Towards the syntheses of α-trifluoromethylated oxygenated heterocycles and their precursors

Article abstract of DOI:10.1055/s-2005-872076

The possibilities of synthesizing various α-trifluoro-methylated oxygenated heterocycles, starting from trifluoroacetaldehyde methyl hemiacetal, through classic cyclization reactions, have been demonstrated. In this way, six-membered cyclic compounds, δ-l

Full text of DOI:10.1055/s-2005-872076

FEATURE ARTICLE
2253
Towards the Syntheses of a-Trifluoromethylated Oxygenated Heterocycles
and Their Precursors
a
-Trifluorome
t
thylate
e
O
xygenat
v
e
d
H
eterocy
eles n Harthong, Thierry Billard,* Bernard R. Langlois
Laboratoire SERCOF (UMR CNRS 5181), Université Claude Bernard-Lyon 1, Bât. Chevreul, 43 Bd du 11 novembre 1918,
69622 Villeurbanne, France
Fax +33(4)72431323; E-mail: billard@univ-lyon1.fr
Received 21 June 2005
The majority of previously described synthetic methods
Abstract: The possibilities of synthesizing various a-trifluoro-
were directed towards the synthesis of trifluoromethyl
sugars. Such compounds could be obtained by nucleo-
philic trifluoromethylation of aldose derivatives with
methylated oxygenated heterocycles, starting from trifluoroacetalde-
hyde methyl hemiacetal, through classic cyclization reactions, have
been demonstrated. In this way, six-membered cyclic compounds,
d-lactones and also macrolactones, bearing a CF3 group in a posi- CF3SiMe3.4,5 The trifluoromethylation of glyceraldehydes
tion a to the oxygen could be easily obtained through a RCM reac-
tion. Bicyclic compounds could be also synthesized by an
intramolecular Pauson–Khand reaction.
fluorohexoses.6
Key words: fluorine, metathesis, cyclizations, oxygenated hetero-
with CF3I/Zn, followed by 5 steps of ring construction
have been also employed to prepare 6-deoxy-6,6,6-tri-
Hetero Diels–Alder strategy has been also envisaged to
cycles, trifluoroacetaldehyde
synthesize trifluoromethylated sugars. The fluorinated
starting materials can be either the dienophile (1,1,1-tri-
fluoro-4-alkoxybut-3-en-2-one)7 or the heterodiene (tri-
fluoroacetaldehyde).8 Longer strategies have been also
Introduction
elaborated to obtain, via multistep syntheses, trifluoro-
Oxygenated heterocycles constitute an important core in
various compounds implicated in different applications.
The most known molecules presenting such a structure
are, of course, the carbohydrates which play an important
role in the biological processes. On the other hand, since
Henri Moissan’s first isolation of elemental fluorine in
1886, the interest in fluorine chemistry has seen an expo-
nential growth because of the specific properties of fluor-
inated molecules, which opened the various field of
applications.1 Among these fluorinated compounds, tri-
fluoromethylated ones play an important role, in particu-
lar in the design of bioactive compounds. Indeed, the
introduction of such a moiety onto organic compounds
generally increases the lipophilicity of the molecules and,
consequently, contributes, often, to the amelioration of the
biological activity.2 Nevertheless, despite this growing in-
terest for the trifluoromethylated compounds, methods to
synthesize such products are still scarce and new method-
ologies are required.
methylated sugars. The fluorinated starting material was
trifluoromethylated allylic alcohols9 or trifluoromethylat-
ed butenolides arising from the corresponding trifluoro-
methylated furanols.10 Some other methods can be also
found in literature about the access to trifluoromethyltet-
rahydrofuran or -tetrahydropyran like structures by cyclo-
oxymercurations of unsaturated trifluoromethylated
alcohols11 or by Prins cyclization of homoallylic hemiac-
etals of trifluoroacetaldehyde.12 Finally, a-trifluorometh-
ylated g-butyrolactones and d-valerolactones could be
obtained either by palladium-mediated intramolecular
lactonization of allylic alcohols13 or by classical lacton-
ization of corresponding hydroxy acids coming from a
Kolbe electrolysis.14
We wish to present here the first attempts in the elabora-
tion of other efficient methods to achieve, quickly and ef-
ficiently, a-trifluoromethylated oxygenated heterocycles
or bicyclic molecules, starting from a commercially avail-
able fluorinated starting material not too expensive and
easy to handle. In the following work, we want essential-
ly, to demonstrate the feasibility and the potentialities of
these methods and underline certain occurring problems.
In search of synthesis of potentially new bioactive fluori-
nated compounds, we were interested in the synthesis of
trifluoromethylated oxygenated heterocycles, and in par-
ticular the a-trifluoromethylated ones. Some six-mem-
bered a-trifluoromethylated oxygenated heterocycles
have been previously described in the literature. Such de-
rivatives have found interesting applications in liquid
crystal compositions3 and in the design of derivatives of
doxorubicin in the antitumor field.4
Retrosynthetic Analysis
In previous work, we have described efficient methods for
the synthesis of a-trifluoromethylated nitrogen heterocy-
cles or bicycles starting from the trifluoromethylated ho-
moallylamine.15 Consequently, by analogy, the following
retrosynthetic strategies have been envisioned
(Scheme 1).
SYNTHESIS 2005, No. 13, pp 2253–2263
x
x.
x
x
.
2
0
0
5
Advanced online publication: 20.07.2005
DOI: 10.1055/s-2005-872076; Art ID: Z12005SS
© Georg Thieme Verlag Stuttgart · New York

2254
S. Harthong et al.
FEATURE ARTICLE
In these strategies, the key steps of cyclization are the available and not so easy to handle. Other techniques,
well-known reactions, namely ring-closing metathesis starting directly from the commercial methyl hemiacetal
(RCM)16 and Pauson–Khand reaction (PKR).17 The ad- form of fluoral, have been also developed. These methods
vantage of such strategies was their convergent aspect, are based on the reaction of allylorganometallic deriva-
since the common key intermediates were a-trifluoro- tives under Lewis acid activation conditions.19 We have
methylated homoallylic alcohols 2 which should be focused our interest on these latest methods as we wished
obtained from the stable commercial form of tri- to start from the fluoral methyl hemiacetal 1.
fluoroacetaldehyde, the methyl hemiacetal 1.
By comparison with our previous work on the synthesis of
a-trifluoromethylhomallylamine, via iminium spe-
cies,15a,20 we first tried to condense the allyltrimethylsi-
lane onto an oxonium species, which could arise from a
Synthesis of a-Trifluoromethylated Homoallylic Al-
cohols
Lewis activation of 1, to obtain the a-trifluoromethylho-
moallylic alcohol 2a (Table 1).
Some methods have been already described to obtain such
a-trifluoromethylated homoallylic alcohols. Most of them
consist in the nucleophilic addition of organometallic re-
Only ZnBr2 gave moderate yields (entry 3), the other
Lewis acids leaving the starting material unchanged.
Faced with these disappointing results, we turned our in-
agents (zinc, magnesium, lithium,18a boron,18b,c and
terest towards the interesting chemistry of indium species
stannyl18d derivatives) onto the gaseous trifluoroacetalde-
in water developed by Loh et al.19c (Scheme 2). Because
hyde (fluoral) which is very reactive, non-commercially
Biographical Sketches
Steven Harthong was born finishing the Master of Sci- Thierry Billard on the fluor-
in Lannion (France) in ence at the ENS-Lyon. Dur- inated oxygenated heterocy-
1981. After studies in ing his studies he spent four cles.
Rennes (France), he joined months in Dr. Langlois’ lab-
ENS-Lyon in 2002. He is oratory, working with Dr.
