Diels-Alder reactions of perfluoroketene dithioacetals with electron-rich 1,3-dienes: A new access to polysubstituted aromatic sulfides

Article abstract of DOI:10.1016/j.tet.2012.07.054

The present paper describes the Diels-Alder reactions of perfluoroketene dithioacetals with electron-rich 1,3-dienes (2,3-dimethylbuta-1,3-diene, isoprene, penta-1,3-diene) followed by spontaneous HF and thiol elimination, leading to polysubstituted aromatic sulfides in moderate to good yields. Reactions seem to be dependent on the substitution patterns of perfluoroketene dithioacetals; the best results were obtained from trifluoromethyl or pentafluoroethyl and ethylsulfanyl derivatives. Theoretical calculations performed at the DFT level are in good agreement with the experimental results and show that the overall process is controlled by the cycloaddition step.

Full text of DOI:10.1016/j.tet.2012.07.054

Tetrahedron 68 (2012) 8663e8669
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
DielseAlder reactions of perfluoroketene dithioacetals with electron-rich
1,3-dienes: a new access to polysubstituted aromatic sulfides
,
a,
Jean-Philippe Bouillon a , Sergiy Mykhaylychenko b, Sigismund Melissen a, Agathe Martinez c,
*
Dominique Harakat c, Yuriy G. Shermolovich a
,
b
a
ꢀ
ꢀ
ꢀ
Universite et INSA de Rouen, Laboratoire Chimie Organique Bioorganique Reactivite et Analyses (COBRA), UMR CNRS 6014, IRCOF, F-76821 Mont Saint Aignan Cedex, France
b Institute of Organic Chemistry, National Academy of Sciences of Ukraine, 5, Murmanska, 02094 Kiev, Ukraine
c
ꢀ
ꢀ
Universite de Reims Champagne-Ardenne, Institut de Chimie Moleculaire de Reims, UMR CNRS 6229, UFR Sciences Exactes et Naturelles, BP 1039, F-51687 Reims Cedex 2, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 June 2012
Received in revised form 11 July 2012
Accepted 16 July 2012
Available online 21 July 2012
The present paper describes the DielseAlder reactions of perfluoroketene dithioacetals with electron-
rich 1,3-dienes (2,3-dimethylbuta-1,3-diene, isoprene, penta-1,3-diene) followed by spontaneous HF
and thiol elimination, leading to polysubstituted aromatic sulfides in moderate to good yields. Reactions
seem to be dependent on the substitution patterns of perfluoroketene dithioacetals; the best results
were obtained from trifluoromethyl or pentafluoroethyl and ethylsulfanyl derivatives. Theoretical cal-
culations performed at the DFT level are in good agreement with the experimental results and show that
the overall process is controlled by the cycloaddition step.
Dedicated to the memory of Professor H. G.
Viehe
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Perfluoroketene dithioacetals are unique versatile building-
blocks because of the presence of electron-accepting substituents
The DielseAlder reaction is one of the most powerful synthetic
tools in modern organic chemistry.1 Its classical version supposes
the use of electron-deficient olefins or acetylenes and electron-rich
1,3-dienes. In organofluorine chemistry, the use of electron-
deficient polyfluoroolefins in DielseAlder reactions affords new
(fluoro or perfluoroalkyl chains) in the molecule attached to one
carbon atom of the C]C bond and of electron-donating groups (S-
Alk or S-Ar) bonded to the other. The aim of our work was the
experimental and theoretical investigation of perfluoroketene
dithioacetals in [4þ2] cycloaddition reactions with electron-rich
1,3-dienes.
