Electrophilic aromatic fluorination with fluorine: meta-Directed fluorination of anilines

Article abstract of DOI:10.1016/j.jfluchem.2005.02.013

Anilines are mainly or selectively fluorinated in the meta-position with F₂ when dissolved in triflic acid, sometimes in the presence of small quantities of antimony pentafluoride. The regioselectivity is increased when an electron-donating substituent is present at the para-position.

Full text of DOI:10.1016/j.jfluchem.2005.02.013

Journal of Fluorine Chemistry 126 (2005) 661–667
www.elsevier.com/locate/fluor
Electrophilic aromatic fluorination with fluorine:
meta-Directed fluorination of anilines
*
J.P. Alric, B. Marquet, T. Billard, B.R. Langlois
´ ˆ
Laboratoire SERCOF (UMR CNRS 5181), Universite Claude Bernard-Lyon1, Bat. Chevreul,
43 Bd du 11 Novembre 1918, 69622 Villeurbanne, France
Received 5 October 2004; accepted 15 February 2005
Dedicated to Prof. R.D. Chambers for his 70th birthday.
Abstract
Anilines are mainly or selectively fluorinated in the meta-position with F2 when dissolved in triflic acid, sometimes in the presence of small
quantities of antimony pentafluoride. The regioselectivity is increased when an electron-donating substituent is present at the para-position.
# 2005 Elsevier B.V. All rights reserved.
Keywords: Anilines; Fluorine; Fluoroanilines; Triflic acid; Fluorination
1. Introduction
in the difficulty to control the direct fluorination of organic
compounds, which was claimed, for a long time, to be a very
exothermic radical chain process. Indeed, homolytic
dissociation of fluorine is very easy (F–F bond
energy = 159 kJ molÀ1) [11] and predominates when no
polarization of this molecule is induced by its environment.
This was especially the case when almost apolar fluorotri-
chloromethane (CFC-11), which is outstandingly inert to F2,
was chosen as solvent [12–14].
Fluoroaromatics constitute an important class of fluori-
nated compounds since they are key-intermediates in the
manufacture of prominent pharmaceuticals and agrochem-
icals [1–4]. Usually, they arise from the nucleophilic
substitution of diazonium moieties with hydrogen fluoride or
that of activated halides with fluoride anions [5–10]. In
principle, electrophilic fluorination of unsubstituted aro-
matic nuclei with fluorine could provide shorter routes to
fluoroaromatics.
However, during the last decade and especially under the
impetus given by Chambers et al., important work has been
devoted to the development of new conditions allowing the
polarization of fluorine, and thus the tamed electrophilic
fluorination of organic substrates. Since this time, a steadily
growing number of papers have appeared [15–24]. Con-
cerning electrophilic aromatic fluorination [25–30], an
outstanding contribution has been given again by Chambers
et al., who demonstrated that aromatics bearing an electron-
donating group in para-position to an electron-withdrawing
group, can be cleanly fluorinated with F2 at room
temperature, provided that a strong acid, such as 98%
sulfuric or formic acid, is used as solvent [20,31,32]. Such
solvents are polar enough and, more importantly, protic
enough to strongly polarize fluorine and avoid uncontrolled
Nevertheless, the use of fluorine has been hindered, for a
long time, for two main reasons. The first one is the
hazardous storage and handling of this reagent. Such a
problem can be now solved, at least on the laboratory scale,
by the commercial availability of bench electrolysers
(‘‘Fluorodec1’’ from Fluorogas Ltd.), able to deliver
fluorine on demand, even at low rates, without any storage
of this corrosive gas. Moreover, safety is improved by
diluting the fluorine thus generated, with nitrogen (typically,
a 10:90, v/v, F2/N2 mixture is used). The second problem lies
* Corresponding author. Tel.: +33 472 448 163; fax: +33 472 431 323.
E-mail address: Bernard.Langlois@univ-lyon1.fr (B.R. Langlois).
0022-1139/$ – see front matter # 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2005.02.013

662
J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
Scheme 1. Fluorination of disubstituted benzenes [20,31,32].
