A facile route to aryl amines: Nucleophilic substitution of aryl triflates

Article abstract of DOI:10.1055/s-1999-2886

The aromatic nucleophilic substitution (S(N)Ar) between aryl triflates and secondary amines has been studied. In the absence of solvent, the reaction proceeds at room temperature for nitro and cyano activated aryl triflates and requires higher temperatures in the case of carboxy activation. Variable triflate reactivity could be explained in terms of frontier molecular orbital theory. This methodology has been applied for the synthesis of substituted piperidyl pyridines.

Full text of DOI:10.1055/s-1999-2886

LETTER
1559
A Facile Route to Aryl Amines: Nucleophilic Substitution of Aryl Triflates
Laurent Schio*,a Guy Lemoine,b Michel Klicha
a Medicinal Chemistry, b Molecular Modelling, Hoechst Marion Roussel ; 102, Route de Noisy, 93235 Romainville Cedex, France
Fax +33 (0) 1 49 91 50 87; E-mail: laurent.schio@hmrag.com
Received 24 June 1999
Abstract: The aromatic nucleophilic substitution (SNAr) between
aryl triflates and secondary amines has been studied. In the absence
of solvent, the reaction proceeds at room temperature for nitro and
cyano activated aryl triflates and requires higher temperatures in the
case of carboxy activation. Variable triflate reactivity could be ex-
plained in terms of frontier molecular orbital theory. This method-
ology has been applied for the synthesis of substituted piperidyl
pyridines.
EWG: NO2, CN, OTf, PhCO
reactions.3, 4 In 1990, Kotsuki et al.3 described the reaction
between electron poor aryl triflates and various amines
(mainly secondary).
Key words: Triflate, SNAr, amination, aryl amine, frontier orbitals
This methodology for the preparation of aromatic amines
is very attractive since triflates are readily available from
the corresponding phenols.5 However, the experimental
conditions reported by Kotsuki are rather drastic with long
reaction times in refluxing acetonitrile and high pressures.
In parallel to Kotsuki’s work, nucleophilic displacement
of triflate by amines has been extensively investigated in
Aryl triflates have been widely used in transition metal ca-
talysed coupling reactions with organometallic reagents.1
Recently, palladium complexes have been shown to be ef-
fective catalysts for the amination of aryl triflates.2 How-
ever, very little attention has been devoted to the study of
displacement of the triflate group by amines in SNAr type
a The yields are given for isolated products and refer mainly to single runs. b In a typical procedure, a mixture of 1 equiv. triflate and 5 equiv.
piperidine is stirred at RT or 100°C until completion of the reaction (monitored by TLC). The crude mixture is purified by flash chromatogra-
1
phy on silica gel to afford aryl amine as pure compound (satisfactory H & 13 C NMR, IR and mass spectral data). c 2.5 equiv. piperidine. d Lit.
mp3: 101-104°C. e Not determined.
Synlett 1999, No. 10, 1559–1562 ISSN 0936-5214 © Thieme Stuttgart · New York

1560
L. Schio et al.
LETTER
our laboratory. We report herein the scope and limitations Morpholine, N-methyl piperazine and tetrahydropiperi-
of this reaction as a synthetic tool and from a mechanistic dine gave satisfactory yields in the absence of solvent (see
point of view. An application in medicinal chemistry is Table 2). However, we observed two limitations to the
presented as well.
scope of the reaction:
Noting the detrimental effect of dilution in kinetics of sec- – With a bulky amine such as diisopropylamine, no reac-
ond order reactions, preliminary experiments were con- tion occurred and the starting material was recovered.
ducted in the absence of solvent. Indeed, we observed that
– In the case of octylamine, phenol 9 was isolated as the
aryl triflates react with piperidine at room temperature in
main product. According to MNDO calculations,9 octy-
neat conditions, when activated with the nitro or nitrile
lamine is a harder nucleophile than piperidine. As a result,
group (Table 1, entries 4 and 5). In the case of an ester
nucleophilic attack occurs at the harder electrophilic site
group, the reaction requires higher temperatures (ª 100
of the molecule, the sulfur atom, leading to aminolysis of
the triflate moiety.
