Unexpected reactivity of trifluoromethylated olefins with indole: A mechanistic investigation

Article abstract of DOI:10.1016/j.tetlet.2012.02.055

Several trifluoromethylated compounds were reacted with indole sodium salt, leading to monofluorinated compounds. The unexpected products formation was rationalized by DFT calculations.

Full text of DOI:10.1016/j.tetlet.2012.02.055

Tetrahedron Letters 53 (2012) 2132–2136
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Unexpected reactivity of trifluoromethylated olefins with indole:
a mechanistic investigation
Mónica S. Estevão a, Filipe J. S. Duarte a, Eduarda Fernandes b, A. Gil Santos a, M. Manuel B. Marques a,
⇑
a REQUIMTE – Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
b Departamento de Ciências Químicas, Faculdade de Farmácia, Universidade do Porto, Rua Aníbal Cunha 164, 4099-030 Porto, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
Several trifluoromethylated compounds were reacted with indole sodium salt, leading to monofluorinat-
ed compounds. The unexpected products formation was rationalized by DFT calculations.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 4 January 2012
Revised 8 February 2012
Accepted 13 February 2012
Available online 20 February 2012
Keywords:
Trifluoromethyl alkenes
C–F bond
Indole
DFT calculations
Fluoride elimination
Organic fluorinated compounds have received great attention in
all fields of science. Currently approximately 30% of all agrochem-
icals and 20% of pharmaceuticals contain fluorine. Incorporation of
fluorine has also been applied in material sciences such as the use-
ful polymer polytetrafluoroethylene (Teflon) that is perfluorinated
and exemplifies the strength of the C–F bond.1 Fluorine electroneg-
ativity, small size, and the significant electrostatic character of the
C–F bond are responsible for the unique properties of fluorinated
compounds.2 The introduction of fluorine atoms into organic com-
pounds imparts diverse properties, such as improved metabolic
stability,3 lipophilicity, or basicity,4 enhancing, in some cases, the
drug like properties of these molecules.5 Thus, the drug candidates
with one or more fluorine atoms have become a commonplace in
medicinal chemistry. Indeed, the number of fluorinated drugs in
the total number of drugs launched worldwide has been increasing
during the last 5 years.5
In an attempt to explore the effect of fluorine in the anti-inflam-
matory activity of some indole derivatives, we focused on the syn-
thesis of trifluoromethylated compounds.6 In the course of our
studies, we envisaged to introduce a 3,3,3-trifluorobut-2-en chain
at position N-1 of the indole scaffold.
In the presence of nucleophiles (preferentially metallic), trifluo-
romethylated alkenes without other reactive functional groups are
well known to easily react via SN20 mechanisms (Scheme 1A).2,6
However, we expected that compounds 1a,7 1b,8 and 1c9 would
react as electrophiles at the mesyl or tosyl group (Scheme 1B), based
on the previous reports on the SN2 reaction of these substrates.
To our surprise and in strong contrast with the literature, we
have to conclude that under our experimental conditions fluorine
is a better leaving group10 than mesyl or tosyl, while in the literature
it is shown that trifluoromethylated allyl bromide 1a has been used
in indium-mediated allylations to afford c-coupling products when
reacting with aldehydes11 (route c, the SN2 product was observed
only in a small amount),11a and methanesulfonate 1b and the tolu-
ene-4-sulfonic acid derivative 1c have been reported to afford exclu-
sively the SN2 product (route d) when reacted with nucleophiles.8,9
Our first attempts to introduce the trifluoromethyl allylic chain
at the indolic nitrogen atom consisted on the treatment of 1b with
2, in DMF at 0 °C (Scheme 2). Surprisingly, product 3 was isolated
as a mixture of Z/Z and Z/E isomers in a 38% yield, instead of the
expected SN2 product, together with a little amount of isomerized
5a (16%, exclusively the Z isomer). In order to understand the role
of the solvent and base in the reaction outcome, several reaction
conditions were tested (Table 1). First experiments were carried
in DMF and the base loading was investigated (entries 1 and 3).