Scheme 3. Fluorination of acetanilide with acetyl hypofluorite [35].
radical processes. This explains why the reaction can be
carried out at room temperature. Of course, because of the
aromatic substitution pattern, a completely regioselectivity
was observed (Scheme 1).
from fluorine and sodium acetate [35]. In this case, ortho-
fluoroaniline was the major product, may be (but it is not still
clear), because of an interaction between the nitrogen atom
and the fluorinating agent, at least in the transition state
(Scheme 3).
Curiously, few fluorinations have been carried out with
perfluoroalkanesulfonic acids as solvents or additives. For
example, Hoechst patented results, very similar to those of
Chambers et al., from fluorination in perfluorobutanesulfo-
nic acid [22], whereas Coe et al. fluorinated fluorobenzene,
at low temperature in fluorotrichloromethane, in the
presence of small quantities of triflic acid (CFC-
11:TfOH = 95:5 to 90:10) [33] (Scheme 2). It should be
noticed that, probably because of the rather low protic
character of this medium, chemical selectivity is limited.
Another strategy could be to protonate anilines strongly
enough to protect them against oxidation. Moreover, it
could be anticipated that protonated anilines should be
fluorinated mainly at the meta-position, because of the
inductive electron-withdrawing effect of the ammonium
substituent.
Thus, we studied the fluorination of aniline itself in strong
acidic media, namely pure sulfuric acid, triflic acid or triflic
acid with several added Lewis acids [B(OTf)3, SbF5, SbCl5,
1–4 mol%]. The results are summarized in Table 1.
As resulting from Table 1, aniline was fluorinated, as
expected, without extensive oxidation or degradation and
meta-fluoroaniline was always predominant over its ortho-
and para-isomers. This indicates that aniline was protonated
to a large extent during the reaction. However, the results
were dependent on the medium composition. Sulfuric acid
(entry 1) was not the most suitable solvent since, though
conversion of 1a was rather high (74%), global selectivity
(51%) and chemical balance (64%) were not satisfactory.
Possibly, water-soluble by-products, arising from sulfona-
tion and oxidation by H2SO4, could have been formed and
eliminated from the organic phase during the aqueous work
up. Better results were obtained in triflic acid (entry 2),
which is not an oxidant. In this case, aniline was probably
protonated more efficiently. Indeed, reactivity was lower (as
indicated by a lower conversion), but the reaction was
cleaner, as indicated by a higher global selectivity (70%) and
a more satisfying balance of aromatic compounds (80%).
Additional proof was brought by a higher meta-regioselec-
tivity (m-2a:p-2a:o-2a = 1.5:1.2:1 with TfOH compared to
1.3:1.1:1 with H2SO4).
2. Results and discussion
Fluoroanilines are important building blocks for the
manufacture of numerous bioactive products. ortho- and
para-fluoroaniline are usually produced, on the industrial
scale, through a two-step nitration-reduction of fluoroben-
zene. meta-Fluoroaniline is less easily available, but can be
prepared by a cine-substitution of ortho-chlorofluoroben-
zene with sodium amide in liquid ammonia [34]. Thus, it
would be interesting to examine the possibility to obtain
fluoroanilines from direct electrophilic fluorination of
anilines.
Of course, the free amino group, as well as the
methylamino moiety, cannot survive the action of such an
oxidant as F2, as demonstrated by Purrington and Woodward
during the fluorination of N-methylaniline, even in the
presence of boron or aluminum trichlorides [29]. This is the
reason why Rozen and co-workers fluorinated acetanilide,
and not aniline, with acetyl hypofluorite, generated in situ
The addition of boron triflate (1 mol%, entry 3) did not
change dramatically the conversion, the chemical balance
and the isomer ratio but resulted in a lower selectivity. This
additive, known as one of the strongest Lewis acids [36],
could have activated the fluorine atoms in p-2a and o-2a to
promote their substitution by free anilines that leads to tarry
Scheme 2. Fluorination of fluorobenzene [33].

J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
663
Table 1
Fluorination of aniline in acidic medium
Entry
Solvent
Lewis acid (mol%)a
Conv. 1a (%)b
Selectivities (%)c
Balance (%)d
m-2a
p-2a
o-2a
Total
1
2
3
4
5
6
a
H2SO4
TfOH
TfOH
TfOH
TfOH
TfOH
None
74
59
62
48
61
59
20
28
24
31
28
19
16
23
21
27
21
17
15
19
15
20
15
8
51
70
60
78
64
44
64
80
77
86
71
67
None
B(OTf)3 (1)
SbF5 (1)
SbF5 (4)
SbCl5 (4)
vs. solvent.
b
c
d
Conv. = [consumed 1a (mol)/introduced 1a (mol)] Â 100.