types of alicyclic fluorinated compounds. As
a rule, poly-
fluoroolefins contain electron-withdrawing substituents attached
at the carbon atoms of the C]C bond.2 Perfluoroketene dithioa-
cetals are one of the types of fluorine-containing olefins introduced
into synthetic practice and have been extensively used in the last 10
years in our laboratories.3 These compounds are convenient re-
agents4 for the synthesis of fluorine-containing cyclic and hetero-
cyclic derivatives; structures of increasing interest over the past
decade.5 The use of perfluoroketene dithioacetals in [4þ2] cyclo-
addition reactions with 1,3-dienes has not yet been studied, al-
though the pioneering papers of Viehe6a,b and Purrington6c have
shown the ability of such compounds to participate in [2þ2] and
[2þ3] cycloaddition reactions. It is worth noting that ketene
dithioacetals, especially 1,4-benzodithiafulvenes, are useful dien-
ophiles in the aza-DielseAlder reactions with N-arylimines leading
to tetrahydroquinoline derivatives.7
2. Results and discussion
2.1. Preliminary molecular orbital calculation study
In order to study the mutual influence of the substituents
(perfluoroalkyl and sulfanyl groups) on the nature of the C]C bond
and their possible participation in [4þ2] cycloaddition reactions
with 1,3-dienes, we performed molecular orbital calculations at the
DFT/B3LYP/6-31G** level of theory8 on several models of ketene
dithioacetals (Fig. 1).
The corresponding energy values of the LUMOs of the model
compounds are collected in Table 1. Evaluation of the frontier or-
bital energies confirmed the preferential interaction between the
HOMO of 2,3-dimethylbuta-1,3-diene and the LUMO of any of the
described perfluorinated dienophiles (Table 1: entries 4e9) and
thus the normal electron demand character of the cycloaddition
process. It should be noticed that our calculations also confirm the
known fact9 that the non-fluorinated dienophiles react with 1,3-
dienes according to an inverse electron demand process (Table 1:
* Corresponding author. Tel.: þ33 (0) 235522422; fax: þ33 (0) 235522959;
e-mail address: jean-philippe.bouillon@univ-rouen.fr (J.-P. Bouillon).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2012.07.054

8664
J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
R2
H
SR1
SR1
2.2. Reactions of perfluoroketene dithioacetals 1aef
RF
F
SR1
SR1
Perfluoroketene dithioacetals 1a,b did not react with 2,3-
dimethylbuta-1,3-diene at room temperature. When the reaction
mixture was heated for 40 h in a glass cylinder at 190 ꢀC, conversion
of ketene dithioacetals 1a,b did not exceed 40% and the corre-
sponding yields of compounds 3a,b were low. Indeed, the reaction
was accompanied by the formation of unidentified fluorinated
products most likely formed from perfluoroketene dithioacetals
decomposition.
The optimized reaction conditions used N-methylpyrrolidone
(NMP) as a polar solvent in a closed glass cylinder for w24 h at
190 ꢀC. Under these conditions, S-ethyl perfluoroketene dithioa-
cetals 1a,b reacted with an excess of 2,3-dimethylbuta-1,3-diene to
afford aromatic sulfides 3a,b in 31% and 63% yields, respectively
(Scheme 1, Table 2: entries 1 and 2). Obviously, the formation of
aromatic compounds 3a,b is the result of the loss of hydrogen
fluoride and ethylmercaptan from the DielseAlder cycloadducts
2a,b, probably due to the drastic thermolysis conditions
(Scheme 1).
1g: R1 = Me, R2 = H
1h: R1 = Me, R2 = Me
1i: R1 = Me, R2 = F
1a: RF = F, R1 = Et
1b: RF = CF3, R1 = Et
1c: RF = CF3, R1 = Ph
1d: RF = C2F5, R1 = Et
1e: RF = C2F5, R1 = Ph
1f: RF = (CF2)3H, R1 = Et
1j: RF = F, R1 = Me
1k: RF = CF3, R1 = Me
1l: RF = C2F5, R1 = Me
1m: RF = (CF2)3H, R1 = Me
1n: RF = (CF2)2CF3, R1 = Et
Fig. 1. Various ketene dithioacetals.