The yield of 3 is dependent on the amount of NaH, with smaller
amounts leading to lower product yields. In all experiments unre-
acted indole (starting material) was recovered. The dilution of the
mixture had no influence on the reaction outcome (entry 2). How-
ever, the use of 2 equiv of the sodium salt of indole 2 (entry 4), sig-
nificantly increased the yield of 3. The nature of the metal revealed
to be crucial for the fluorine elimination, since the formation of 3
was suppressed when BuLi was used (entry 9).
⇑
Corresponding author. Tel.: +351 21 2948300; fax: +351 21 2948550.
E-mail address: mmbmarques@fct.unl.pt (M.M.B. Marques).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2012.02.055

M. S. Estevão et al. / Tetrahedron Letters 53 (2012) 2132–2136
2133
(A) Reacctionnss of trifluoropropene derivatives with nucleophhiles
R
Nu
F
a
b
R
a
b
F
Nu
CF3
R
Nu
CF3
R = H, Aryl, Ph, Br, F, CO2H, SiMe2Ph; Nu = Nucleophile
(B) Reactions of β-substituted trifluoropropene
Nu
F3C
d
1a
1b
1c
X = Br
X = OMs
X= OTs
E
X
c
Inverted addition
SN2
c
d
F3C
E
F3C
Nu
X = Br, OMs, OTs;
Nu = Nucleophile; E = Electrophile
Scheme 1. Possible reactions of the trifluoromethylated alkenes.
2D experiments. The doublets found in the 13C NMR strongly
suggest that only one fluorine atom is present. The 19F NMR (pro-
ton coupled) of 3 (Z/Z) shows a doublet signal at ꢀ93.0 ppm with
28.9 Hz, while for 3 (Z/E) a doublet of 5.9 Hz is observed at
ꢀ81.8 ppm, thus confirming the proposed configurations.12
As stated in the introduction, fluoride elimination from trifluoro
allylic chains is known, as a side product, under metal-mediated
reactions.13 Yamazaki group reported the formation of difluorinat-
ed products during the Michael addition of organocopper species
to 3-[(E)-4,4,4-trifluorobut-2-enoyl]oxazolidin-2-ones.13a The
same group observed the formation of a difluorinated diene as a
product in SN2 reactions of Grignard reagents toward CF3-contain-
ing allylic acetates, in the presence of CuCN and TMSCl.13
+
+
F3C
X
N
Na
1b X = OMs
1c X = OTs
2
DMF, 0°C
X
N
F
F3C
X
3
4
Z/Z, Z/E)
5a X = OMs, 16%
5b X = OTs, 51%
(
(1:0.6), X = OMs, 38%
(Z/Z: Z/E) (1:0.3), X = OTs, 17%
Additionally, the reaction of
a-substituted trifluoromethylated
alkenes to give functionalized 1,1-difluoroalkenes, in the presence
of a nucleophile in aprotic solvents has already been reported
(Scheme 1, route a).2,6
Scheme 2. Reaction of 1b/1c with the sodium salt of 2.
The change in solvent influenced the products formation (entries
1and 6), and best results were obtained when DMF was used as sol-
vent. Inverting the order of the reagents addition, addition of a sus-
pension of sodium salt 2 to a solution of 1b, enhanced the yield of
the isomerized 5a and almost suppressed the formation of 3 (entry
8), supporting that two competitive mechanisms take place. When
1c was used the corresponding products 4 were isolated (entry
10). In any case was observed the SN2 product. The importance of
the electron-withdrawing effect of the CF3 group on the reaction
outcome was confirmed, as when the CF3 group was replaced by a
methyl group, compound 1d,11 the SN2 product was obtained with
both NaH and BuLi (entry 11). The corresponding SN2 product was
also isolated when 1b reacted with cylohexylamine instead of 2.
