Synthesis of organofluorine compounds using ene type reaction of N-(p-toluenesulfonyl)trifluoroacetaldehyde imine

Article abstract of DOI:10.1016/S0022-1139(99)00027-5

N-Tosyltrifluoroacetaldehyde imine (4) reacted as an enophile, but it is very sensitive to moisture and the yields of ene products were low. N-(2,2,2-Trifluoro-1-ethoxyethyl)tosylamide (11), obtained by the reaction of trifluoroacetaldehyde ethyl hemiacetal (1) with tosylamide (3) in the presence of TiCl₄ followed by addition of ethanol, was found to react as a good substitute for 4 to give the same products from the ene reaction of 4 in much better yields.

Full text of DOI:10.1016/S0022-1139(99)00027-5

Journal of Fluorine Chemistry 97 (1999) 61±63
Short communication
Synthesis of organo¯uorine compounds using ene type reaction
of N-(p-toluenesulfonyl)tri¯uoroacetaldehyde imine
Itsumaro Kumadaki*, Shuichi Jonoshita, Atsuhiro Harada,
Masaaki Omote, Akira Ando
Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho, 573-0101 Hirakata, Japan
Received 20 October 1998; received in revised form 5 November 1998; accepted 7 November 1998
Dedicated to Prof. Yoshiro Kobayashi on the occasion of his 75th birthday
Abstract
N-Tosyltri¯uoroacetaldehyde imine (4) reacted as an enophile, but it is very sensitive to moisture and the yields of ene products were low.
N-(2,2,2-Tri¯uoro-1-ethoxyethyl)tosylamide (11), obtained by the reaction of tri¯uoroacetaldehyde ethyl hemiacetal (1) with tosylamide
(3) in the presence of TiCl4 followed by addition of ethanol, was found to react as a good substitute for 4 to give the same products from the
ene reaction of 4 in much better yields. # 1999 Elsevier Science S.A. All rights reserved.
Keywords: Tri¯uoroacetaldehyde; Imine; Ene reaction; Amidoacetal; Titanium tetrachloride
Nowadays, organo¯uorine compounds are attracting
much attention in biomedicinal ®elds, and some new med-
icines containing ¯uorine substituents have been developed
[1]. New methodologies for ¯uorine compounds have been
developed, too. However, new methods for syntheses of new
types of ¯uorine compounds are still required. We are
engaged in ®nding new synthones for synthesis of tri¯uoro-
methyl compounds, and reported that tri¯uoromethyl car-
bonyl compounds reacted as a good enophile in the presence
or absence of a Lewis acid and gave a wide range of ꢀ-
(tri¯uoromethyl)homoallyl alcohols [2].
without puri®cation.1
(1)
Reaction of 4 with terminal ene compounds, 1-decene (5)
and allylbenzene (7), proceeded in boiling xylene to give
E-N-[1-(tri¯uoromethyl)-3-undecenyl]tosylamide (6) and
In this paper, we would like to present the ene reaction
of N-tosyltri¯uoroacetaldehyde imine (4) and related
reactions.
To compare the ene reaction of tri¯uoromethyl carbonyl
compounds with that of imine with a tri¯uoromethyl group,
N-tosyltri¯uoroacetaldehyde imine (4) was synthesized.
Thus, tri¯uoroacetaldehyde (2), generated from tri¯uoro-
acetaldehyde ethyl hemiacetal (1), was treated with p-tolu-
enesulfonamide (3), and the intermediate was treated with
thionyl chloride to give the imine (4) (Eq. (1)). This imine
was very sensitive to moisture and used for ene reaction
Scheme 1.
1CF3CHO (2), generated from CF3CH(OH)OEt (1, 3 ml) and concen-
trated H2SO4 at 808C, was introduced to a solution of p-toluenesulfona-
mide (3, 2.00 g, 11.7 mmol) in anhydrous THF (10 ml) containing pyridine
(45 ꢁl, 5 mol%). After standing in a sealed tube at room temperature for
two days, the mixture was concentrated using a vacuum-line. The residue
was refluxed with thionyl chloride (2 ml) in C6H6 (5 ml) for 5 h, and the
mixture was concentrated using a vacuum line. The residue, N-
tosyltrifluoroacetaldehyde imine (4), was used for the ene reaction.