Selectivity = [produced 2a (mol)/consumed 1a (mol)] Â 100.
P
Balance = {[ produced 2a (mol) + recovered 1a (mol)]/introduced 1a (mol)} Â 100.
polynuclear products. The addition of antimony pentachlor-
ide (4 mol%, entry 6) was not beneficial, especially
concerning selectivity and chemical balance. SbCl5 could
be suspected to act as a chlorinating agent; nevertheless, no
chlorinated product was detected by gas phase chromato-
graphy but our apparatus was not very sensitive. Finally,
small quantities of antimony pentafluoride (1 mol%, entry 4)
provided the best results. SbF5 is known to enhance the
acidity of anhydrous protic acids [36]. Thus, aniline was
more extensively protonated than in pure triflic acid.
Consequently, its conversion was lower but the reaction
was cleaner (chemical balance = 86%), more selective
(global selectivity = 78%) and slightly more regioselective
(m-2a:p-2a:o-2a = 1.6:1.4:1). The enhanced selectivity
could also arise from the fact that, as the medium became
more acidic, polarization of the fluorine molecule was more
efficient and elimination of FÀ was facilitated. Conse-
quently, the electrophilic process was less contaminated by
radical or redox side-reactions. When the SbF5 content was
increased to 4 mol% (entry 5), the conversion of aniline
increased dramatically (up to 61%), but at the same time, the
selectivity and the chemical balance decreased, probably
because of side-reactions induced by the strong acidity of the
medium. Finally, the best results, in terms of chemical
selectivity and regioselectivity, were obtained in triflic acid
with 1 mol% SbF5 added. This result may be connected to
results reported by Firnau and co-workers concerning the
fluorination of tyrosine: in pure hydrogen fluoride,
fluorination occurred at ortho-position to the hydroxyl
group whereas in HF/BF3, fluorine was mainly introduced at
the meta-position [37,38].
fluorination of aniline was compared with that of
nitrobenzene under the same conditions (Table 2).
These results were consistent with the rules governing
electrophilic aromatic substitution. The nitro group, which is
not sensitive at all to F2 and exhibits more electron-
+
withdrawing power than the NH3 group, decreased the
conversion, but selectivity and chemical balance were
complete. Moreover, the regioselectivity of meta-fluorina-
tion was far better. Thus, it appeared that the NH3+ moiety,
which is a purely inductive electron-withdrawing substi-
tuent, is not so efficient as the nitro group, which acts as an
inductive and mesomeric electron-withdrawing substituent,
to direct fluorination predominantly enough to the meta-
position. In order to force fluorination to occur at the meta-
position (relative to NH2), we then used substrates bearing,
in para-position to the amino group, an electron-donating
substituent, by analogy with Chambers’ strategy. The first
substrate we examined was 4-chloroaniline (Table 3).
From Table 3, results demonstrate that, when the quantity
of fluorine increased, conversion of 1b increased steadily,
but global selectivity and chemical balance decreased.
Correspondingly, the amount of difluorinated products 3b
increased. Among these difluorinated products, 2,5-difluoro-
4-chloroaniline 2,5-3b was predominant and could be
produced with a yield similar to that of m-2b with three
equivalents of fluorine. 2,5-3b obviously resulted from
fluorination of both m-2b and o-2b, whereas 3,5-3b was
produced from m-2b only. Of course, fluorination of m-2b
+
and o-2b was deactivated by the inductive effect of NH3
that also acts as a meta-directing substituent. Nevertheless,
2,5-3b was produced about three times faster than 3,5-3b
(entry 4), since the formation of 2,5-3b either from m-2b or
o-2b, was activated by the synergistic conjugative effects of
In order to get additional proof of the electrophilic
character of these reactions and to evaluate their interest,

664
J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
Table 2
Comparison of the fluorinations of aniline and nitrobenzene
Entry
Z
Conv. PhZ (%)a
Selectivities (%)b
Balance (%)c
m-F
p-F
o-F
2,5-F2
Total
1
2
a
NH2
NO2
59
36
28
62
23
9
19
20
0
9
70
80
100
100
Conv. = [consumed ArZ (mol)/introduced ArZ (mol)] Â 100.