In all experiments the Z/Z isomer 3 was the major product.
Structure 3 (Z/Z and Z/E) was assigned by 1H, 13C, 19F NMR, and
In contrast with the literature known substrates, compound 1
has a good leaving group at b-position (mesyl or tosyl group),
and as it is accepted that the C–F bond is strong and chemically sta-
ble, the formation of 3 from 1b under our experimental conditions
was an unexpected reaction outcome.
Thus, aiming at a proposal of a possible mechanism, we envis-
aged a full density functional theory (DFT) study,14 by calculating
all the intermediates (some of them observed by MS) and transi-
tion states (TSs) along the possible reaction pathways. In Scheme
3 is depicted a proposed mechanism on the basis of our calculated
data, which indicates that the reaction occurs in three steps.
Scheme 4 details the mechanism by presenting all the important
TSs and intermediates as well as their relative energies. In all TSs
it is very important to consider the explicit participation of a so-
dium ion. In the absence of this ion the TS energies become extre-

2134
Table 1
M. S. Estevão et al. / Tetrahedron Letters 53 (2012) 2132–2136
Reaction conditions
R1
OR2
base
R1
1
solvent
0°C
solvent
0°C
N
OR2
N
N
H
X
F
6
2
3-4
5
R2O
Entry
1
1 R1; R2
Solvent
DMF
Base (equiv)
NaH (1)
Products
Yielda (%)
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
1b CF3; Ms
3 (Z/Z)/(Z/E) (1:0.6)
5a
3 (Z/Z)/(Z/E) (1:0.6)
5a
3 (Z/Z)/(Z/E) (1:0.6)
5a
3 (Z/Z)/(Z/E) (1:0.5)
5a
3 (Z/Z)/(Z/E) (1:0.5)
5a
3 (Z/Z)/(Z/E) (1:0.9)
5a
3 (Z/Z)/(Z/E) (1:0.6)
5a
3 (Z/Z)/(Z/E) (1:0.5)
38
16
47
1
22
Trace
83
Trace
40
7
3
9
29
6
25
2b
3
DMF
DMF
DMF
MeCN
THF
NaH (1)
NaH (0.5)
NaH (2)
NaH (1)
NaH (1)
t-BuOK (1)
NaH (1)
4c
5d
6
7d
8e
THF
THF
5a
—
9f
10
1b CF3; Ms
1c CF3; Ts
THF
DMF
n-BuLi (1)
NaH (1)
19
17
51
34
72
4 (Z/Z)/(Z/E) (1:0.3)
5b
11
1d CH3; Ms
1b CF3; Ms
DMF
THF
NaH (1)
—
SN2 product
SN2 product
12g
a
The yields were calculated based on 1H NMR of the isolated mixture of 3 and 5, except entries 10, 11 and 12. 1 equiv of indole 6 has been used except in entry 5.
b
c
d
e
f
The reaction was diluted 5 times.
2 equiv of 2 were used.
Addition of crown-ether (1 mol %).
Inverted order of reagents addition.
No reaction.
g
Cyclohexylamine was used as the nucleophile.
O
S
O
mely high or the TS structures cannot even be obtained, which is in
agreement with the experimental data.
O
O
CF3
Na
S
O
F
O
- NaF
1b (
+
E
)
For the reaction of 1b with 2, we studied six different pathways
(see SI). However, Scheme 4 shows only the two pathways (TS-1
and TS-2) that lead to experimentally observed products (5a (Z)
and 3). All the other calculated TSs are more energetic than TS-1
and TS-2 in, at least, 20 kJ molꢀ1, which means that they are irrel-
evant for the discussion.
N
F
N
7
2
2
- NaF
The energy difference between the double-bond migration (TS-
1) and the indole addition with concerted elimination of fluorine
anion (TS-2) is minimal. Thus, these two pathways can co-exist
and one or the other can be dominant depending on the reaction
conditions (solvent, temperature, concentration), as was experi-
mentally observed. The Z isomer of 5a is preferentially formed
due to strong restrictions on the TS conformation, as the sodium
ion has to coordinate between the mesylic and the indolic groups.