Table 1
Energy values of the LUMOs of various ketene dithioacetals
R2
SR1
SR1
RF
F
SR1
SR1
Table 2
DielseAlder reactions of S-ethyl perfluoroketene dithioacetals 1a,b,d,f,n with 2,3-
dimethylbuta-1,3-diene
H
Entry
Dienophile
RF
Reaction time
(h)
Conversion
(%)
Yielda (%)
1c,e,j-m
1g-i
Entry Compds R1 RF/R2
EHOMO (eV) ELUMO (eV) Normal/Inversea
1
2
3
4
5
1a
1b
1d
1f
F
CF3
C2F5
(CF2)3H
(CF2)2CF3
23
22
40
48
41
100
100
100
<5
3a (31%)
3b (63%)
3d (60%)
d
1
2
3
4
5
6
7
8
9
1g
1h
1i
Me
Me Me
H
ꢁ5.92
ꢁ5.66
ꢁ5.87
ꢁ9.50
ꢁ9.85
ꢁ7.25
ꢁ9.84
ꢁ6.17
ꢁ0.102
ꢁ0.132
ꢁ0.094
ꢁ0.257
ꢁ1.07
ꢁ1.13
ꢁ1.22
ꢁ1.28
ꢁ1.15
6.05/5.80
6.02/5.54
6.24/5.75
5.89/9.38
5.08/9.73
5.02/7.13
4.93/9.72
4.87/6.05
5.00/6.20
Me R2¼F
Me RF¼F
Me CF3
Ph CF3
Me C2F5
Ph C2F5
1n
<5
d
1j
a
Isolated yield after silica gel column chromatography.
1k
1c
1l
1e
1m
In a second experiment, the trifluoromethyl group was replaced
by a longer polyfluoroalkyl chain. Compound 1d (RF¼C2F5) reacted
with 2,3-dimethylbuta-1,3-diene affording the cycloadduct 3d in
60% yield, after heating at 190 ꢀC for 40 h (Scheme 1, Table 2: entry
3). In contrast, compounds 1f and 1n bearing a (CF2)3H and
(CF2)2CF3 polyfluoroalkyl chain, respectively, did not react with 2,3-
dimethylbutadiene during 41e48 h at 190 ꢀC in NMP (Scheme 1,
Table 2: entries 4 and 5). Thus, the increase of polyfluoroalkyl chain
length had a detrimental effect on the course of the reaction (see
discussion Section 2.3).
Me CF2CF2CF2H ꢁ6.33
a
HOMOdieneeLUMOdienophile gaps for normal and inverse control. The smallest
value gives rise to the type of control of the cycloaddition process.
entries 1 and 2). Moreover, when a single fluorine atom is present
(Table 1: entry 3), the reaction is still under inverse control.
The reactivity of the system may parallel the decreasing LUMO
energy of the fluorinated dienophiles. The addition of an electron-
withdrawing perfluoroalkyl group lowers the energies of both the
HOMO and the LUMO of dienophile (Table 1: entries 5e9), thus
decreasing the HOMOdieneeLUMOdienophile gap would increase the
rate of the cycloaddition step. So, at least from the theoretical point
of view, the DielseAlder reactions of perfluoroalkyl ketene
dithioacetals with 2,3-dimethylbuta-1,3-diene are quite possible.
We experimentally checked this possibility with perfluoroketene
dithioacetals 1aef.
We then studied the influence of SR1 substituents of per-
fluoroketene dithioacetals by replacing the ethylsulfanyl group by
a
phenylsulfanyl. The reaction of dienophile 1c with 2,3-
dimethylbuta-1,3-diene afforded only 29% yield (conversion:
100%) of cycloadduct 3c, accompanied by several other fluorinated
compounds (Scheme 2). Among them, a mixture (w60/40) of two
trifluoromethyl derivatives was isolated from the crude product,
RF
RF
RF
F
SEt
SEt
NMP, 190°C
22-48h
F
+
SEt
- HF, -EtSH
SEt
SEt
1a,b,d,f,n
3a,b,d
2a,b,d
1a: RF = F; 1b: RF = CF3; 1d: RF = C2F5; 1f: RF = (CF2)3H; 1n: RF = (CF2)2CF3
Scheme 1. DielseAlder reactions of S-ethyl perfluoroketene dithioacetals 1a,b,d,f,n with 2,3-dimethylbuta-1,3-diene.

J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
8665
8
9
2
RF
5
1
6
S
CF3
6
RF
F
SPh
SPh
3
NMP
4
7
+
7
9
190°C, 48h
SPh
1
4
CF3
3
10
S
8
10
5
2
1c,e
3c: 29%
3e:
-
5a (minor)
5b (major)
1c: RF = CF3; 1e: RF = C2F5
Fig. 3. Assignment of regioisomers 5a,b.
