
(56.4 MHz for 19F). Chemical shifts for 19F were reported
respective to CFCl3. Positive shifts are down®eld from the
reference. Mass spectra were measured on a Shimadzu GC/
MS-7000 instrument at 70 eV and a Shimadzu GC/MS-
QP5000 instrument ®tted with a capillary column (Neutra
Bond-1, 30 m long, 0.25 mm i.d., 1.5 mm thick).
3.1. Fluorination of N,N-diethylamine (1)
Sample 1 (24.3 g, 0.333 mol) was charged into a cell
containing 420 ml electrically puri®ed AHF, and the solu-
tion was subjected to ¯uorination with an anodic current
density of 3.7 A/dm2, a cell voltage of 5.2±5.5 V, and a
cell temperature of 7±88C over a period of 490 min
(222 Ah). At the ®nal stage of the ¯uorination, the voltage
reached 6.2 V.
3. Experimental details
Boiling points were uncorrected. All samples which were
subjected to ECF were used as received. The purity of
anhydrous hydrogen ¯uoride (AHF) (Daikin Industries
Co.) was more than 99.9%. The electrolytic ¯uorination
apparatus and operating procedures were similar to those
described previously [10]. Glass vacuum line equipped with
The ef¯uent gases from the cell were passed over NaF
pellets and then condensed in a trap cooled to 788C. The

40
T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
gaseous products which did not condense in the 788C trap
were then bubbled through a ¯uoropolymer bottle contain-
ing water and a gas washing bottle containing an aqueous
solution of a mixture of K2CO3, KOH and KI. All products
except new ones were identi®ed by comparison of their
infrared spectra and GLC retention times with those of
authenticated samples. When unknown compounds existed
among the ¯uorination products whose IR data were not
available in our laboratory, they were separated from other
products by use of semi-preparative GLC, and their struc-
tures were determined on the basis of 19F NMR and mass
spectra.
The products (8.3 g) condensed in the 788C trap con-
sisted of C2F6 (29) (0.3 g), C2F5N=CF(CF3) (30) [26]
(2.2 g), (C2F5)2NF (31) [26] (5.6 g) and unidenti®ed pro-
ducts (0.2 g). The GC yields of 30 and 31 were 3.1 and 6.2%,
respectively. Spectral data (IR and 19F NMR) of 30 and 31
are shown below.
3.3. Fluorination of N-ethyl,N-iso-propylamine (3)
Sample 3 (18.0 g, 0.207 mol) was ¯uorinated similarly
under the following conditions: 2.9 A/dm2, 5.7±6.0 V, 7±
88C, 161 Ah (388 min). The work-up gave the following
products in the 788C trap (6.6 g): C2F6 (29)‡C3F8 (14)
(0.6 g), iso-C3F7NF2 (16) (1.1 g), iso-C3F7N=CF(CF3) (17)
[28] (2.0 g), iso-C3F7(C2F5)NF (24) [29] (1.6 g), and uni-
denti®ed products (1.3 g). The GC yields of 20 and 24 were
3.3 and 2.4%, respectively. Spectral data of 16, 17 and 24 are
shown below.
iso-C3F7NF2 (16): IR (gas): 1320 (ms), 1300 (s), 1263
(vs), 1220 (ms), 1178 (s), 1131 (m), 1017 (s), 958 (m), 927
(ms), 740 (m).
iso-C3F7N=CF(CF3) (17): IR (gas): 1794 (ms) n(C=N),
1373 (w), 1219 (s), 1297 (ms), 1262 (vs), 1187 (s), 1113
(ms), 1097 (w), 1003 (s), 838 (w), 747 (w), 722 (w), 671 (m),
555 (w). 19F NMR (CDCl3): d 33.4 (m, 1F, =CF(CF3)), d
74.4 (d, 3F, JF±Fˆ2.0 Hz, ±CF(CF3)), d 79.2 (m, 6F, ±
CF(CF3)2), d 153.0 (d, 1F, JF±Fˆ26.6 Hz, ±CF(CF3)2).
iso-C3F7(C2F5)NF (24): bp 38±398C. IR (gas): 1311 (m,
sh), 1286 (s, sh), 1261 (vs), 1188 (ms), 1154 (m), 1138 (m),
1113 (m), 1079 (w), 1023 (w), 979 (w), 741 (w), 712 (w).