P
Selectivity = [ fluorinated products (mol)/consumed ArZ (mol)] Â 100.
b
c
Balance = {[Sfluorinated products (mol) + recovered ArZ (mol)]/introduced ArZ (mol)} Â 100.
chlorine and fluorine, whereas the formation of 3,5-3b was
activated by the conjugative effect of chlorine only.
Moreover, it is known that fluorine activates the electrophilic
substitution of aromatic nuclei more efficiently than
chlorine. The observation that 2,5-3b was more abundant
than 3,5-3b is also consistent with the fact that the
conjugative ortho- and para-directing effect of chlorine
and fluorine is more efficient than the inductive meta-
Thus, the same conditions were used to compare the
fluorination of different anilines substituted by an electron-
donating group at the para-position. The results are
summarized in Table 4.
In all cases, selectivities and chemical balance were good
to excellent. As expected from the rules governing
electrophilic aromatic substitution, conversion of the
substrate decreased when the inductive electron-withdraw-
ing power of the substituent increased. However, meta-
regioselectivity, relative to NH2, increased with the
conjugative electron-donating power of the substituent.
Thus, the direct fluorination of readily available 4-
substituted anilines with F2 provides a rapid access to a
variety of meta-fluoroanilines. Some of them, prepared in
the present work (Scheme 4), are very important for the
design of bioactive molecules and are not easy to synthesize
by other routes. Nevertheless, the separation of these
valuable products, by distillation from the crude mixture,
+
directing effect of NH3 .
The use of 1.2 equivalent of fluorine (entry 2) constituted
the best compromise in terms of conversion, chemical
selectivity, regioselectivity and chemical balance. Under
these conditions, an excellent chemical balance (96%) was
reached with an acceptable conversion (50%), in terms of
production, and the amount of difluorinated products 3b
remained rather low (9% of the fluorinated products).
Moreover, the meta-regioselectivity was excellent (m-2b:o-
2b = 74:8).
Table 3
Fluorination of 4-chloroaniline
b
Entry
F2 (eq)
Conv. 1b (%)a
Selectivities (%)
Balance (%)c
m-2b
o-2b
3,5-3b
2,5-3b
2,3-3b
2,6-3b
Total
3b/2b
1
2
3
4
a
0.5
1.2
2.0
3.0
15
50
78
74
57
23
15
8
0
1
3
6
0
6
0
2
4
7
0
0
0
5
92
91
81
60
0
11
98
96
86
68
75
7
10
16
27
100
3
131
Conv. = [consumed 1b (mol)/introduced 1b (mol)] Â 100.
P
b
c
Selectivity = [produced 2b or 3b (mol)/consumed 1b (mol)] Â 100 (estimated by 19F NMR).
Balance = {[ produced 2b and 3b (mol) + recovered 1b (mol)]/introduced 1b (mol)} Â 100.

J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
665
Table 4
Fluorination of substituted anilines
Entry
R
Conv. 1 (%)a
Selectivity (%)b
Balance (%)c
P
m-2
o-2 +
3
Total
1
2
3
4
5
6
a
H (1a)
Cl (1b)
F (1c)
Me (1d)
OMe (1e)
2,4-F2 (1f)
59
50
30
57
38
33
28
74
74
56
68
78
42
17
8
70
91
82
70
76
91
80
96
89
80
86
94
14
8
13
Conv. = [consumed 1 (mol)/introduced 1 (mol)] Â 100.
P
b
c
Selectivity = [produced 2 or 3 (mol)/consumed 1 (mol)] Â 100 (estimated by 19F NMR).
Balance = {[ produced 2 and 3 (mol) + recovered 1 (mol)]/introduced 1 (mol)} Â 100.
should need a deeper chemical engineering study to define
the required conditions and was not carried out in the present
work.
verted substrates were estimated by GC using 4-fluoroto-
luene as standard (and also by 19F NMR for 4-fluoro- and
2,4-difluoro-aniline). If necessary, HPLC was carried out on
an apparatus fitted with a column filled with Hypersil silica
gel (length: 250 mm, diameter: 4.6 mm) and a UV–vis
detector. Eluent flow: 1 mL minÀ1. Mass spectrometry was
carried out under electron impact at 70 eV and eventually
coupled with GC.