GC-MS experiments were carried (see SI, Table 4) and an interme-
diate with m/z (300) consistent with intermediate 7 was observed.
While compound 5a (Z) is a dead-end, intermediate 7 can un-
dergo fluorine substitution by indole anion to afford intermediate
8. According to our calculations, only the E isomer of 8 can be
formed, due to the lack of important electrostatic interactions in
the TS with Z configuration. An alternative mechanism for the for-
mation of 8 was tested, via an addition–elimination process, in
which the indole anion adds to the intermediate 7 to form an an-
ionic intermediate that undergoes fluorine elimination to originate
8.
O S O
O
O
S
O
F
- Indole
N
O
N
N
F
3 (Z/Z ) ; (Z/E
)
8 (
E
)
Scheme 3. Proposed mechanism for the formation of 3.
eliminated from intermediate 8 via two diastereomeric concerted
TSs (TS-4), but high selectivity is expected, as TS-4 (Z/E) is ca.
11.5 kJ molꢀ1 less energetic than TS-4 (E/E). Thus, product 3a (Z/
E) is predicted to be preferentially formed, which does not fully
agree with the experiment. Indeed, while the double-bond at the
mesyl group side was experimentally obtained in Z configuration,
the second double-bond was obtained as a Z/E mixture, with
preference for the Z configuration. However, this result can be ex-
However, this possibility is not discussed in Scheme 4, as it orig-
inates considerably higher energetic TSs (see SI). Indole can be

M. S. Estevão et al. / Tetrahedron Letters 53 (2012) 2132–2136
2135
O
+
S
N
O
CF3
O
Na
O
1b (
E
)
2
O
S
F
F
N
H
O
Na
F
N
Na
CF3
O
O
S
O
TS-1 (
= 115.92 kJ mol
Z
)
TS-2
-1
-1
Δ
G
ΔG
= 116.93 kJ mol
2
-
- NaF
O
O
S
O
F
O
S
O
O
N
F
CF3
5a (
Z
)
7
= 1.66 kJ mol-1
ΔG
= -52.14 kJ mol-1
2
Δ
G
O
O S O
O
S
O
O
F
- NaF
N
N
F
N
Na
N
F
8 (
E
)
TS-3 (
E )
-1
= 65.53 kJ mol-1
Δ
Δ
G
G = -156.35 kJ mol
O
S
F
O
N
O
H
O
N
F
H
S
N
O
O
N
TS-4 (E/E
= 65.40 kJ mol
)
TS-4 (Z/E )
= 53.74 kJ mol
-1
-1
ΔG
ΔG
- Indole
6
NaF
3 (Z/E
O
) Δ
:
G
= -186.13 kJ mol-1
F
O
N
S O
O
S O
O
F
F
H
N
Na
6
-
N
- NaF
-1
3 (Z/Z
) Δ
:
G
= -190.72 kJ mol
TS-5
= 51.74 kJ mol-1
Δ
G
Scheme 4. Proposed mechanism for the formation of 3. B3LYP/6-311++G(2df,2p)//B3LYP/6-31G(d,p), DMF, T = 25 °C (similar conclusions at 0 °C), radii = uaks.
plained, as the indole can catalyze the conversion of 3a (Z/E) into
3a (Z/Z), via TS-5, thus allowing for a mixture of configurations,
as experimentally observed. The conclusion is that while structure
3a (Z/E) is formed under kinetic control, isomer 3a (Z/Z) is formed
under thermodynamic control, which explains the relative variable
amounts of both isomers, depending on the reaction conditions.