*Corresponding author. Tel.: +81-720-66-3140; fax: +81-720-50-7020;
e-mail: kumadaki@pharm.setsunan.ac.jp
0022-1139/99/$ ± see front matter # 1999 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 0 2 7 - 5

62
I. Kumadaki et al. / Journal of Fluorine Chemistry 97 (1999) 61±63
E-N-[4-phenyl-1-(tri¯uoromethyl)-3-butenyl]tosylamide
(8), respectively, while non-terminal ene compounds,
cyclohexene (9) and trans-2-octene (10), hardly reacted
(Scheme 1).2
The ene reaction of tri¯uoroacetaldehyde itself does not
proceed thermally [3]. This means that a tosylimino group is
a better substituent as an enophile than a carbonyl group,
while the two large substituents of 4, a tri¯uoromethyl and a
p-toluenesulfonyl groups, make non-terminal disubstituted
ole®ns unreactive. Thus, 4 was found to be useful as an
enophile in the reaction with terminal ole®ns. However,
synthesis of 4 needs multistep and it is highly sensitive to
moisture and dif®cult to do with. This could be the reason
why the yields of the reaction were not satisfactory.
If our previous ®nding that 1 reacted with ene compounds
in the presence of Lewis acids [4] could be applied here, the
above dif®culties would be removed (Scheme 2).
Namely, if 1 reacts with 3 in the presence of a Lewis acid
to give ꢀ-tosylamidoalcohol and this forms the imine 4 (a)
or the amidoalcohol reacts directly with an ene compound in
the presence of the Lewis acid (b), the same product from
the ene reaction will be obtained.
In practice, the best result was obtained when 1, 3 and two
equivalents of 5 were treated in the presence of about two
equivalents of titanium chloride in methylene chloride. By
this reaction the ene product 6 was obtained in 66% yield
with 19% of N-(2,2,2-tri¯uoro-1- ethoxyethyl)tosylamide
(11). This reaction is superior to the former ene reaction of
the imine (4), since this procedure needs not preliminary
formation of tri¯uoroacetaldehyde 2 and its imine 4.
The same reaction of allylbenzene (7) gave the ene
product (8) and N-(2,2,2-tri¯uoro- 1-ethoxyethyl)tosyla-
mide (11) in 34% and 25% yields, respectively.3 These
results showed that the same products from the ene reaction
Scheme 2.
of 4 were obtained from monosubstituted ole®ns by a much
more simple procedure. However, this procedure was not
effective for disubstituted ole®ns. These results are shown in
Scheme 3.
The above results suggest that disubstituted ole®ns are
much less reactive than monosubstituted ones. This may be
due to the larger steric hindrance of the reactive intermedi-
ate of this reaction than those in the reaction of tri¯uor-
omethyl carbonyl compounds.
The above method made the procedure for the ene
products much simpler, but the yields were not satisfactory,
and the reaction hardly proceeded with disubstituted ene
compounds. The byproduct of the above reaction, N-(2,2,2-
tri¯uoro-1-ethoxyethyl)tosylamide (11), seemed to be a
good intermediate for the ene products. We could obtain
11 in a high yield by reaction of 1 and 3 in the presence of
TiCl4, followed by treatment with ethanol.4
Treatment of 5 and 11 with TiCl4 gave 6 in 29% yield
after 20 h. Namely, the reaction proceeded very slowly. We
presumed that the reaction could be accelerated, if an
intermediate like the imine 4 was formed. Thus, 11 was
treated with sodium hydride then TiCl4 followed by addition
of 5 to the mixture. Remarkably, 6 was obtained in 86% by
this procedure.5 Investigation of the mixture by 19F-NMR
showed no signal of 4 but a new peak at 14.1 ppm from
benzotri¯uoride temporarily assigned to the intermediate 12
(Scheme 4).