°C) and improved yields were observed with hindered es-
ters that minimise amidation as a competing process (en-
tries 2 and 3). Moreover, it is noteworthy that
naphthyltriflate 3 undergoes amination (albeit in poor
yield) even though activated by a remote ester group (en-
try 7).
Surprisingly, 2- and 4-cyanophenyltriflates 5 and 1e ex-
hibited similar reaction times at room temperature (entries
5 and 9). The para attack might have been expected to be
faster considering steric hindrance in the ortho position.
In addition, it has been reported that amination of haloni-
trobenzenes proceeds much faster at the para position in
agreement with optimised HOMOamine - LUMOaryl halide in-
teractions when compared to the ortho reaction.6 There-
fore, we embarked on semi-empirical molecular orbital
calculations in order to estimate the contributions of fron-
tier orbital versus coulombic interactions in triflate reac-
tivity. LUMO energy (EEWG) and electron deficiency at
the reaction site (d-EWG) were calculated using MNDO and
CNINDO programmes respectively.7 Reactivity of select-
ed electron poor triflates (kEWG) was assessed from de-
tailed kinetic studies.8
Interestingly, the relative reactivity of triflates 1b, 1d, 1e
a The yields are given for isolated products and refer to single runs .b
24h at room temperature. c 48h at room temperature. d Reactions were
performed with 5 equiv. amine and followed by TLC. After comple-
tion, crude mixtures were purified by flash chromatography on silica
gel. All the isolated compounds gave satisfactory analytical data (1H
and 13 C NMR, IR and MS). e 8b: mp = 110°C. f 8c: mp = 84°C, lit.
and 5 was found to be correlated with softness. The lower
the LUMO, the faster the reaction: kNO2 (ENO2 = -3.655
eV) > korthoCN (EoCN = -3.416 eV) ª kparaCN (EpCN = -3.466
eV) > kCO2CHPh2 (E = -3.262 eV).
Moreover, 2- and 4-cyanophenyltriflates turned out to
possess LUMOs of equivalent energy in agreement with mp3: 85°C.
the observed similar reaction times. On the other hand, no
correlation was obtained regarding electron deficiency
With developed methodology for the synthesis of aryl
calculated at the reaction site. For example, it appeared
from detailed kinetics studies that the amination rate is 60
times higher in the case of 4-nitrotriflate 1d when com-
pared to 4-cyanotriflate 1e whereas the electron deficien-
cy is very similar (d-NO2 = 0.215, d-CN = 0.202). This study
leads to the conclusion that the reactivity of aryl triflates
towards piperidine is primarily influenced by frontier or-
bitals interactions rather than coulombic forces.
amines, we investigated the amination of aryl triflate 1010
in relation to a medicinal project.
Next, we investigated the reactivity of 4-cyanophenyltri-
flate 1e towards various amines (eq.2).
Triflate 10 turned out to be very reactive towards several
functionalised amines (Table 3). Indeed, detailed kinetic
studies revealed that 10 reacts with piperidine 10 times
faster than 4-cyanophenyltriflate 1e (and hence 6 fold
slower than 4-nitrophenyltriflate 1d). According to CN-
INDO calculations, 10 is a softer electrophile than 1e and
Synlett 1999, No. 10, 1559–1562 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
A Facile Route to Aryl Amines: Nucleophilic Substitution of Aryl Triflates
1561
is harder than 1d regarding their LUMO energy.11 Fur-
thermore, 10 exhibits a reduced electron deficiency at the
reaction site (d- = 0.177) when compared to 1e and 1d con-
firming predominant frontier orbital interactions in triflate
reactivity.
References and Notes
(1) Ritter, K. Synthesis 1993, 735. For recent papers on cross-
coupling reactions with various organometallic reagents, see
for example: Baston, E.; Hartmann, R. W. Synth. Commun.