Thierry Billard was born in glois. After a post-doctoral particularly interested in
Dole (France) in 1971. After training with Prof. L. finding new methods for in-
studies at the ‘Institut de Ghosez
(Université troduction of fluorinated
Chimie et Physique Indust- Catholique de Louvain, Bel- moiety into molecules and
rielles de Lyon’, he received gium), he was appointed in their application in the syn-
his Ph.D. from the Universi- 1999 as a CNRS permanent thesis of fluorinated sub-
ty of Lyon in 1996 under the researcher in the laboratory strates with biological
supervision of Dr. B. Lan- of Dr. B. Langlois. He is interests.
Bernard R. Langlois was In this context, he was in- University of Lyon where
born near Paris (France) in volved in fluorinated aro- he is appointed as first class
1947. After receiving his di- matics and heterocycles, Research Director by the
ploma from the ‘Ecole trifluoromethoxy-contain-
CNRS. There, his research
Supérieure de Physique et ing products, trifluoroacetic group is developing new
Chimie Industrielles de Par- acid derivatives, triflic acid methodologies for the intro-
is’ in 1970, he entered the derivatives and radical tri- duction of fluorine-contain-
Rhône-Poulenc Co. where fluoromethylation. In the ing moieties in organic
he worked, in different re- meantime, he got a Ph.D. substrates. As far as possi-
search centres (near Paris degree from the University ble, these methodologies are
and Lyon), on organofluo- of Lyon in 1984 on the stud- illustrated by the synthesis
rine chemistry, from bench ies about difluorocarbene. of potentially bioactive
synthesis to development. In 1991, he moved to the compounds.
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

FEATURE ARTICLE
a-Trifluoromethylated Oxygenated Heterocycles
2255
Table 1 Reaction of 1 with Allyltrimethylsilane with Lewis Acid
Activation
R2
R1
R2
R1
R3
R4
RCM
Entry Lewis acid
Solvent
Time Yield
(%)a
F3C
O
(h)
F3C
O
R4
5
1
2
3
4
5
6
7
BF3·OEt2 (1.2 equiv)
CH2Cl2
Et2O
18
0
R3
3
MgBr2 (1 equiv)
ZnBr2 (1 equiv)
48
120
48
6
0
Et2O
32
0
R1
Yb(OTf)3 (0.1 equiv) THF–H2O (4:1)
Yb(OTf)3 (0.1 equiv) DMF
R2
OH
OMe
OH
0
F3C
F3C
1
2
Yb(OTf)3 (0.1 equiv) THF
6
0
Yb(OTf)3 (0.1 equiv) THF–H2O (9:1)
24
0
O
R2
a Crude yields determined by 19F NMR titration with internal standard
(PhOCF3).
R1
R1
PKR
R3
R2
F3C
O
F3C
O
amount of indium to only one equivalent without modify-
ing the yield (entries 1,3). Better yield could be also ob-
tained in a shorter time than in the previous work. The
same reaction can be also realized in an organic solvent
(DMF) without the presence of water in a satisfactory
yield (entry 2). If we succeeded to obtain quantitative
yield for the formation of 2, the isolated yield was gener-
ally lower (around 50%) because of its very high volatili-
ty. This method has been applied to the synthesis of other
a-trifluoromethylated homoallylic alcohols 2b–e
(Table 3).
6
R3
4
Scheme 1 Retrosynthetic analysis.
Br
In
(2 equiv)
OMe
OH
1 (1 equiv)
+
F3C
OH
F3C
H2O
(2 equiv)
2a (81%)
The yields are better when the reaction is carried out in
DMF rather than water. With our conditions, the com-
pound 2b could be obtained in good yield, whereas with
Scheme 2 Synthesis of 2a using In-chemistry.19c
of the use of an excess of the expensive indium, we tried their conditions, Loh et al. failed.19c Because of their vol-
to optimize this reaction (Table 2).
atility, 2d and 2e gave some problems of isolation and will
not be used in the following steps. Concerning the diaste-
reoselectivity, compounds 2b and 2c were obtained with
Attempts to replace indium by less expensive metal failed
(entries 4–8). However, we succeeded to decrease the
Table 2 Synthesis of 2a
Entry
Allyl bromide
Metal
Solvent
Time (h)
Yield
(%)a
1
2
3
4
5
1 equiv
1 equiv
1 equiv
1 equiv
1 equiv
In (1 equiv)
In (1 equiv)
In (1 equiv)
Zn (1 equiv)
H2O
DMF
H2O
H2O
H2O
2
5
5
6
6
84
69
96
0
Zn (1 equiv) +
In (0.1 equiv)
1
6
7
8
1 equiv
2 equiv
2 equiv
Mg (1 equiv)
THF
THF
THF
24
24
24
0
0
0
Mg (2 equiv)
Mg (2 equiv) + USb
a Crude yields determined by 19F NMR titration with internal standard (PhOCF3).
b US: Ultrasonic activation.
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

2256
S. Harthong et al.
FEATURE ARTICLE
Table 3 Synthesis of a-Trifluoromethylated Homoallylic Alcohols
2
8
Entry Allyl bromide
Solvent Product
Yield (%)a
F3C
O
F3C
O
O
3aa (43%)
3ag (60%)
1
Ph
Br
H2O
(10)
(de = 80%)
OH
7b
F3C
Ph
Ph
Ph
2b
2b
F3C
O
F3C
O
2
3
7b
DMF
H2O
90
(de = 80%)
Br
3ba (77%)
(de = 80%)
3bd (35%)
(de = 80%)
(13)
(de = 60%)
HO
Br
Ph
Ph
F3C
7c
7c
2c
2c
F3C
O
O
F3C
O
4
5
DMF
H2O
80
(de = 60%)
CO2Et
3bf (60%)
(de = 80%)
3be (95%)
(de = 80%)
0
Br
OH Br
Ph
Br
F3C
7d
7d
2d
2d
6
7
DMF
DMF
(25)
F3C
O
F3C
O
3cb (82%)
(de = 60%)
3ca (50%)
(de = 60%)
(50) 10
OH CO2Et
CO2Et
Br
F3C
7e
2e
a Isolated yields. In parentheses: crude yields determined by 19F NMR
titration with internal standard (PhOCF3).
3cc (41%)
(dr = 10:47:10:33)
F3C
O
Ph
modest to good diastereoselectivity. In the case of 2b, the
anti-conformation could be demonstrated. All these re-
sults are in accordance with literature.21
F3C
O
F3C
O
4a (91%)
(de = 80%)
4b (90%)
(de = 60%)
With these a-trifluoromethylhomoallylic alcohols in
hand, we turned our attention to synthesize the precursors
for cyclization by etherification or esterification of the OH
group with convenient unsaturated moieties.
Figure 1 Precursors synthesized for cyclization.
compound without need for further purification. A one-
pot two-step procedure from fluoral hemiacetal 1, has
been then elaborated. The key step is the use of a phase-
transfer catalyst to allylate 2a. At the end of the reaction,
only the expected product 3aa stays in the organic phase,
the unreacted allyl bromide being hydrolyzed into allylic
alcohol, consequently remains in the aqueous basic phase
with the residual 2a (Scheme 3).
Synthesis of Precursors for Cyclization
Classically, the etherification of 2 was realized by depro-
tonation of the OH group and condensation with alkenyl
halide. Because of the low nucleophilicity of the oxygen
atom, due to the presence of the high electron withdraw-
ing CF3 moiety, the esterification with acyl chlorides re-
quired also the preliminary deprotonation of the OH
group. The reaction was carried out in DMF using NaH as
base, and the precursor products obtained are illustrated in
Figure 1.