2A typical procedure for the ene reaction: 1-Decene (5, 1.35 ml,
7.1 mmol) and xylene (10 ml) was added to 4 obtained as above, and the
mixture was refluxed for 16 h. The mixture was worked up the usual way,
and the product was purified by column chromatography (SiO2, hexane±
CH2Cl2, 7:3) to give N-[1-(trifluoromethyl)-3-undecenyl]tosylamide (6,
0.992 g, 34%). 6: Colorless crystals. Mp 46±478C. Mass spectrum (MS)
4Preparation of 11: In a stream of Ar, 1 (1.32 ml, 11 mmol) and TiCl4
(2.18 ml, 20 mmol) was added dropwise to a solution of 3 (1.71 g,
10 mmol) in CH2Cl2 (20 ml), and the mixture was stirred at room
temperature for 14 h, followed by treatment with EtOH (3.50 ml,
60 mmol), to give 11 (2.49 g, 84%) after separation by column
chromatography (SiO2, hexane±CH2Cl2±AcOEt, 7:2:1). 11: Colorless
‡
m/z: 391 (M ). HRMS Calcd C19H28F3NO2S: 391.179. Found: 391.179.
1
IR (KBr) cm : 3292 (N±H), 1338, 1180 (SO). 1H-NMR (CDCl3) ꢂ: 0.88
(3H, t, Jˆ6.7 Hz), 1.13 (10H, m), 1.91 (2H, dt, Jˆ6.7, 6.7 Hz), 2.27 (1H,
ddd, Jˆ7.3, 7.3, 14.7 Hz), 2.35 (1H, ddd, Jˆ5.2, 7.3, 14.7 Hz), 2.43 (3H,
s), 3.91 (1H, m, ddq on treatment with D2O (Jˆ5.2, 7.3, 7.3 Hz)), 5.04
(1H, d, Jˆ9.2 Hz, disappeared on treatment with D2O), 5.10 (1H, ddd,
Jˆ7.3, 7.3, 15.2 Hz), 5.47 (1H, dtt, Jˆ15.2, 6.7, 1.2 Hz), 7.30 (2H, d,
Jˆ8.2 Hz), 7.73 (2H, d, Jˆ8.2 Hz). 19F-NMR (CDCl3) ppm (from
C6H5CF3): 11.2 (3F, d, Jˆ7.3 Hz). All the spectral data of other products
support the assigned structures.
3The typical procedure of the reaction of 1, 3 and an ene compound in
the presence of TiCl4: In a stream of Ar, CF3CH(OH)OEt (1, 132 ꢁl,
1.1 mmol) and TiCl4 (218 ml, 2.0 mmol) was added to a solution of TsNH2
(3, 0.171 g, 1.0 mmol) in CH2Cl2 (2 ml) dropwise, then the mixture was
stirred for 4 h. Allylbenzene (7, 300 ꢁl, 2.3 mmol) was added to the
mixture, and the whole was stirred for 6 h in a sealed condition. The
mixture was worked up the usual way, and the product was separated by
column chromatography (SiO2, hexane±AcOEt±Et2O, 8:1:1) to give 8
(124 mg, 34%) and 11 (74 mg, 25%).
‡
crystals. MS m/z: 349 (M ). HRMS Calcd C16H22F3NO2S: 349.132.
1
Found: 349.132. IR (KBr) cm : 3292 (N±H), 1330, 1166 (SO2). 1H-NMR
(CDCl3) ꢂ: 0.88 (3H, t, Jˆ7.3 Hz), 1.26 (4H, m), 1.42 (2H, m), 2.42 (3H,
s), 2.52 (1H, m), 3.97 (1H, ddq, Jˆ2.8, 12.5, 8.0 Hz), 4.71 (1H, d,
Jˆ12.5 Hz), 5.16 (1H, ddd, Jˆ0.9, 1.2, 17.1 Hz), 5.27 (1H, dd, Jˆ1.2,
10.7 Hz), 5.62 (1H, dddq, Jˆ8.2, 10.7, 17.1, 0.9 Hz), 7.29 (2H, d,
Jˆ8.2 Hz), 7.74 (2H, d, Jˆ8.2 Hz). 19F-NMR (CDCl3) ppm: 8.78 (3F, d,
Jˆ8.0 Hz).