3. Experimental
TLC analyses were carried out on silica gel (Kieselgel
60F254) deposited on aluminum plates, detection being done
by UV (254 nm). Flash-chromatographies were performed
on silica gel Geduran SI 60. Unless stated otherwise, NMR
Trifluoromethanesulfonic acid was distilled prior use and
kept under nitrogen. Other reagents were used as received or
dried according to literature [39]. Fluorine was generated by
a Fluorodec1 electrolyzer (from Fluorogas Co.), fitted with
a cartridge filled with KFÁ2 HF and heated up to 85 8C to
melt the electrolyte. The electric plugs of the electrolyzer
were connected to an electric generator delivering a
stabilized intensity (typically: 0.25 A), which governed
fluorine flow. The anodic gas outlet of the fluorine generator
was connected to a sodium fluoride cartridge (to trap
residual HF), then to a gas mixture system, for dilution of F2
with dry N2 (F2:N2 = 1:9), and finally to a cylindrical glass
reaction vessel (capacity: 40 mL), bearing internal counter-
flow baffles, and fitted with an efficient mechanical stirrer, a
thermometer, a gas inlet and a gas outlet. This gas outlet was
connected to a sulfuric acid trap then to an alumina cartridge
to trap residual fluorine. Hydrogen produced at the cathodic
outlet of the fluorine generator was directed to an efficient
hood.
1
spectra were recorded in CDCl3. H NMR spectra were
recorded at 300 MHz and 13C NMR spectra at 75 MHz. The
substitution patterns of the different carbons were deter-
mined by a ‘‘DEPT 135’’ sequence. 19F NMR spectra were
recorded at 188 or 282 MHz. Chemical shifts (d) are given in
ppm versus TMS (1H, 13C) or CFCl3 (19F) used as internal
references. Coupling constants are given in Hertz. Crude
yields were determined by 19F NMR versus 4-fluorotoluene
used as standard. GC was carried out on an apparatus fitted
with an apolar semi-capillary column (length: 15 m,
diameter: 0.53 mm, film thickness (DB1): 1 mm) or a polar
semi-capillary column (length: 15 m, diameter: 0.53 mm,
film thickness (VA-WAX): 1 mm) and a catharometric
detector. Helium flow: 25 mL minÀ1. Amounts of uncon-
3.1. General procedure
Before switching on the power, the whole apparatus
(electrolyzer + reaction vessel) was swept with dry N2 to
expel oxygen, which could react violently with F2. Then,
electricity was switched on and F2/N2 (1.2 equivalent)
Scheme 4. 3-Fluoro-4-substituted anilines.

666
J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
introduced, under efficient stirring, into a solution of the
substrate (2 mmol) in triflic acid (40 mL). After reaction, the
anodic compartment of the fluorine generator was purged
with dry N2, to evacuate remaining fluorine into the reaction
flask. Then, the reaction mixture was poured on ice,
neutralized with 10% NaOH until basic and extracted with
dichloromethane (4 Â 50 mL). The organic phase was dried
over MgSO4 and CH2Cl2 was evaporated at room
temperature under vacuum. If necessary, the coloured oil
or the solid thus obtained was separated from tarry by-
products by a very rapid chromatography on silica gel. The
spectroscopic data (19F NMR, GC–MS) of the resulting
products were compared, in the reaction mixture, to those of
authentic commercial compounds. The crude mixture was
also compared, by gas chromatography or HPLC, to a
mixture of these commercial products.
2,3,4-F3-aniline: 19F NMR d: À150.2 (m), À156.79 (broad
s), À162.12 (broad s).
References
[1] R. Filler, Y. Kobayashi, L.M. Yagulpolskii, Organofluorine Com-
pounds in Medicinal Chemistry and Biomedical Applications, Else-
vier, Amsterdam, 1993.
[2] R.E. Banks, B.E. Smart, J.C. Tatlow, Organofluorine Chemistry: Prin-
ciples and Commercial Applications, Plenum Press, New York, 1994.
[3] T. Welch, S.E. Ewarakrishnan, Fluorine in Biorganic Chemistry, John
Wiley & Sons, New York, 1991.
[4] A. Becker, Inventory of Industrial Fluoro-Biochemicals, Eyrolles,
Paris, 1996.