Ring-Closing Metathesis
2-(Trifluoromethyl)-3,6-dihydro-2H-pyran Deriva-
tives
The low yield observed for 3aa is essentially due to the
volatility of this compound (crude yield: 80%) which led
to a loss of product during evaporation of chromato-
graphed fractions. To circumvent this drawback, an alter-
native way has been envisaged to obtain directly the pure
The RCM reactions of the synthesized precursors were
then studied with the commercially available Grubbs I or
II catalysts (Scheme 4 and Table 4).
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

FEATURE ARTICLE
a-Trifluoromethylated Oxygenated Heterocycles
2257
The reactions give satisfactory yields for 5aa and 5ba (en-
tries 1,2) and more modest yields for 5be (entry 4). This
last result can be explained by the electronic deficiency of
the conjugated double bond that involves a decreased re-
activity towards metathesis. Moreover, the more active
Grubbs II catalyst was required in this reaction.
Br
OMe
(1 equiv)
F3C
OH
In (1 equiv)
H2O
F3C
OH
(1 equiv)
2a
Br
KOHaq (50%)
Et2O
No reaction was observed with the brominated diene 3bd.
Such result is in accordance with literature, which puts
forward an inhibition of the reaction by the brominated
double bond.22 Attempts to realize ring-opening/ring-
closing metathesis sequences with 3c failed since only
complex mixtures of fluorinated compounds were ob-
served (entries 8–10). These results led to suppose that all
the possibilities of ring closing occurred during the reac-
tion, accompanied certainly by cross-coupling reactions.
(1 equiv)
Bu4NCl (1 equiv)
F3C
O
3aa (70%)
From a practical point of view, the obtained compounds
are very difficult to purify, except 5ba. Indeed, 5aa was
too volatile to be extracted from the reacting mixture and
could not be characterized other than by 19F NMR spec-
trum (shift analogy with 5ba) and by molecular mass de-
termination by GC/MS. Compound 5be was obtained as
mixture with starting material 3be that could not be sepa-
rated, and again its formation has been proven by 19F
NMR spectrum and GC/MS.
Scheme 3 Alternative synthesis of 3aa.
R2
R4
R1
R3
R1
R3
CH2Cl2
Grubbs I or II
(10%)
F3C
O
F3C
O
3
5
These results confirm the possibility to generate easily
and rapidly a-trifluoromethylated oxygenated heterocy-
cles. However, the major drawback of this reaction comes
Scheme 4 RCM of 3.
Table 4 Synthesis of 5 by RCM of 3
Entry
1
Diene 3
Conditions
5
Yield (%)a
(70)
3aa
Grubbs I (c = 0.1 M), r.t., 24 h
F3C
O
O
5aa
2
3ba
Grubbs I (c = 0.1 M), r.t., 24 h
72
(de = 80%)
Ph
F3C
5ba
–
3
4
3bd
3be
Grubbs II (c = 0.01 M), 50 °C, 24 h
Grubbs II (c = 0.01 M), 50 °C, 24 h
0
(42)
(de = 60%)
Ph
CO2Et
F3C
O
5be
–
5
6
3cb
3cb
3cb
3cb
3cb
3ba
3cc
Grubbs I (c = 0.1 M), r.t., 24 h
0
Grubbs I (c = 0.1 M), 50 °C, 24 h
Grubbs I (c = 0.1 M), 110 °C (in toluene), 24 h
Grubbs II (c = 0.1 M), 50 °C, 24 h
Grubbs II (c = 0.01 M), 50 °C, 24 h
Grubbs II (c = 0.1 M), 50 °C, 24 h
Grubbs II (c = 0.01 M), 50 °C, 24 h
–
–
–
–
–
–
0
7
0
8
complex mixture
complex mixture
complex mixture
0
9
10
11
a Isolated yields. In parentheses: crude yields determined by 19F NMR titration with internal standard (PhOCF3).
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

2258
S. Harthong et al.
FEATURE ARTICLE
from the products, which are often difficult to purify be-
cause of their volatility. To extend the potentiality of
RCM to a-trifluoromethylated precursors, the enyne met-
athesis has been also tested, starting from 4a to achieve
the synthesis of conjugated diene (Scheme 5).
Ph
F3C
O
8a (40%)
Disfavored
Grubbs II (10%)
Ph
Ph
(c = 0.1 M)
Ph
Ph
Grubbs II
CH2Cl2
50 °C
Ru
O
F3C
O
F3C
O
4a
F3C
O
8a (40%)
F3C
4a
Scheme 5 RCM of the enyne 4a.
Ru
Ph
Grubbs II
The expected diene 8a was obtained only in moderate
yield. Such a result is not necessarily surprising compared
to literature, since generally, most of enyne metathesis re-
quired the addition of ethylene to reach good yields. How-
ever, in the same conditions, good yields were obtained
with the corresponding nitrogen compounds.15a This dif-
ference of reactivity could be rationalized by supposing a
chelation between oxygen and ruthenium in the reaction
intermediate arising from the attack of the Grubbs catalyst
onto the triple bond. This chelation would stabilize this
new carbene, disfavoring its reaction with double bond
and then prohibiting the cyclization (Scheme 6). In the
case of nitrogen compounds, the chelation is disfavored
because of the Cbz protecting group of nitrogen, which
delocalizes the free electron pair (Scheme 6). Despite its
modest yield of formation, the obtained conjugated diene
should constitute interesting starting material for the de-
sign of bicyclic structures, for example by Diels–Alder
cycloadditions.
F3C
N
Cbz
F3C
N
BnO
O
No chelation
Favored
F3C
N
Cbz
(75%)
Scheme 6 Enynes metathesis: oxygen vs nitrogen compounds.
Grubbs II
(10%)
(c = 0.1 M)
Ph
Ph
CH2Cl2
50 °C
F3C
O
O
F3C
O
O
9b (80%)
3bf
Synthesis of Lactones
The lactone core is present in many of the structures of
natural products and bioactive molecules. For instance,
macrolide antibiotics possess such substructures. In these
last years, RCM has proved to be an efficient method to
achieve such structures. Consequently, we looked into the
possibility to obtain a-trifluoromethyl lactones starting
from the esters 3ag and 3bf (Scheme 7).
Grubbs II
(10%)
(c = 0.001 M)
8
CF3
CH2Cl2
50 °C
F3C
O
O
O
3ag
O
9a (60%)
The expected a-trifluoromethyl d-lactone was obtained
with good yield confirming the power of the RCM in such
synthesis. More interestingly, for the first time, 14-mem-
bered a-trifluoromethyl macrolactone could be also syn-
thesized, opening interesting perspectives in the synthesis
of fluorinated analogues of macrolides. Such applications
are under study in our laboratory and will be published in
due course.
Scheme 7 Synthesis of a-trifluoromethylated lactones.
use of such reaction into the synthesis of trifluoromethyl-
ated nitrogen bicyclic enones,15b we have been interested
to verify if the same reaction is possible in oxygen series.
Compound 4a was used as the model substrate for this re-
action (Scheme 8).
The expected a-trifluoromethylated oxygenated bicyclic
enone 6a was obtained in good yield. This cyclization is
also highly diastereoselective since, starting from a pure
anti diastereomer of 4a, only one diastereomer has been
detected by NMR. The resulting diastereomer possesses
an anti-anti-anti configuration, which was determined by
NOESY experiment (Figure 2).
Pauson–Khand Cyclization
The Pauson–Khand reaction (PKR) is a well-known co-
balt-mediated cyclization to achieve the preparation of bi-
cyclic ketones. Because of our recent work concerning the
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

FEATURE ARTICLE
a-Trifluoromethylated Oxygenated Heterocycles
2259
O
O
H
H
F3C Co
Co2(CO)8
Ph
Ph
Ph
NMO
Me
O
X
H
CH2Cl2 / THF
F3C
O
F3C
O
F3C
O
Ph
T1
r.t.
4a
6a (67%)
(de > 95%)
O
H
H
Scheme 8 Pauson–Khand cyclization of 4a.