1998, 28, 2725; Olofsson, K.; Larhed, M.; Hallberg, A. J. Org.
Chem. 1998, 63, 5076; Kamikawa, T.; Hayashi, T. Synlett
1997, 163; Badone, D.; Cardamone, R.; Guzzi, U.
Tetrahedron Lett. 1994, 35, 5477; Quesnelle, C. A.; Familoni,
O. B.; Snieckus, V. Synlett 1994, 349.
(2) For excellent recent reviews see: Hartwig, J.F. Synlett 1997,
329; Hartwig, J. F. Angew. Chem. Int. Ed. Engl. 1998, 37,
2046; Frost, C.G.; Mendonca, P. J. Chem. Soc., Perkin Trans.
1 1998, 2615.
(3) Kotsuki, H.; Kobayashi, S.; Suenaga, H.; Nishizawa, H.
Synthesis 1990, 1145.
(4) For the nucleophilic amination of N2-aromatic triflates see:
Chayer, S.; Essassi, E. M.; Bourguignon, J. J. Tetrahedron
Lett. 1998, 39, 841; Cappelli, A.; Anzini, M.; Vomero, S.;
Mennuni, L.; Makovec, F.; Doucet, E.; Hamon, M.; Bruni, G.;
Romeo, M. R.; Menziani, M. C.; De Benedetti, P. G.; Langer,
T. J. Med. Chem. 1998, 41, 728; Arcadi, A.; Cacchi, S.;
Fabrizi, G.; Manna, F.; Pace, P. Synlett 1998, 446;
Steinbrecher, T.; Wameling, C.; Oesch, F.; Seidel, A. Angew.
Chem. Int. Ed. Engl. 1993, 32, 404; Megati, S.; Goren, Z.;
Silverton, J. V.; Orlina, J.; Nishimura, H.; Shirasaki, T.;
Mitsuya, H.; Zemlicka, J. J. Med. Chem. 1992, 35, 4098;
Saari, W. S.; Halczenko, W.; King, S. W.; Huff, J. R.; Guare
Jr, J. P.; Hunt, C. A.; Randall, W. C.; Anderson, P. S.; Lotti,
V. J.; Taylor, D. A.; Clineschmidt, B. V. J. Med. Chem. 1983,
26, 1696.
a 5 equiv. unless otherwise indicated. b Heating was necessary to ob-
c
d
e
tain a homogeneous medium. 2 equiv. amine. 11a: mp=121°C.
11b: mp=128°C. f 11f: mp=143°C.
(5) A typical procedure is as follows: To a mixture of 4-
hydroxybenzonitrile (5g, 42 mmol) in dichloromethane (200
ml) was added at 0°C triethylamine (8.7 ml, 62.4 mmol) and
then slowly trifluoromethanesulfonic anhydride (8.4 ml, 49.9
mmol). The resulting mixture was stirred 2 h at 0°C then
quenched with water. The organic layer was separated,
washed with water and dried over anhydrous magnesium
sulfate. After filtration and concentration at reduced pressure,
the crude mixture was purified by flash chromatography on
silica gel (diethyl ether/cyclohexane: 20/80 v/v) to afford 1e
(8.46 g, 80 % yield). TLC, SiO2, ether/cyclohexane: 1/1 (v/v),
Rf = 0.48. IR (CHCl3, n cm-1): 2235 (CN); 1430, 1140
(OSO2CF3). 1H NMR (250 MHz, CDCl3, ppm): d 7.80, 7.43
(4H, AA’BB’ system). Analyses for the elements C, H, F, N,
S were within 0.2% of theoretical values.
(6) For a comprehensive textbook on frontier orbital theory, see:
Fleming, I. in Frontier Orbitals and Organic Chemical
Reactions, Wiley, London, 1976. For a discussion on aromatic
nucleophilic substitution with aryl halides see chapter 3, 68-69
and references cited therein.
(7) Low energy conformations were identified using force field
calculations (in-house software). Following MNDO or
CNINDO/2 calculations (programmes implemented in QCPE
package), relevant frontier orbitals were assigned based on
electronic density at the reaction site.