C2F5N=CF(CF3) (30): IR (gas): 1785 n(C=N), 1344 (s),
1255 (s), 1240 (vs), 1215 (vs), 1193 (s), 1111 (m), 1095 (m),
1072 (m), 850 (w), 745 (w), 700 (w), 19F NMR (CDCl3): d
29.0 (s, 1F, =CF), d 74.1 (m, 3F, =CCF3), d 86.3 (s, 3F,
CF3), d 97.8 (m, 2F, CF2).
‡ ‡
‡
‡
(C2F5)2NF (31): IR (gas): 1358 (w), 1227±1265 (vs), 1190
(vs), 1129 (ms), 1104 (s), 1086 (ms), 1027 (w), 857 (w), 820
(w), 740 (m), 701 (s), 651 (w), 533 (w). 19F NMR (CDCl3): d
82.9 (d, 6F, JF±Fˆ16.1 Hz, CF3), d 108.9 (d, 4F, JF±
MS: 264 [M±F] (36.2), 214 C4F8N (100), 169 C3F7
(18.1), 164 C3F6N (66.5), 119 C2F5 (48.8), 114 C2F4N
‡
‡
‡
‡
(25.2), 100 C2F4 (7.9), 69 CF3 (74.4). 19F NMR (CDCl3):
d
74.9 (t, d, 6F, JF±Fˆ8.7, 3.7 Hz, ±CF(CF3)2), d 83.5 (d,
ˆ23.1 Hz, CF2), d 92.2 (m, 1F, NF). The equation log P
3F, JF±Fˆ17.8 Hz, ±CF2CF3), d 89.9 (m, 1F, ±NF), d
108.2 (m, 2F, ±CF2CF3), d 158.6 (m, 1F, ±CF(CF3)).
F
(mm)ˆ7.84 1474/T describes the vapor pressure curve
from which DHn (6.8 kcal/mol), DSn (22.7 eu), and the
bp of 24.18C ([22]: bp 30.58C) are obtained.
3.4. Fluorination of N,N-di-n-propylamine (4)
3.2. Fluorination of N-ethyl,N-n-propylamine (2)
Sample 4 (25.4 g, 0.252 mol) was ¯uorinated similarly
under the following conditions: 2.9 A/dm2, 5.6±5.9 V, 7±
88C, 207 Ah (655 min). The products (18.9 g) condensed in
the 788C trap and the cell drained products (0.9 g) were
combined and were analyzed by GC similarly. The work-up
gave the following products: C3F8 (14) (9.5 g), n-C3F7NF2
(15) (0.7 g), n-C3F7N=CF(C2F5) (13) [29] (1.5 g), (n-
C3F7)NF (12) [30] (7.3 g) and unidenti®ed products
(0.8 g). The yields of 13 and 12 were 1.9 and 6.2%,
respectively. Spectral data of 15, 13 and 12 are shown
below.
Sample 2 (18.2 g, 0.209 mol) was ¯uorinated similarly
under the following conditions: 2.9 A/dm2, 5.8±6.1 V, 7±
88C, 164 Ah (393 min). The work-up gave the following
products in the 788C trap (5.0 g): C3F8 (14) (0.8 g),
C2F5N=CF(C2F5) (22) (1.3 g), n-C3F7(C2F5)NF (23)
(1.7 g), and unidenti®ed products (1.2 g). The yield of 22
and 23 were 2.4 and 2.8%, respectively. Spectral data of 22
and 23 are shown below.
C2F5N=CF(C2F5) (22): IR (gas): 1780 (s) n(C=N), 1344
(m), 1316 (ms), 1238 (vs), 1211 (vs), 1186 (s), 1133 (m),
1091 (ms), 1011 (m), 930 (w), 755 (w), 714 (w), 19F NMR
(CDCl3): d 22.5 (m, 1F, =CF(C2F5)), d 83.4 (d, t, 3F, JF±
n-C3F7NF2 (15): IR (gas): 1352 (m), 1312 (m), 1265 (vs),
1213 (ms), 1175 (m), 1145 (m), 1123 (w), 1020 (m), 939
(m), 885 (m), 800 (w), 746 (m).