In summary, an unexpected reactivity was observed when
compounds 1b and 1c were treated with indole sodium salt. A

2136
M. S. Estevão et al. / Tetrahedron Letters 53 (2012) 2132–2136
6. (a) Bergeron, M.; Johnson, T.; Paquin, J.-F. Angew. Chem., Int. Ed. 2011, 50,
11112–11116. and references cited therein; b) Kühnel, M. F.; Lentz, D. Angew.
Chem., Int. Ed. 2010, 49, 2933–2936.
7. (a) Loh, T.-P.; Li, X.-R. Angew. Chem., Int. Ed. 1997, 36, 980–982; (b) Loh, T.-P.; Li,
X.-R. Eur. J. Org. Chem. 1999, 1893–1899.
8. Sankyo Company, Ube industries, Ltd.; U.S. Patent 6063782A1, 2000.
9. a) Kelly, S. M.; Skelton, G.; Jones, C.; Minter, V.; Tuffin, R. Mol. Cryst. Liq. Cryst.
Sci. Technol., Sect. A 2001, 364, 873–880; b) UCB S.A.; Patent; WO2007/31263A1,
2007.; c) UCB S.A.; Patent; WO2005/121082A1, 2005.
mechanism was proposed, based on DFT studies, which fully ratio-
nalizes the experimental data. Six possible reaction pathways were
considered, but only two lead to experimentally observable
products (3 and 5a). The proposed mechanism identifies the
importance of the metal ion and the reaction conditions on the fi-
nal reaction outcome.
10. For fluorine as a leaving group see: (a) Pigeon, X.; Bergeron, M.; Barabé, F.;
Dubé, P.; Frost, H. N.; Paquin, J.-F. Angew. Chem., Int. Ed. 2010, 49, 1123–1127;
(b) Paquin, J.-F. Synlett 2010, 289–293.
Acknowledgments
11. a) Loh, T.-P.; Li, X.-R. Tetrahedron Lett. 1997, 38, 869–872; b) Chen, Q.; Qiu, X.-
L.; Qing, F.-L. J. Org. Chem. 2006, 71, 3762–3767.
12. Dolbie, W. R. A guide to fluorine NMR to organic chemists; John Wiley Sons:
Hoboken, 2009.
The authors acknowledge the Fundação para a Ciência e Tecno-
logia for financial support (PTDC/QUI/65187/2006) and for the PhD
fellowship to M. E. (SFRH/BD/46234/2008).
13. a) Yamazaki, T.; Shinohara, N.; Kitazume, T.; Sato, S. J. Fluorine Chem. 1999, 97,
91–96; b) Yamazaki, T.; Umetani, H.; Kitazume, T. Tetrahedron Lett. 1997, 38,
6705–6708; c) Yamamoto, S.; Sugimoto, H.; Tamura, O.; Mori, T.; Matsuo, N.;
Ishibashi, H. Tetrahedron 2004, 60, 8919–8927; d) Yamazaki, T.; Umenati, H.;
Kitazume, T. Isr. J. Chem. 1999, 39, 193–205; e) Hanzawa, Y.; Ishizawa, S.;
Kobayashi, Y.; Taguchi, T. Chem. Pharm. Bull. 1990, 38, 1104–1106.
14. Gaussian 09, Revision. B.01., Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.;
Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.;
Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi,
R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma,
K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian, Inc., Wallingford CT, 2009.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2012.02.055.
References and notes
1. Amii, H.; Uneyama, K. Chem. Rev. 2009, 109, 2119–2183. and references cited
therein.
2. O’Hagan, D. Chem. Soc. Rev. 2008, 37, 308–319.
3. Clader, J. W. J. Med. Chem. 2004, 47, 1–19.
4. (a) Müller, K. C.; Faeh, F. Diedrich Science 2007, 317, 1881–1886; b) Böhm, H.-J.;
Banner, D.; Bendels, S.; Kansy, M.; Kuhn, B.; Müller, K.; Obst-Sander, U.; Stahl,
M. ChemBioChem 2004, 5, 637–643.
5. Hagmann, W. K. J. Med. Chem. 2008, 51, 4359–4369.