[5] M. Hudlicky, A.E. Pavlath, Chemistry of Organic Fluorine Com-
pounds II. A Critical Review, ACS Monograph 187, American Che-
mical Society, Washington, DC, 1995, pp. 1092–1100.
[6] L.R. Subramanian, G. Siegemund, in: B. Baasner, H. Hagemann, J.C.
Tatlow (Eds.), Organo-Fluorine Compounds. Houben-Weyl: Methods
of Organic Chemistry, vol. E10a, Thieme, Stuttgart, 1999, pp. 548–596.
[7] B.R. Langlois, in: B. Baasner, H. Hagemann, J.C. Tatlow (Eds.),
Organo-Fluorine Compounds. Houben-Weyl: Methods of Organic
Chemistry, vol. E10a, Thieme, Stuttgart, 1999, pp. 686–740.
[8] T. Hiyama, Organofluorine Compounds: Chemistry and Properties,
Springer-Verlag, Berlin, 2000, pp. 137–182.
Fluorination of aniline 1a (186 mg, 2 mmol).
Crude mixture: 203 mg.
m-2a: 19F NMR d: À113.9 (m); MS m/z: 111 (Mꢀ+, 100), 85,
84.
p-2a: 19F NMR d: À127.6 (broad s); MS m/z: 111 (Mꢀ+,
100), 85, 84.
[9] J.H. Clark, D. Wails, T.W. Bastock, Aromatic Fluorination, CRC
Press, Boca Raton, 1996.
o-2a: 19F NMR d: À136.1 (m); MS m/z: 111 (Mꢀ+, 100), 85,
84.
Fluorination of 4-chloroaniline 1b (255 mg, 2 mmol).
Crude mixture: 290 mg.
[10] J.S. Moilliet, in: R.E. Banks, B.E. Smart, J.C. Tatlow (Eds.), Organo-
fluorine Chemistry: Principles and Commercial Applications, Plenum
Press, New York, 1994, pp. 195–219.
m-2b: 19F NMR d: À116.2 (m); MS m/z: 145 (Mꢀ+, 100).
o-2b: 19F NMR d: À133.1 (m); MS m/z: 145 (Mꢀ+, 100).
3,5-3b: 19F NMR d: À114.9 (broad d, 3JHF = 7.6); MS m/z:
163 (Mꢀ+, 100).
[11] R.E. Banks, J.C. Tatlow, Fluorine: The First Hundred Years (1886–
1986), Elsevier Sequoia, Lausanne, 1986.
[12] R.J. Lagow, J.L. Margrave, Prog. Inorg. Chem. 26 (1979) 161.
[13] R.F. Merritt, T.E. Stevens, J. Am. Chem. Soc. 88 (1966) 1822–1823.
[14] V. Grakauskas, J. Org. Chem. 34 (1969) 2835–2839.
[15] J. Hutchinson, G. Sandford, in: R.D. Chambers (Ed.), Organofluorine
Chemistry, Techniques and Synthons, Springer, Berlin, 1997, pp. 1–
43.
2,5-3b: 19F NMR d: À140.2 (m), À157.0 (m); MS m/z: 163
(Mꢀ+, 100).
2,3-3b: 19F NMR d: À121.7 (m), À139.4 (m); MS m/z: 163
(Mꢀ+, 100).
[16] J.S. Moilliet, J. Fluorine Chem. 109 (2001) 13–17.
[17] S. Rozen, in: G.A. Olah, R.D. Chambers, G.K.S. Prakash (Eds.),
Synthetic Fluorine Chemistry, John Wiley & Sons, New York, 1992,
pp. 143–161(Chapter 7).
2,6-3b: 19F NMR d: À138.0 (broad d, 3JHF = 6.2); MS m/z:
163 (Mꢀ+, 100).
[18] S.T. Purrington, B.S. Kagen, T.B. Patrick, Chem. Rev. 86 (1986) 997–
1018.
Fluorination of 4-fluoroaniline 1c (222 mg, 2 mmol).
Crude mixture: 231 mg.
[19] G. Resnati, Tetrahedron 49 (1993) 9385–9445.
[20] R.D. Chambers, C.J. Skinner, J. Hutchinson, J. Thomson, J. Chem.