ˆ5.1, 1.5 Hz, CF3CF2N=), d 86.4 (d, 3F, JF±Fˆ4.8 Hz,
n-C3F7N=CF(C2F5) (13): IR (gas): 1780 (s) n(C=N), 1345
(m), 1318 (m), 1293 (m), 1238 (vs), 1195 (s), 1175 (ms),
1138 (s), 1023 (m), 988 (ms), 968 (w), 829 (w), 751 (w), 730
(w). 19F NMR (CDCl3): d 23.0 (m, 1F, =CF(CF2F3)), d
81.4 (t, 3F, JF±Fˆ8.7 Hz, CF3F2CF2N), d 83.6 (m, 3F,
=CF(CF2CF3)), d 94.6 (m, 2F, CF3CF2CF2N), d 121.3
(d, 2F, JF±Fˆ13.5 Hz, =CF(CF2CF3)), d 129.7 (m, 2F,
CF3CF2CF2N)).
F
=C(F)CF2CF3), d 98.0 (d, 2F, JF±Fˆ19.7 Hz, CF3CF2N=),
d
121.1 (d, q, 3F, JF±Fˆ12.8, 1.5 Hz, =C(F)CF2CF3).
n-C3F7(C2F5)NF (23): IR (gas): 1352 (w), 1314 (m), 1256
(vs), 1206 (ms), 1180 (ms), 1136 (m), 1106 (m), 1029 (w),
986 (w), 744 (w), 714 (w), 693 (w). 19F NMR (CDCl3): d
82.4 (t, d, 3F, JF±Fˆ9.0, 4.0 Hz, CF2CF2CF3), d 83.3 (d,
3F, JF±Fˆ16.9 Hz, CF2CF3), d 106.9 (m, 2F, CF2CF2CF3),
d
108.7 (m, 2F, CF2CF3), d 127.2 (d, 2F, JF±Fˆ9.7 Hz,
(n-C3F7)2NF (12): IR (gas): 1624±1356 (ms), 1302 (s),
1226±1256 (vs), 1204 (s), 1181 (s), 1161 (s), 1031 (s), 1012
CF2CF2CF3).

T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
41
(s), 991 (ms), 971 (m), 826 (w), 788 (w), 746 (m), 729 (m),
706 (ms), 651±681 (w). 19F NMR (CDCl3): d 90.4 (m, 1F,
NF), d 82.1 (m, 6F, CF3), d 106.1 (m, 4F, CF3CF2CF2), d
126.9 (m, 4F, CF3CF2CF2).
(iso-C3F7)2NF (18): bp 72.8±73.58C, n2D0<1.28, d420
1.7610. IR (gas): 250±1278 (vs), 1223 (m), 1196 (s),
1166 (s), 1122 (s), 1080 (m), 1013 (m), 987 (s), 963
‡
(ms), 800 (w), 750 (s), 708 (s), 532 (w). MS: 371 M
‡
‡
(1.1), 352 [M±F] (1.1), 314 C6F12N (7.0), 302
C5F12N (34.6), 264 C3F12 (15.9), 214 C4F8N (11.0),
‡ ‡ ‡
3.5. Fluorination of 4 at a lower solute concentration
‡
‡
‡
‡
169 C3F7 (36.8), 164 C3F6N (11.1), 119 C2F5 (12.5),
‡
‡
Sample 4 (12.8 g, 0.126 mol) was ¯uorinated under simi-
lar conditions as above except for the amount of the sample
fed: 3.0 A/dm2, 6.2±6.7 V, 7±88C, 104 A h (283 min). The
products (7.8 g) condensed at 788C were similarly ana-
lyzed by GC. The work-up gave the following products: 14
(4.0 g), 15 (0.1 g), 13 (0.5 g), 12 (2.7 g) and unidenti®ed
products (0.5 g). The yields of 14, 15, 13 and 12 were 16.8,
0.3, 1.2 and 5.8%, respectively.
114 C2F4N (6.0), 100 C2F4 (6.1), 76 C2F2N (3.1), 69
‡
CF3 (100). 19F NMR (CDCl3): d 85.1 (m, 1F, NF), d
75.5 (s, 12F, (CF3)2CF±), d 155.3 (s, 2F, (CF3)2CF±).