F3C Co
Ph
H
Me
O
F3C
O
Ph
T2
O
H
Ph
Figure 4 Transition states of PKR with 4a.
H
H
O
F3C
nOe
Figure 2 NOESY correlation of 6a.
Co2(CO)8
Co2(CO)6
Such diastereoselectivity can be rationalized by consider-
ing the potential transition states. The NMR analysis of 4a
(anti diastereomer) has shown a gauche coupling constant
between H-1 and H-2 (3JH1,H2 = 4.3 Hz) leading to envis-
age a preferential conformation pseudo-diaxial for 4a
(Figure 3). This major conformation is certainly due to the
high electrostatic repulsion between the two electron-rich
groups CF3 and Ph.23
F3C
O
F3C
O
4b
NMO X
or
DMSO
O
F3C
O
H
CF3
CF3
H
6b
H2
H1
gauche
interaction
Scheme 9 PKR with 4b.
O
H2
H
O
H1
Ph
H1
Ph
H
step fails. This led to conclude that the cyclic tension in
the final product is too high and thus disfavors its forma-
tion.
H1
H
Ph
electrostatic
repulsion
O
Ph H
O
F3C
CF3
H2
H2
Figure 3 Preferential conformation of 4a.
Conclusion
In this work, we have shown that methyl hemiketal of flu-
oral constitutes an interesting fluorinated commercially
available starting material, to synthesize oxygenated cy-
clic structure bearing a CF3 group in a position. The ring
closing metathesis allowed not only the formation of 6-
membered cycles, bearing various substituents, but also
the synthesis of lactones and macrolactones. With the
Pauson–Khand reaction, bicyclic structure could be
achieved with a high diastereoselectivity. Nevertheless,
the main drawback of these reactions is essentially the dif-
ficult purification of the products due to the high volatility
of certain compounds. These volatilities are linked to the
presence of the CF3 group, which is well-known to de-
crease the boiling point of molecules containing this enti-
ty.1
By this fact, the two potential transition states for the PKR
are T1 and T2 (Figure 4). Because of a higher steric inter-
action between vinylic hydrogen and phenyl group in T1
than between the vinylic hydrogen and CF3 in T2, the tran-
sition state T2 is favored, leading to the formation of the
anti-anti-anti diastereomer. Such results are in accor-
dance with similar observation previously described in lit-
erature.24
With this result in hand, we have envisioned the construc-
tion of a fused tricyclic structure, starting from 4b
(Scheme 9). Tricycle formations by PKR have been de-
scribed,25 but never with the fused 3 cycles.
Whatever the conditions used for this reaction, even with
the use of DMSO as described in literature,11a the expect-
ed tricyclic compound was never obtained. With the aim
to understand the reasons of this failure, the reaction has
been followed, step by step. If the complexation of the tri-
ple bond by cobalt occurs quantitatively, the cyclization
However, the easy synthesis of the precursors for cycliza-
tions and the efficiency of the RCM and PKR to obtain
rapidly and diastereoselectively fluorinated cyclic struc-
tures should constitute effective synthetic tools for the de-
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

2260
S. Harthong et al.
FEATURE ARTICLE
13C NMR: d = 133. 5, 131.5, 126.5 (q, J = 1.5 Hz), 123.9 (q, J = 1.6
Hz), 125.7 (q, J = 283.2 Hz), 125.6 (q, J = 283.4 Hz), 73.7 (q,
J = 29.6 Hz), 73.35 (q, J = 29.3 Hz), 36.6 (q, J = 1.3 Hz), 36.2 (q,
J = 1.3 Hz), 26.8 (q, J = 1.1 Hz), 22.9 (q, J = 1.3 Hz), 25.1, 25.0,
21.5, 21.0.
sign of new fluorinated molecules. The synthesis of
fluorinated analogues of natural products, using these
methods, is under development in our laboratory. Another
interest of these strategies is their potential extension to
other fluoroalkyl group. Indeed, others hemiketal of fluo-
roalkyl aldehydes are available (with CF3CF2 or CF2Cl
moieties for example) and they can be used in the same
manner to reach other a-fluoroalkylated oxygenated cy-
cles. Such extension is also under study in our laboratory.
19F NMR: d = –76.02 (d, 0.2 F, J = 6.9 Hz), –77.01 (d, 0.8 F, J = 6.9
Hz).
Anal. Calcd for C8H11F3O: C, 53.33; H, 6.15. Found: C, 53.57; H,
5.97.
4-Bromo-1,1,1-trifluoropent-4-en-2-ol (2d)
1H NMR: d = 5.75 (m, 1 H), 5.60 (m, 1 H), 4.11 (m, 1 H), 2.3 (m, 2
H).
Prior to use, THF was freshly distilled over Na/benzophenone,
CH2Cl2 was dried over molecular sieves. Anhydrous DMF was pur-
chased from Aldrich. Other reagents were used as received. NMR
19F NMR: d = –80.17 (d, J = 6.9 Hz).
spectra were recorded on Bruker Avance 300 spectrometer. 1H, 13
C
Anal. Calcd for C5H6BrF3O: C, 27.42; H, 2.76; Br, 36.49. Found: C,
27.51; H, 2.72; Br, 36.13.
and 19F NMR spectra were recorded in CDCl3 at 300, 75 and 282
MHz, respectively. Chemical shifts are given in ppm relative to
TMS (1H, 13C) or CFCl3 (19F) as internal references and coupling
constants are given in Hz. Flash chromatography was performed on
silica gel 60 M (0.04–0.063 mm). Melting points (uncorrected)
were determined in capillary tubes on a Büchi apparatus.
Ethyl 5,5,5-Trifluoro-4-hydroxy-2-methylenepentanoate (2e)
1H NMR: d = 6.34 (d, 1 H, J = 1.1 Hz), 5.80 (m, 1 H), 4.25 (q, 2 H,
J = 7.0 Hz), 4.12 (ddq, 1 H, J = 9.5, 2.8, 6.6 Hz), 3.85 (br s, 1 H),
2.77 (br dd, 1 H, J = 14.4, 2.8 Hz), 2.58 (br dd, 1 H, J = 14.4, 9.5
Hz), 1.32 (t, 3 H, J = 7.0 Hz).
a-Trifluoromethylated Homoallylic Alcohols 2; General Proce-
dure
13C NMR: d = 168.2, 135.9, 129.9, 125.38 (q, J = 282.1 Hz), 70.0
(q, J = 30.9 Hz), 61.9, 33.7 (q, J = 2.0 Hz), 14.4.
To a suspension of In powder (115 mg, 1 mmol) in H2O or DMF (1
mL) was added allylic halide 7 (1 equiv). After stirring for 10 min,
fluoral hemiacetal 1 (160 mg, 1.2 mmol) was added. The mixture
was then stirred at r.t. for 5 h. A 5% aq solution of KHSO4 (5 mL)
was then added and the mixture was extracted with Et2O (10 mL).
The organic phase was dried (MgSO4) and the solvent was evapo-
rated in vacuo. Generally, the expected alcohol 2 was obtained pure
and was engaged in the following reaction. If necessary, purifica-
tion by flash chromatography could be realized.
19F NMR: d = –80.17 (d, J = 6.9 Hz).
Anal. Calcd for C8H11F3O3: C, 45.29; H, 5.23. Found: C, 45.07; H,
5.51.
Ethers or Esters 3 and 4; General Procedure
To a solution of 2 (1 mmol) in DMF (2 mL) was added NaH (1.2
equiv) at 0 °C. After stirring for 10 min at 0 °C, allylic halide 7 (1
equiv) was added. The mixture was then stirred at r.t. for 5 h. Pen-
tane (10 mL) was then added and the organic phase was washed
with H2O (10 mL). The organic phase was dried (MgSO4) and the
solvent was evaporated in vacuo. The crude product was purified by
flash chromatography (eluent: pentane–Et2O).