Soc. Perkin Trans. 1 (1996) 605–609.
m-2c: 19F NMR d: À138.7 (m), À153.2 (m).
o-2c: 19F NMR d: À125.8 (m), À132.2 (m).
Fluorination of 4-methylaniline 1d (214 mg, 2 mmol).
Crude mixture: 216 mg.
[21] R.D. Chambers, J. Hutchinson, G. Sandford, J. Fluorine Chem. 100
(1999) 63–73.
[22] A.G. Hoechst, US Patent US, 5,756, 834 (1996).
[23] S. Rozen, Chem. Rev. 96 (1996) 1717–1736.
m-2d: 19F NMR d: À117.9 (m).
o-2d: 19F NMR d: À136.3 (m).
[24] G.S. Lal, G.P. Pez, R.G. Syvret, Chem. Rev. 96 (1996) 1737–1756.
[25] V. Grakauskas, J. Org. Chem. 35 (1970) 723–728.
[26] F. Cacace, A.P. Wolf, J. Am. Chem. Soc. 100 (1978) 3639–3641.
[27] F. Cacace, P. Giacomello, A.P. Wolf, J. Am. Chem. Soc. 102 (1980)
3511–3515.
3,5-3d: 19F NMR d: À116.1 (m).
2,5-3d: 19F NMR d: À123.6 (m), À142.3 (m).
Fluorination of 4-methoxyaniline 1e (246 mg, 2 mmol).
Crude mixture: 252 mg.
[28] S. Misaki, J. Fluorine Chem. 17 (1981) 159–171;
S. Misaki, J. Fluorine Chem. 21 (1982) 191–199.
3
4
m-2e: 19F NMR d: À134.8 (dd, JHF = 12.6, JHF = 9.2).
o-2e: 19F NMR d: À131.5 (d, JHF = 12.6).
[29] S.T. Purrington, D.L. Woodward, J. Org. Chem. 56 (1991) 142–145.
[30] M. Napoli, L. Conte, G.P. Gambaretto, C. Fraccaro, E. Legnaro, F.M.
Carlini, J. Fluorine Chem. 54 (1991) 240;
3
3
3,5-3e: 19F NMR d: À130.0 (d, JHF = 10.3).
Fluorination of 2,4-difluoroaniline 1f (258 mg, 2 mmol).
Crude mixture: 269 mg.
M. Napoli, L. Conte, G.P. Gambaretto, C. Fraccaro, E. Legnaro, F.M.
Carlini, J. Fluorine Chem. 60 (1993) 19–25;
2,4,5-F3-aniline: 19F NMR d: À138.22 (m), À144.00 (m),
M. Napoli, L. Conte, G.P. Gambaretto, C. Fraccaro, E. Legnaro, F.M.
Carlini, J. Fluorine Chem. 70 (1995) 175–179.
À149.00 (m).

J.P. Alric et al. / Journal of Fluorine Chemistry 126 (2005) 661–667
667
[31] R.D. Chambers, C.J. Skinner, J. Thomson, J. Hutchinson, J. Chem.
Soc. Chem. Commun. (1995) 17–18.
[35] O. Lerman, Y. Tor, S. Rozen, J. Org. Chem. 46 (1981) 4629–4631.
[36] G.A. Olah, G.K.S. Prakash, J. Sommer, Superacids, John Wiley &
Sons, New York, 1985 (Chapters 1 and 2).
[32] R.D. Chambers, J. Hutchinson, M.E. Sparrowhawk, G. Sandford, J.S.
Moilliet, J. Thomson, J. Fluorine Chem. 102 (2000) 169–173.
[33] P.L. Coe, A.M. Stuart, D.J. Moody, J. Chem. Soc. Perkin Trans. 1
(1998) 1807–1812.
[37] R. Chirakal, G. Firnau, E.S. Garnett, J. Fluorine Chem. 45 (1989) 125.
[38] R. Chirakal, K.L. Brown, G. Firnau, E.S. Garnett, D.W. Hughes, B.G.
Sayer, R.W. Smith, J. Fluorine Chem. 37 (1987) 267–278.
[39] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemicals,
second ed., Pergamon, New York, 1982.
[34] N.N. Vorozhtsov Jr., G.G. Yakobson, T.D. Rubina, Dokl. Akad. Nauk
SSSR 127 (1959) 1225–1227.