3.8. Fluorination of N-methyl,N-n-butylamine (6)
Sample 6 (21.8 g, 0.251 mol) was ¯uorinated similarly
under the following conditions: 2.9 A/dm2, 6.1±7.1 V, 7±
88C, 196 Ah (434 min). The work-up gave the following
products in the 788C trap (29.6 g): C4F10 (25.1 g), n-
C4F9NF2 (32) [26] (2.2 g), CF3N=CF(n-C3F7) (26) [31]
(1.1 g), n-C4F9(CF3)NF (25) (0.8 g), and unidenti®ed
products (0.4 g). The GC yields of 26 and 25 were 1.4
and 1.2%, respectively. Spectral data of 25, 26 and 32 are
shown below.
n-C4F9NF2 (32): IR (gas): 1366 (ms), 1310 (m, sh), 1293
(ms, sh), 1265 (s, sh), 1255 (vs), 1237 (s, sh), 1203 (m), 1168
(m), 1148 (ms), 1116 (m), 1095 (w), 1065 (w), 990 (w), 947
(m), 892 (ms), 830 (w), 795 (w), 742 (ms).
3.6. Fluorination of 4 at a higher solute concentration
Sample 4 (41.2 g and 0.408 mol) was ¯uorinated under
the similar conditions as above except for the amount of the
sample fed and the addition of 120 ml of AHF after 202 Ah
had passed: 3.0 A/dm2, 5.9±7.1 V, 7±88C, 320 Ah (877 min).
The products (28.5 g) condensed in the 788C trap and the
cell drained products (1.3 g) were combined and were
similarly analyzed by GC. The work-up gave the following
products: 14 (14.0 g), 15 (1.2 g), 13 (2.8 g), 12 (8.9 g) and
unidenti®ed products (1.8 g). The yields of 14, 15, 13 and 12
were 18.2, 1.3, 2.0 and 6.7, respectively.
CF3N=CF(n-C3F7) (26): IR (gas): 1777 (m) n(C=N), 1350
(ms), 1287 (ms), 1256 (s, sh), 1230 (vs), 1176 (m), 1158 (w),
1140 (w), 1103 (w), 1026 (m), 978 (w), 823 (w), 742 (w),
651 (w). 19F NMR (CDCl3): d 23.3 (m, 1F, =CF(n-C3F7)),
3.7. Fluorination of N,N-di-iso-propylamine (5)
d
57.9 (d, 3F, JF±Fˆ15.0 Hz, CF3N=), d 81.2 (t, d, 3F,
Sample 5 (27.0 g, 0.267 mol) was ¯uorinated similarly
under the following conditions: 3.5 A/dm2, 5.7±6.2 V, 7±
88C, 221 Ah (580 min). The work-up gave the following
products in the 788C trap (21.0 g): C2F6 (29)‡C3F8 (14)
(1.6 g), iso-C3F7NF2 (16) (2.9 g), iso-C3F7N=CF(CF3) (17)
(1.1 g), iso-C3F7N=CF(C2F5) (20)‡iso-C3F7N=C(CF3)2
(21) (1.0 g), n-C3F7N=CF(C2F5) (13) (0.8 g), iso-
C3F7NCF3(C2F5) [28] (1.5 g), iso-C3F7(n-C3F7)NF (19)
(2.0 g), (iso-C3F7)2NF (18) (6.0 g), and unidenti®ed pro-
ducts (4.1 g). The GC yield of 18 was 5.3%. The physico-
chemical properties and spectral data of 19 and 18 are shown
below.
JF±Fˆ6.5, 6.8 Hz, =C(F)CF2CF2CF3), d 118.8 (q, d, 2F,
JF±Fˆ9.6 Hz, =C(F)CF2CF2CF3),
JF±Fˆ9.6 Hz, =C(F)CF2CF2CF3).
d
127.0 (d, 2F,
n-C4F9(CF3)NF (25): IR (gas): 1309 (s, sh), 1281 (vs),
1248 (vs), 1236 (s, sh), 1188±1218 (s, sh), 1146 (ms), 1101
(w), 1063 (w), 1032 (m), 915 (w), 861 (m), 814 (w), 743
(ms), 710 (w). 19F NMR (CDCl3): d 88.0 (m, 1F, NF), d
68.3 (d, t, 3F, JF±Fˆ13.5 Hz, CF3N), d 82.0 (t, t, 3F, JF±F
ˆ0.4, 1.7 Hz, CF2CF2CF2CF3),
d
108.7 (m, 2F,
CF2CF2CF2CF3), d 124.1 (m, 2F, CF2CF2CF2CF3), d
127.6 (m, 2F, CF2CF2CF2CF3).