1,1,1-Trifluoropent-4-en-2-ol (2a)
1H NMR: d = 5.85 (m, 1 H), 5.26 (m, 2 H), 4.02 (m, 1 H), 2.54 (m,
1 H), 2.41 (m, 1 H), 2.28 (d, J = 5.84 Hz, 1 H).
19F NMR: d = –79.91 (d, J = 6.9 Hz).
4-(Allyloxy)-5,5,5-trifluoropent-1-ene (3aa)
Anal. Calcd for C5H7F3O: C, 42.86; H, 5.04. Found: C, 43.01; H,
4.82.
1H NMR: d = 5.78–5.98 (br, 2 H), 5.13–5.38 (br, 4 H), 4.30 (dd, 1
H, J = 12.4, 5.5 Hz), 4.11 (dd, 1 H, J = 12.4, 6.3 Hz), 3.71 (tq, 1 H,
J = 6.4, 6.6 Hz), 2.44 (m, 2 H).
1,1,1-Trifluoro-3-phenylpent-4-en-2-ol (2b)
Two diastereomers, 90:10 ratio.
19F NMR: d = –77.18 (d, J = 6.6 Hz).
1H NMR: d = 7.36–7.41 (br, 5 H), 6.33 (ddd, 1 H, J = 8.5, 9.1, 17.3
Hz), 5.40 (dd, 1 H, J = 9.1, 1.0 Hz), 5.31 (dd, 1 H, J = 17.2, 1.0 Hz),
4.31 (dq, 1 H, J = 5.2, 6.9 Hz), 3.85 (dd, 1 H, J = 5.2, 8.5 Hz), 2.96
(br s, 1 H).
Anal. Calcd for C8H11F3O: C, 53.33; H, 6.15. Found: C, 53.47; H,
6.39.
1-(Trifluoromethyl)but-3-en-1-ylundec-10-enoate (3ag)
1H NMR: d = 5.64–5.89 (br, 2 H), 5.38 (ddq, 1 H, J = 9.2, 4.3, 6.1
Hz), 4.92–5.22 (br, 4 H), 2.35–2.65 (br, 4 H), 2.06 (m, 2 H), 1.65
(br m, 2 H), 1.27–1.43 (br, 10 H).
13C NMR: d = 140.4, 135.2, 129.3, 128.6, 127.8, 125.3 (q, J = 284.4
Hz), 73.6 (q, J = 29.5 Hz), 50.6 (q, J = 1.3 Hz).
19F NMR: d = –75.21 (d, 0.1 F, J = 6.9 Hz), –75.14 (d, 0.9 F, J = 6.9
Hz).
13C NMR: d = 172.0, 139.1, 130.9, 123.7 (q, J = 280.8 Hz), 119.4,
114.2, 68.68 (q, J = 31.9 Hz), 33.6, 33.8, 32.6 (q, J = 1.6 Hz), 29.3,
29.1, 29.0, 28.9, 28.9; 24.8.
Anal. Calcd for C11H11F3O: C, 61.11; H, 5.13. Found: C, 60.99; H,
5.04.
19F NMR: d = –77.37 (d, J = 6.1 Hz).
1-Cyclohex-2-en-1-yl-2,2,2-trifluoroethanol (2c)
Two diastereomers, 80:20 ratio.
Anal. Calcd for C16H25F3O2: C, 62.73; H, 8.23. Found: C, 62.81; H,
8.17.
1H NMR: d = 5.99 (ddd, 0.8 H, J = 10.2, 6.2, 3.8 Hz), 5.89 (ddd, 0.2
H, J = 10.1, 6.2, 3.8 Hz), 5.71 (br d, 0.8 H, J = 10.2 Hz), 5.52 (br d,
0.2 H, J = 10.1 Hz), 3.89 (ddq, 0.2 H, J = 5.7, 5.7, 6.9 Hz), 3.75
(ddq, 0.8 H, J = 7.8, 4.2, 6.9 Hz), 3.00 (d, 0.2 H, J = 5.7 Hz), 2.82
(d, 0.8 H, J = 7.8 Hz), 2.58 (m, 1 H), 2.02 (m, 2 H), 1.8 (m, 2 H),
1.59 (m, 2 H).
{1-[1-(Allyloxy)-2,2,2-trifluoroethyl]prop-2-en-1-yl}benzene
(3ba)
Two diastereomers, 90:10 ratio.
1H NMR: d = 7.27–7.36 (br, 5 H), 6.34 (ddd, 1 H, J = 8.8, 9.1, 17.7
Hz), 5.76 (ddt, 1 H, J = 5.6, 10.9, 16.7 Hz), 5.13–5.29 (br, 4 H),
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

FEATURE ARTICLE
a-Trifluoromethylated Oxygenated Heterocycles
2261
4.23 (dd, 1 H, J = 12.2, 5.6 Hz), 3.85–4.00 (br, 2 H), 3.73 (dd, 1 H,
J = 8.7, 4.0 Hz).
1H NMR: d = 6.70–6.01 (br, 3 H), 5.28 (m, 2 H), 4.32 (m, 1 H), 4.09
(m, 1 H), 3.62 (dq, 0.2 H, J = 5.3, 7.0 Hz), 3.47 (dq, 0.8 H, J = 6.6,
7.0 Hz), 2.56 (br m, 1 H), 2.02 (br m, 2 H), 1.45–1.90 (br, 4 H).
13C NMR: d = 141.2, 135.6, 133.8, 129.1, 128.6, 127.5, 125.4 (q,
J = 286.0 Hz), 118.6, 118.3 (C-1), 80.9 (q, J = 30.0 Hz), 74.8 (q,
J = 1.1 Hz), 50.35 (q, J = 1.5 Hz).
13C NMR: d = 134.3, 134.1, 130.5, 130.1, 127.2 (q, J = 1.3 Hz),
126.1 (q, J = 1.3 Hz), 126.0 (q, J = 286.0 Hz), 118.7, 118.5, 80.5 (q,
J = 27.8 Hz), 79.6 (q, J = 27.8 Hz), 74.6 (q, J = 1.1 Hz), 74.4 (q,
J = 1.1 Hz), 36.6 (q, J = 1.5 Hz), 26.3 (q, J = 1.6 Hz), 25.2, 25.2,
21.9.
19F NMR: d = –72.98 (d, 0.1 F, J = 5.7 Hz), –73.56 (d, 0.9 F, J = 6.9
Hz).
Anal. Calcd for C14H15F3O: C, 65.62; H, 5.90. Found: C, 65.89; H,
5.99.
19F NMR: d = –73.23 (d, 0.2 F, J = 7.0 Hz), –73.47 (d, 0.8 F, J = 7.0
Hz).
(1-{1-[(2-Bromoprop-2-en-1-yl)oxy]-2,2,2-trifluoroethyl}prop-
2-en-1-yl)benzene (3bd)
Two diastereomers, 90:10 ratio.
Anal. Calcd for C11H15F3O: C, 59.99; H, 6.87. Found: C, 60.21; H,
6.65.
[(1E)-3-(1-Cyclohex-2-en-1-yl-2,2,2-trifluoroethoxy)prop-1-en-
1-yl]benzene (3cb)
Two diastereomers, 80:20 ratio.
1H NMR: d = 7.26–7.41 (br, 5 H), 6.37 (m, 1 H), 5.93 (m, 1 H), 5.63
(m, 1 H), 5.14–5.32 (br, 2 H), 4.32 (d, 1 H, J = 13.4 Hz), 4.06 (d, 1
H, J = 13.4 Hz), 3.99 (dq, 1 H, J = 4.2, 6.7 Hz), 3.79 (dd, 1 H,
J = 8.7, 4.2 Hz).