5In a stream of Ar, a solution of 11 (0.300 g, 1.0 mmol) in CH2Cl2
(0.50 ml) and TiCl4 (218 ml, 2.0 mmol) were added in this order to a
suspension of 60% NaH (50 mg) in CH2Cl2 (1.50 ml), and the mixture was
stirred at room temperature for 1.5 h, then 1-decene (5, 190 ꢁl, 1.0 mmol)
was added. After stirring for 4 h, the mixture was worked up as usual and
the product was separated by column chromatography (SiO2, hexane±
CH2Cl2±AcOEt, 7:2:1) to give 6 (0.339 g, 86%).

I. Kumadaki et al. / Journal of Fluorine Chemistry 97 (1999) 61±63
63
Scheme 3.
Scheme 4.
This method can be applied to other ene compounds.
Allylbenzene (7) gave 8 in 70% yield. Surprisingly, non-
terminal ole®ns gave products from the ene reaction. Thus,
cyclohexene (9) gave a mixture of ene products (13 and 14),
their isomerization product (15) and an HCl adduct of 15
(16) in 33%, 26% and 9% yields, respectively (Scheme 5).
The same reaction of cis-2-heptene (17) gave two ene
products (18 and 19) in 68% and 19% yields, respectively.
Methylenecyclohexane (20) gave the ene type product (21)
in 74% yield. ꢃ-Methylstyrene (22) did not give the ene type
product but an adduct of the cation from 12 and chloride ion.
These results are shown in Scheme 5.
These results suggest that the above reaction is not a
concerted ene reaction but a stepwise reaction with cation
intermediates. Namely, a cationic carbon from 12 attacks
the double bond of the ene compounds. In the last example,
a stable benzyl cation, shown in the bracket was formed and
reacted with chloride ion to give 23. Formation of other ene
type products in the above reaction seems to be explained by
similar cationic intermediates.
In conclusion, N-tosyltri¯uoroacetaldehyde imine (4)
reacts as an enophile to give ene reaction products, but it
is so sensitive to moisture that it must be synthesized just
before use and the yields of the ene products are unsatis-
factory. Non-terminal ole®ns do not react with 4 at all. N-
(2,2,2-Tri¯uoro-1-ethoxyethyl)tosylamide (11), which is
obtained by the reaction of the hemiacetal (1) and tosyla-
mide (3) in the presence of TiCl4 followed by addition of
ethanol, serves as a good substitute for 4 to obtain the ene
products. Treatment of 11 with sodium hydride and TiCl4,
followed by addition of ene compounds at room tempera-
ture, gives high yields of the ene reaction products even
from non-terminal ole®ns. This reaction would serve for
synthesis of new type of ¯uorine compounds.
Acknowledgements
Authors express sincere gratitude to Central Glass, Inc.
for generous gift of tri¯uoroacetaldehyde hemiacetal.
References
[1] R. Filler, Y. Kobayashi, L.M. Yagupolskii (Eds.), in: Organic
Compounds in Medicinal Chemistry and Biomedicinal Applications,
Elsevier, Amsterdam, 1993.
[2] I. Kumadaki, Reviews Heteroatom Chem. 9 (1993) 183 and
references therein.
[3] K. Ogawa, T. Nagai, M. Nonomura, T. Takagi, M. Koyama, A. Ando,
T. Miki, I. Kumadaki, Chem. Pharm. Bull. 39 (1991) 1707.
[4] K. Sakumo, N. Kuki, T. Kuno, T. Takagi, M. Koyama, A. Ando, I.
Kumadaki, J. Fluorine Chem. 93 (1999) 165.
Scheme 5.