iso-C3F7(n-C3F7)NF (19): bp 74.2±74.88C, n2D0<1.28, d420
1.7260. IR (gas): 1357 (m), 1235±1290 (vs), 1205 (s), 1179
(s), 1157 (s), 1129 (ms), 1106 (m), 1030 (ms), 997 (s), 800
3.9. Reaction of F-(N-fluoro-N-ethyl,N-n-propylamine)
with triphenylphosphine
‡
(w), 742 (ms), 654 (w), 539 (w). MS: 352 [M±F] (36.5),
In a 50 ml round bottom ¯ask which contained 1.99 g of
P(Ph)3, F-(N-¯uoro,N-ethyl,N-n-propylamine) (0.78 g,
2.43 mmol), which had been puri®ed by GC among the
¯uorination products obtained from 6, was condensed using
a vacuum line, and the contents were allowed to stand
overnight. The volatile products (0.59 g), which was sepa-
rated from the solid residue, were obtained as clear volatile
compounds. By studying the spectroscopic analysis, it was
determined to be C2F5N=CF(C2F5) (22). The yield for 22
was 86.0%.
‡
‡
‡
‡
‡
314 C6F12N (34.9), 302 C5F12N (21.4), 264 C5F10N
‡
‡
(47.1), 252 C4F10N (10.0), 214 C4F8N (16.2), 169 C3F7
‡
‡
(38.7), 164 C3F6N (13.6), 119 C2F5 (22.2), 114 C2F4N
‡
‡
‡
(7.2), 100 C2F4 (6.8), 76 C2F2N (6.1), 69 CF3 (100), 50
‡
CF2 (3.9). 19F NMR (CDCl3): d 89.2 (m, 1F, NF), d
75.4 (s, 6F, (CF3)2CF±), d 82.6 (t, d, 3F, JF±Fˆ9.0,
4.2 Hz, CF3CF2CF2±), d 105.5 (m, 2F, CF3CF2CF2±),
d 126.5 (d, 2F, CF3CF2CF2±), d
(CF3)2CF±).
157.8 (m, 1F,

42
T. Abe et al. / Journal of Fluorine Chemistry 106 (2000) 35±42
[2] N.L. Weinberg, in: N.L. Weinberg (Ed.), Technique of Electroorganic
Synthesis, Part II, Wiley/Interscience, New York, 1975, p. 1.
[3] R.E. Banks, Fluorocarbons and Their Derivatives, MacDonald & Co.
Ltd., London, 1970, p. 70.
3.10. Reaction of F-(N-fluoro-N,N-di-iso-propylamine)
with triphenylphosphine
The reaction of F-(N-¯uoro-N,N-di-iso-propylamine)
(0.87 g, 2.35 mmol) and P(Ph)3 (1.99 g, 7.59 mmol) was
conducted similarly. The work-up of the product yielded
0.58 g of iso-C3F7N=C(CF3)2 (21) (Yˆ74.0%). The physi-
cochemical properties and spectral data of 21 are shown
below.
iso-C3F7N=C(CF3)2 (21): bp 50.7±51.08C, n2D0<1.28, d420
1.6117. IR (gas): 1746 n(C=N), 1337 (ms), 1310 (s), 1250±
1280 (vs), 1220 (vs), 1187 (m), 1171 (w), 1143 (ms), 1099
(ms), 998 (s), 987 (ms, sh), 743 (m), 733 (m), 687 (m), 655
(w). 19F NMR of this compound shows a very peculiar
spectra at room temperature due to the slow conversion
between two tutomers, which has been published by And-
reads [12]. 19F NMR (CDCl3): d 78.4 (m, 6F, (CF3)2CF±),
[4] I.T. Horvath, J. Rabai, Science 266 (1994) 72.
[5] J.H. Simons (to 3M), US Patent 2,519,983 (1950).
[6] B. Chang, H. Yanase, K. Nakanishi, N. Watanabe, Electrochim. Acta
16 (1971) 1179.
[7] K. Omori, S. Nagase, H. Baba, K. Kodaira, T. Abe, J. Fluorine Chem.
9 (1977) 279.
[8] T.C. Simons, F.W. Hoffmann, R.B. Beck, H.V. Holler, T. Katz, R.J.