1H NMR: d = 7.28–7.48 (br, 5 H), 6.69 (d, 1 H, J = 16.2 Hz), 6.25–
6.39 (br, 1 H), 5.89–5.97 (br, 1 H), 5.86 (br d, 0.8 H, J = 11.3 Hz),
5.61 (d, 0.2 H, J = 10.9 Hz), 4.47–4.59 (br, 1 H), 4.31 (dd, 1 H,
J = 12.2, 6.8 Hz), 3.73 (dq, 0.2 H, J = 5.4, 6.9 Hz), 3.59 (dq, 0.8 H,
J = 6.4, 6.9 Hz), 2.63 (m, 1 H), 2.08 (m, 2 H), 1.77–1.95 (br, 2 H),
1.52–1.70 (br, 2 H).
13C NMR: d = 140.8, 135.3, 129.2, 128.52, 127.62, 127.4, 125.1 (q,
J = 286.0 Hz), 118.9, 118.6, 81.8 (q, J = 28.0 Hz), 77.2 (q, J = 1.1
Hz), 50.2 (q, J = 1.6 Hz).
19F NMR: d = –72.60 (d, 0.19 F, J = 6.7 Hz), –73.24 (d, 0.9 F,
J = 6.7 Hz).
13C NMR: d = 136.9, 134.0, 133.9, 130.6, 130.2, 129.0, 128.4,
127.1, 127.0, 127.2 (q, J = 1.5 Hz), 126.1 (q, J = 1.3 Hz), 126.1 (q,
J = 285.8 Hz), 126.1 (q, J = 285.2 Hz), 125.5, 125.2, 80.4 (q,
J = 27.8 Hz), 79.7 (q, J = 27.8 Hz), 74.3 (q, J = 1.3 Hz), 74.0 (q,
J = 1.3 Hz), 36.6 (q, J = 1.5 Hz), 26.4 (q, J = 1.6 Hz), 23.6 (q,
J = 1.3 Hz), 25.3, 25.3, 21.9, 21.3.
Anal. Calcd for C14H14BrF3O: C, 50.17; H, 4.21; Br, 23.84. Found:
C, 50.39; H, 4.26; Br, 23.96.
Ethyl 2-({[2-Phenyl-1-(trifluoromethyl)but-3-en-1-
yl]oxy}methyl)acrylate (3be)
Two diastereomers, 90:10 ratio.
19F NMR: d = –72.99 (d, 0.2 F, J = 6.9 Hz), –73.47 (d, 0.8 F, J = 6.9
Hz).
1H NMR: d = 7.24–7.37 (br, 5 H), 6.34 (m, 0.9 H), 6.31 (m, 0.9 H),
6.22 (m, 0.1 H), 6.12 (m, 0.1 H), 5.91 (m, 0.9 H), 5.65 (m, 0.1 H),
5.10–5.24 (br, 2 H), 4.46 (d, 0.9 H, J = 13.2 Hz), 4.44 (d, 0.1 H,
J = 13.6 Hz), 4.13–4.26 (br, 3 H), 4.03 (dq, 1 H, J = 4.5, 6.8 Hz),
3.67–3.79 (br, 1 H), 1.31 (t, 2.7 H, J = 7.2 Hz), 1.30 (t, 0.3 H, J = 7.2
Hz).
Anal. Calcd for C17H19F3O: C, 68.90; H, 6.46. Found: C, 68.96; H,
6.58.
3-[1-(Cyclohex-2-en-1-yloxy)-2,2,2-trifluoroethyl]cyclohexene
(3cc)
Four diastereomers, 10:47:10:33 ratio.
13C NMR: d = 165.8, 140.9, 139.6, 136.9, 136.7, 136.8, 135.6,
129.1, 129.0, 128.9, 128.5, 127.5, 127.5, 127.1, 125.3 (q, J = 286.2
Hz), 118.3, 117.7, 81.9 (q, J = 27.8 Hz), 81.8 (q, J = 28.0 Hz), 71.8,
61.2, 61.2, 51.5 (q, J = 1.3 Hz), 50.3 (q, J = 1.6 Hz), 14.5.
1H NMR: d = 5.54–5.96 (br, 4 H), 4.08 (m, 1 H), 3.48–3.78 (br, 1
H), 2.56 (m, 1 H), 1.50–2.22 (br, 12 H).
19F NMR: d = –73.53 (d, 0.1 F, J = 6.9 Hz), –73.62 (d, 0.47 F,
J = 6.9 Hz), –73.71 (d, 0.1 F, J = 6.9 Hz), –73.90 (d, 0.33 F, J = 6.9
Hz).
19F NMR: d = –72.89 (d, 0.1 F, J = 6.8 Hz), –73.49 (d, 0.9 F, J = 6.8
Hz).
Anal. Calcd for C17H19F3O3: C, 62.19; H, 5.83. Found: C, 62.06; H,
5.56.
Anal. Calcd for C14H19F3O: C, 64.60; H, 7.36. Found: C, 64.80; H,
7.48.
2-Phenyl-1-(trifluoromethyl)but-3-en-1-yl Acrylate (3bf)
Two diastereomers, 90:10 ratio.
{1-[1-(But-2-yn-1-yloxy)-2,2,2-trifluoroethyl]prop-2-en-1-
yl}benzene (4a)
Two diastereomers, 90:10 ratio.
1H NMR: d = 7.22–7.41 (br, 5 H), 6.54 (dd, 1 H, J = 17.3, 1.2 Hz),
5.92–6.38 (br, 3 H), 5.67–5.85 (br, 1 H), 5.14–5.33 (br, 2 H), 3.84
(m, 1 H).
1H NMR: d = 7.25–7.40 (br, 5 H), 6.34 (ddd, 1 H, J = 17.2, 9.5, 8.7
Hz), 5.23 (br d, 1 H, J = 9.5 Hz), 5.15 (br d, 1 H, J = 17.2 Hz), 4.15–
4.36 (br, 3 H), 3.74 (dd, 1 H, J = 4.3, 8.7 Hz), 1.76 (t, 3 H, J = 2.3
Hz).
13C NMR: d = 164.6, 164.3, 138.8, 138.5, 135.6, 135.5, 133.4,
133.1, 129.3, 129.0, 128.7, 128.4, 127.9, 127.8, 127.4, 127.0, 123.9
(q, J = 282.5 Hz), 118.9, 118.8, 72.2 (q, J = 30.2 Hz), 71.6 (q,
J = 30.4 Hz), 50.36 (q, J = 1.3 Hz), 50.18 (q, J = 1.3 Hz).
13C NMR: d = 140.9, 135.8, 128.9, 128.8, 127.4, 125.4 (q, J = 286.0
Hz), 118.2, 84.5, 79.1 (q, J = 28.0 Hz), 74.0, 60.9 (q, J = 1.5 Hz),
50.1 (q, J = 1.6 Hz), 3.9.
19F NMR: d = –73.23 (d, 0.1 F, J = 7.0 Hz), –73.47 (d, 0.9 F, J = 7.0
Hz).
19F NMR: d = –72.70 (d, 0.1 F, J = 6.9 Hz), –73.42 (d, 0.9 F, J = 6.9
Hz).
Anal. Calcd for C14H13F3O2: C, 62.22; H, 4.85. Found: C, 62.26; H,
4.69.
Anal. Calcd for C15H15F3O: C, 67.16; H, 5.64. Found: C, 67.37; H,
5.47.
3-[1-(Allyloxy)-2,2,2-trifluoroethyl]cyclohexene (3ca)
Two diastereomers, 80:20 ratio.
3-[1-(But-2-yn-1-yloxy)-2,2,2-trifluoroethyl]cyclohexene (4b)
Two diastereomers, 80:20 ratio.