In the case of racemic trans-dihydroxy piperidine 7g,12 it
was necessary to add DMF because of the heterogeneity
of the reaction medium. This resulted in a low yield (17
%) of aminated product 11g. The yield was dramatically
improved by protecting the two hydroxyl groups as tert-
butyl dimethylsilyl ethers. Accordingly, the reaction
could be carried out in neat conditions and the racemic
product 11h was isolated in 75 % yield.13
In conclusion, we have demonstrated that the reaction be-
tween secondary amines and electron deficient aryl tri-
flates does not require drastic conditions as reported. In
the absence of solvent, aryl triflates undergo nucleophilic
substitution with various amines either at room tempera-
ture or at 60-100°C depending on the activating substitu-
ent. Kinetic studies have shown that the reactivity of
triflates towards secondary amines is mainly controlled by
frontier orbitals interactions rather than coulombic forces.
In addition, we have successfully applied our methodolo-
gy to the preparation of functionalised piperidylpyridines.
(8) Only triflates reacting cleanly (yield > 70%) were considered.
Amination reactions were performed at 25°C using a
thermostated water bath. Product formation and triflate
consumption were followed by HPLC.
(9) EHOMO piperidine = -10.066 eV; EHOMO octylamine = -10.435
eV
Acknowledgement
We are grateful to the Analytical Department (HMR, Romainville)
for performing the spectral analyses and to Branislav Musicki and
John Weston (Medicinal Chemistry Department) for helpful discus-
sions.
(10) Prepared according to the following Scheme:
Scheme
Synlett 1999, No. 10, 1559–1562 ISSN 0936-5214 © Thieme Stuttgart · New York

1562
L. Schio et al.
LETTER
(11) Calculated LUMO energies with CNINDO: E1b = 0.0349 eV ;
10 = 0.0622 eV ; E1e = 0.0799 eV.
(12) Takemura, S.; Miki, Y.; Uono, M.; Yoshimura, K.; Kuroda,
M.; Suzuki, A. Chem. Pharm. Bull. 1981, 29, 3026.
(13) Preparation: A mixture of triflate 10 (340 mg, 0.78 mmol) and
amine 7h (542 mg, 1.56 mmol) was stirred 6h at 100°C and
then purified by flash chromatography on silica gel with a 15/
85 (v/v) diethyl ether / cyclohexane mixture to give 11h (371
mg, 75% yield). TLC, SiO2, ether/cyclohexane 2/8 (v/v): Rf =
0.40. IR (CHCl3, n cm-1): 1716, 1577, 1555, 1495, 1258, 835.
1H NMR (250 MHz, CDCl3, ppm): d 0.04, 0.05, 0.08 and 0.09
(s, 3H), 0.77 and 0.91 (s, 9H), 1.53 (dq, J = 13 Hz, J= 4 Hz,
1H), 2.12 (m, 1H), 3.4-3.6 (m, 5H), 3.71 (m, 1H), 7.06 (dd, J
= 9 Hz, J= 3 Hz, 1H), 7.18 (s, 1H), 7.25-7.50 (m, 10 H), 8.02
(d, J = 9 Hz, 1H), 8.36 (d, J = 3Hz, 1H). 13C NMR (50 MHz,
CD3COCD3, ppm): d -4.86, -4.77, -4.71, -4.60, 17.81, 17.98,
25.62, 25.80, 28.7, 42.0, 50.0, 69.90, 70.75, 77.05, 118.7,
126.38, 127.32, 127.76, 128.44, 135.36, 136.67, 140.42,
148.88, 164.28. MS m/e (relative intensity): 632 (M+, 25), 575
(12), 422 (75), 167 (100), 73 (12).
E
Article Identifier:
1437-2096,E;1999,0,10,1559,1562,ftx,en;G16899ST.pdf
Synlett 1999, No. 10, 1559–1562 ISSN 0936-5214 © Thieme Stuttgart · New York