Koshar, R.E. Larson, J.E. Mulvaney, K.E. Paulson, F.E. Rogers, B.
Singleton, R.E. Sparks, J. Am. Chem. Soc. 79 (1957) 3429.
[9] H. Baba, T. Abe, E. Hayashi, Nippon Kagaku Kaishi (J. Chem. Soc.
Jpn., Chemistry and Industrial Chemistry) (1986) 1249.
[10] T. Abe, H. Fukaya, T. Ono, E. Hayashi, I. Soloshonok, K. Okuhara, J.
Fluorine Chem. 87 (1997) 193.
[11] W.H. Pearlson, L.J. Hals, US Patent 2,643,267 (1953).
[12] S. Andreads, J. Org. Chem. 27 (1962) 4163.
[13] T. Abe, E. Hayashi, T. Shimizu, Chem. Lett. (1989) 905.
[14] C. Sarwar, R.L. Kirchmeier, J.M. Shreeve, Inorg. Chem. 28 (1989)
2187.
d 149.2 (m, 1F, (CF3)2CF±), d 68.0 (broad, 6F,
=C(CF3)2.
[15] V.A. Petrov, D.D. DesMarteau, Inorg. Chem. 31 (1992) 3776.
[16] M. Haruta, N. Watanabe, J. Fluorine Chem. 7 (1976) 159.
[17] G.P. Gambaretto, M. Napoli, L. Conte, A. Scipioni, R. Armelli, J.
Fluorine Chem. 27 (1985) 149.
3.11. Reaction of F-(N-fluoro-N-methyl,N-n-butylamine)
with triphenylphosphine
[18] R.D. Dresdner, F.N. Tlumac, J.A. Young, J. Am. Chem. Soc. 86
(1964) 5892.
The reaction of F-(N-¯uoro,N-methyl,N-n-butylamine)
(25) (1.01 g, 3.15 mmol) and P(Ph)3 (1.95 g, 7.43 mmol)
was conducted similarly. The work-up of the product yielded
0.61 g (2.16 mmol) of CF3N=CF(n-C3F7) (26) as the
de¯uorination product (Yˆ68.6%).
[19] L. Conte, M. Napoli, G.P. Gambaretto, J. Fluorine Chem. 30 (1985) 89.
[20] T. Abe, E. Hayashi, H. Baba, H. Fukaya, J. Fluorine Chem. 48 (1990)
257.
[21] T. Abe, E. Hayashi, H. Fukaya, H. Baba, J. Fluorine Chem. 50 (1990)
173.
[22] V.J. Davis, R.N. Haszeldine, A.E. Tipping, J. Chem. Soc. (1975)
1263.
[23] P. Sartori, D. Valayutham, N. Ignat'ev, M. Noel, J. Fluorine Chem. 87
(1998) 31.
Acknowledgements
[24] G.S. Lal, G.P. Pez, R.G. Syvret, Chem. Rev. 96 (1996) 1737.
[25] R.E. Banks, E. D. Burling, J. Chem. Soc. (1965) 6077.
[26] R.A. Mitsch, J. Am. Chem. Soc. 87 (1965) 328.
[27] R.E. Banks, M.G. Barlow, M. Nickko-Amiry, J. Fluorine Chem. 14
(1979) 383.
The authors wish to thank Mr. Fumihiro Sato (Aichi
Institute of Technology) for the measurement of vapor
pressure curve of (C2F5)2NF. A part of this investigation
was supported by the New Energy and Industrial Technol-
ogy Development Organization (NEDO).
[28] K.E. Peterman, J.M. Shreeve, Inorg. Chem. 14 (1975) 1223.
[29] V.A. Petrov, G.G. Belen'kii, L.S. German, Izv. Akad. Nauk SSSR.
Ser. Khim. (1985) 1934.
[30] J.B. Hynes, L.A. Bigelow, B.C. Bishop, P. Bandyopdhyay, J. Am.
Chem. Soc. 85 (1963) 83.
References
[31] B.C. Bishop, J.B. Hynes, L.A. Bigelow, J. Am. Chem. Soc. 85 (1963)
1606.
[32] I. Ruppert, K. Schlich and W. Volbach, Tetrahedron Lett. (1984)
2195.
[1] S. Nagase, in: P. Tarrant (Ed.), Fluorine Chemistry Reviews, Vol. 1,
Marcel Dekker, New York, 1967, p. 1.