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

2262
S. Harthong et al.
FEATURE ARTICLE
1H NMR: d = 5.76–5.89 (br, 2 H), 4.35 (m, 2 H), 3.87 (dq, 0.2 H,
J = 5.9, 6.9 Hz), 3.73 (dq, 0.8 H, J = 5.7, 6.9 Hz), 2.55 (br m, 1 H),
2.03 (m, 2 H), 1.70–1.91 (br, 5 H), 1.50–1.66 (br, 2 H).
13C NMR: d = 172.8, 136.0, 124.1 (q, J = 281.6 Hz), 123.7, 69.3 (q,
J = 31.5 Hz), 34.3, 31.6 (q, J = 2.2 Hz), 31.6, 26.6, 25.9, 25.9, 24.4,
24.1.
13C NMR: d = 130.5, 130.5, 126.9 (q, J = 1.5 Hz), 125.96 (q, J = 1.5
Hz), 126.00 (q, J = 285.6 Hz), 125.98 (q, J = 285.6 Hz), 84.3, 84.2,
78.7 (q, J = 28.0 Hz), 78.0 (q, J = 8.0 Hz), 74.5, 74.4, 60.7 (q,
J = 1.5 Hz), 60.4 (q, J = 1.5 Hz), 36.4 (q, J = 1.3 Hz), 36.3 (q,
J = 1.3 Hz), 26.2 (q, J = 1.2 Hz), 23.5 (q, J = 1.2 Hz), 25.20, 25.18,
21.9, 21.1, 3.92, 3.89.
19F NMR: d = –77.37 (d, J = 6.1 Hz).
Anal. Calcd for C14H21F3O2: C, 60.42; H, 7.61. Found: C, 60.69; H,
7.38.
6-(Trifluoromethyl)-5-phenyl-5,6-dihydro-2H-pyran-2-one
(9b)
19F NMR: d = –72.81 (d, 0.2 F, J = 6.9 Hz), –73.20 (d, 0.8 F, J = 6.9
Hz).
1H NMR: d = 7.35–7.52 (br, 3 H), 7.23–7.34 (br, 2 H), 6.83 (ddd, 1
H, J = 10.0, 4.1 Hz), 6.21 (dd, 1 H, J = 10.0, 2.1 Hz), 4.89 (dq, 1 H,
J = 6.1, 6.8 Hz), 3.99 (ddd, 1 H, J = 6.1, 4.1, 2.1 Hz).
Anal. Calcd for C12H15F3O: C, 62.06; H, 6.51. Found: C, 62.17; H,
6.19.
13C NMR: d = 160.3, 146.8, 136.9, 129.9, 129.3, 128.4, 123.3 (q,
J = 282.7 Hz), 120.1, 79.9 (q, J = 31.5 Hz), 39.4 (q, J = 1.5 Hz).
Ring-Closing Metathesis Reaction; General Procedure
To a solution of 3 or 4 (1 equiv) in CH2Cl2 (see Table 4 for concen-
tration) was added Grubbs catalyst (0.1 equiv). The reaction mix-
ture was stirred for 24 h (see Table 4 for temperature). The mixture
was then evaporated in vacuo and purified by flash chromatography
(eluent: pentane–Et2O) (Table 4).
19F NMR: d = –76.56 (d, J = 6.8 Hz).
Anal. Calcd for C12H9F3O2: C, 59.51; H, 3.75. Found: C, 59.60; H,
3.78.
Pauson–Khand Reaction; 3-(Trifluoromethyl)-7-methyl-4-phe-
nyl-3,4,4a,5-tetrahydrocyclopenta[c]pyran-6(1H)-one (6a);
Typical Procedure
To a solution of 4a (268 mg, 1 mmol) in CH2Cl2 (6 mL) was added
Co2(CO)8 (342 mg, 1 mmol). The mixture was stirred at r.t. for 3 h.
Then THF (12 mL) and CH2Cl2 (6 mL) were added, followed by
NMO (1.17 g, 10 mmol). The mixture was stirred at r.t. for 3 h. It
was then filtered on a pad of Celite (eluent: Et2O). After evaporation
of the solvent in vacuo, purification by flash chromatography (elu-
ent: pentane–Et2O, 4:1) afforded a white solid; mp 124 °C.
2-(Trifluoromethyl)-3,6-dihydro-2H-pyran (5aa)
19F NMR: d = –79.7 (d, J = 6.9 Hz).
GC/MS (EI, 70 eV): m/z = 152 (M+·).
3-Phenyl-2-(trifluoromethyl)-3,6-dihydro-2H-pyran (5ba)
1H NMR: d = 7.23–7.40 (br, 5 H), 5.98 (ddt, 1 H, J = 10.3, 2.4, 2.4
Hz), 5.79 (ddt, 1 H, J = 10.3, 2.4, 2.4 Hz), 4.42 (m, 2 H), 4.04 (dq,
1 H, J = 6.8, 7.5 Hz), 3,71 (ddt, 1 H, J = 7.5, 2.4, 2.4, Hz, H-2).
1H NMR: d = 7.21–7.41 (br, 5 H), 4.99 (d, 1 H, J = 14.0 Hz), 4.46
(d, 1 H, J = 14.0 Hz), 4.25 (dq, 1 H, J = 10.2, 5.9 Hz), 3.16 (br m, 1
H), 2.75 (dd, 1 H, J = 10.2, 11.7 Hz), 2.33 (dd, 1 H, J = 19.0, 6.4
Hz), 2.02 (dd, 1 H, J = 19.0, 2.5 Hz), 1.78 (br s, 3 H).
13C NMR: d = 140.4. 129.1, 128.7, 127.8, 127.6, 125.4, 124.7 (q,
J = 282.5 Hz), 77.6 (q, J = 29.3), 65.5, 41.0 (q, J = 1.3 Hz).
19F NMR: d = –75.43 (d, J = 6.8 Hz).
Anal. Calcd for C12H11F3O: C, 63.16; H, 4.86. Found: C, 63.21; H,
4.79.
13C NMR: d = 207.0, 163.5, 137.9, 136.0, 129.4, 128.3, 128.1,
123.9 (q, J = 282.1 Hz), 79.1 (q, J = 29.1 Hz), 65.5, 51.1 (q, J = 1.1
Hz), 43.8, 39.4.
Ethyl 6-(Trifluoromethyl)-5-phenyl-5,6-dihydro-2H-pyran-3-
carboxylate (5be)
Two diastereomers, 80:20 ratio.
19F NMR: d = –74.11 (d, J = 5.9 Hz).
Anal. Calcd for C16H15F3O2: C, 64.86; H, 5.10. Found: C, 64.72; H,
5.34.
19F NMR: d = –74.39 (d, 0.2 F, J = 6.9 Hz), –75.69 (d, 0.8 F, J = 6.9
Hz).
GC/MS (EI, 70 eV): m/z = 300 (M+·), 255 (M+· – OEt).
References
(1) (a) Schofield, H. J. Fluorine Chem. 1999, 100, 7. (b) Groß,
U.; Rüdiger, S. In Organofluorine Compounds, Houben-
Weyl: Methods of Organic Chemistry, Vol. E10a; Baasner,
B.; Hagemann, H.; Tatlow, J. C., Eds.; Georg Thieme:
Stuttgart, 1999, 18–26. (c) Smart, B. E. J. Fluorine Chem.
2001, 109, 3. (d) Hiyama, T. In Organofluorine
Compounds: Chemistry and Properties; Springer-Verlag:
Berlin, 2000. (e) Dunitz, J. D. ChemBioChem 2004, 5, 614.
(f) Dolbier, W. R. Jr. J. Fluorine Chem. 2005, 126, 157.
(2) Biffinger, J. C.; Kim, H. W.; DiMagno, S. G. ChemBioChem
2004, 5, 622.
(3) Ito, K.; Namekawa, M.; Ishizuka, T.; Nayuki, S. Japanese
Patent JP 2000226382, 2000; Chem. Abstr. 2000, 133,
170332.
(4) Nakai, K.; Tagaki, Y.; Ogawa, S.; Tsuchiya, T. Carbohydr.
Res. 1999, 320, 8.
5-Isopropenyl-3-phenyl-2-(trifluoromethyl)-3,6-dihydro-2H-
pyran (8a)
Two diastereomers, 80:20 ratio.
1H NMR: d = 7.22–7.46 (br, 5 H), 5.68 (d, 0.2 H, J = 4.9 Hz), 5.57
(d, 0.8 H, J = 4.1 Hz) 5.35 (br s, 0.2 H), 5.25 (br s, 1 H), 5.21 (br s,
0.8 H), 4.80–5.00 (br, 1.2 H), 4.25–4.50 (br, 1.8 H), 4.05–4.18 (br,
1 H), 2.13 (t, 0.6 H, J = 1.3 Hz), 2.00 (t, 2.4 H, J = 1.5 Hz).
13C NMR: d = 146.3, 145.83; 140.6, 140.5; 129.6, 129.5; 129.19,
129.16; 128.88, 128.87; 127.8, 127.7; 125.2 (q, J = 284.3 Hz),
125.1 (q, J = 284.1 Hz); 117.4, 114.8; 79.9 (q, J = 27.8 Hz), 79.7 (q,
J = 27.6 Hz); 73.5, 68.2; 47.4 (m large), 46.9 (m large); 24.8, 24.1.
19F NMR: d = –74.14 (d, 0.2 F, J = 6.9 Hz), –74.50 (d, 0.8 F, J = 6.9
Hz).
Anal. Calcd for C15H15F3O: C, 67.16; H, 5.64. Found: C, 67.41; H,
5.83.
(5) (a) Takagi, Y.; Nakai, K.; Tsuchiya, T.; Takeuchi, T. J. Med.
Chem. 1996, 39, 1582. (b) Bansal, R. C.; Dean, B.;
Hakomori, S.-I.; Toyokuni, T. J. Chem. Soc., Chem.
Commun. 1991, 796. (c) Nakai, K.; Takagi, Y.; Tsuchiya, T.
Carbohydr. Res. 1999, 316, 47.
(11E)-14-(Trifluoromethyl)oxacyclotetradec-11-en-2-one (9a)
1H NMR: d = 5.55 (ddd, 1 H, J = 14.6, 7.5, 6.8 Hz), 5.36 (m, 1 H),
2.34–2.64 (br, 4 H), 2.04 (m, 2 H), 1.64 (br m, 2 H), 1.27–1.43 (br,
10 H).
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York

FEATURE ARTICLE
a-Trifluoromethylated Oxygenated Heterocycles
2263
(6) Hanzawa, Y.; Uda, J.; Kobayashi, Y.; Ishido, Y.; Taguchi,
T.; Shiro, M. Chem. Pharm. Bull. 1991, 39, 2459.
(7) (a) Hayman, C. M.; Larsen, D. S.; Brooker, S. Aust. J. Chem.
1998, 51, 545. (b) Hayman, C. M.; Hanton, L. R.; Larsen, D.
S.; Guthrie, J. M. Aust. J. Chem. 1999, 52, 921.
(8) Leveque, L.; Le Blanc, M.; Pastor, R. Tetrahedron Lett.
1997, 38, 6001.
(9) Yamazaki, T.; Mizutani, K.; Kitazume, T. J. Org. Chem.
1995, 60, 6046.
(10) (a) Yamazaki, T.; Mizutani, K.; Kitazume, T. Tetrahedron:
Asymmetry 1993, 4, 1059. (b) Yamazaki, T.; Mizutani, K.;
Kitazume, T. J. Org. Chem. 1993, 58, 4346.
(18) (a) Ishikawa, N.; Koh, M. G.; Kitazume, T.; Choi, S. K. J.
Fluorine Chem. 1984, 24, 419. (b) Kumar, D. J. S.;
Madhavan, S.; Ramachandran, P. V.; Brown, H. C.
Tetrahedron: Asymmetry 2000, 11, 4629.
(c) Ramachandran, P. V.; Padiya, K. J.; Rauniyar, V.;
Reddy, M. V. R.; Brown, H. C. Tetrahedron Lett. 2004, 45,
1015. (d) Asao, N.; Liu, P. i.; Maruoka, K. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 2507.
(19) (a) Kubota, T.; Iijima, M.; Tanaka, T. Tetrahedron Lett.
1992, 33, 1351. (b) Ishihara, T.; Hayashi, H.; Yamanaka, H.
Tetrahedron Lett. 1993, 34, 5777. (c) Loh, T.-P.; Li, X.-R.
Tetrahedron 1999, 55, 5611.
(11) (a) Arnone, A.; Bravo, P.; Frigerio, M.; Resnati, G.; Viani,
F. J. Chem. Res., Synop. 1989, 278. (b) Arnone, A.; Bravo,
P.; Frigerio, M.; Resnati, G.; Viani, F. J. Chem. Res.,
Miniprint 1989, 2201.
(12) Rychnovsky, S. D.; Hu, Y.; Ellsworth, B. Tetrahedron Lett.
1998, 39, 7271.
(13) Hanzawa, Y.; Ishizawa, S.; Ito, H.; Kobayashi, Y.; Taguchi,
T. J. Chem. Soc., Chem. Commun. 1990, 394.
(14) Seebach, D.; Renaud, P. Helv. Chim. Acta 1985, 68, 2342.
(15) (a) Gille, S.; Ferry, A.; Billard, T.; Langlois, B. R. J. Org.
Chem. 2003, 68, 8932. (b) Ferry, A.; Billard, T.; Langlois,
B. R. Synlett 2005, 1027.
(20) (a) Blond, G.; Billard, T.; Langlois, B. R. J. Org. Chem.
2001, 66, 4826. (b) Billard, T.; Langlois, B. R. J. Org.
Chem. 2002, 67, 997.
(21) Li, C.-J.; Chan, T.-H. Tetrahedron 1999, 55, 11149.
(22) (a) Katayma, H.; Nagao, M.; Ozawa, F. Organometallics
2003, 22, 586. (b) Chao, W.; Weinreb, S. M. Org. Lett.
2003, 5, 2505.
(23) Such electrostatic repulsion with a CF3 group leading to a
preferential conformation has been observed: Billard, T.;
Langlois, B. R. Tetrahedron 1999, 55, 8065.
(24) Mukai, C.; Kim, J. S.; Hanaoka, M. J. Org. Chem. 1999, 64,
6822.
(16) For reviews on RCM, see: (a) Fürstner, A. Angew. Chem.
Int. Ed. 2000, 39, 3012. (b) Alkene Metathesis in Organic
Synthesis; Fürstner, A., Ed.; Springer: Berlin, 1998.
(17) For recent reviews of the PKR, see: (a) Frühauf, H.-W.
Chem. Rev. 1997, 97, 523. (b) Geis, O.; Schmalz, H.-G.
Angew. Chem. Int. Ed. 1998, 37, 911. (c) Brummond, K.
M.; Kent, J. L. Tetrahedron 2000, 56, 3263. (d) Fletcher, A.
J.; Christie, S. D. R. J. Chem. Soc., Perkin Trans. 1 2000,
1657.
(25) (a) Jiang, B.; Xu, M. Angew. Chem. Int. Ed. 2004, 43, 2543.
(b) Ishizaki, M.; Iwahara, K.; Niimi, Y.; Satoh, H.; Hoshino,
O. Tetrahedron 2001, 57, 2729.
Synthesis 2005, No. 13, 2253–2263 © Thieme Stuttgart · New York