Scheme 2. DielseAlder reactions of S-phenyl perfluoroketene dithioacetals 1c,e with
2,3-dimethylbuta-1,3-diene.
Molecular orbital calculations are in good agreement with the
NMR results. While the largest HOMO coefficient is located at C-1 of
isoprene (0.21 vs 0.18) owing to the donating effect of the methyl
group, the largest LUMO coefficient in 1b is located at the carbon
atom bonded to the fluorine atom and the trifluoromethyl group
(0.41 vs 0.36). However, the difference between the LUMO co-
efficients is very small, explaining both the nature of the major
product and the low regioselectivity.
Dienophile 1b was then reacted with penta-1,3-diene under the
same conditions for 60 h (Scheme 4). The crude reaction mixture
was complex, containing mainly the mixture (90/10) of
regioisomers 6a,b, cycloadducts 7a,b and intermediates 8a,b
(Fig. 4). Increasing the reaction heating time did not change the
conversion of intermediates 7 and 8 into corresponding aromatic
derivatives 6 but led to an important decomposition of the reaction
medium. Silica gel column chromatography allowed us to isolate
a pure mixture of the two regioisomers 6a,b (first fraction) in
a moderate 40% yield. Unfortunately, the two regioisomers were
not separated, and their precise structural assignment was not
successful.
which probably corresponds to the cycloadduct 2c and the in-
termediate 4 resulting from HF elimination (Fig. 2).
CF3
CF3
F
SPh
SPh
SPh
SPh
4
2c
Fig. 2. Structure of possible intermediates 2c and 4.
Moreover, the reaction with the C2F5 derivative 1e failed; the
starting material being almost quantitatively recovered after
heating the reaction mixture for 48 h at 190 ꢀC (Scheme 2). This
observation also emphasizes the strong influence of the SR1 and
perfluoroalkyl substituents of the perfluoroketene dithioacetals on
the course of the cycloaddition reactions.
We then focused our attention to the reactions of un-
symmetrical dienes, such as isoprene and penta-1,3-diene. First,
the reaction of perfluoroketene dithioacetal 1b was performed
Assignment of the mixture of compounds 7a,b and 8a,b was
done by GCeMS study of the second fraction obtained from the
column chromatography (Fig. 4). Cycloadducts 7a,b appeared in
achiral GC as four chromatographic peaks at 19.43 min, 19.52 min,
19.71 min, and 21.34 min, respectively; their mass spectra were
almost the same, exhibiting molecular ions m/z¼302 [Mþ]. This
observation allowed us to assign cycloadducts 7a,b as a mixture of
under the usual conditions affording
a mixture (35/65) of
regioisomers 5a,b in 35% yield after silica gel column chromatog-
raphy (Scheme 3). Three other trifluoromethyl derivatives were
also observed in the 19F NMR spectra of the crude mixture; they
most likely correspond to the mixture of cycloadducts or in-
termediates coming from HF or EtSH elimination (see Experimental
section); nevertheless, more precise assignment was not possible.
two regioisomers, both being
a mixture of stereomers. In-
termediates 8a,b showed two peaks at 21.55 min and 24.12 min and
had, respectively, a molecular ion m/z¼282 [Mþ] clearly indicating
CF3
SEt
SEt
CF3
F3C
F
SEt
SEt
NMP, 190°C,
60h
+
+
1b
5b
5a
35% (35:65)
Scheme 3. DielseAlder reaction of perfluoroketene dithioacetal 1b with isoprene.
The 1H, 13C, COSY, HSQC, and HMBC spectra of the mixture of
compounds 5a and 5b led to unambiguously determine which
compound was the major regioisomer. Four doublets and two
broad singlets aromatic proton signals were separated into two
groups, one per compound, by means of the COSY spectrum. The C-
HF elimination; they were ascribed as
regioisomers.
a mixture of two
All attempts of reacting phenylsulfanyl perfluoroketene
dithioacetal 1c with those two unsymmetrical dienes failed. In all
cases, conversions did not exceed 20% and no trace of the aromatic
sulfide or cycloadduct was detected in the crude mixture by 19F
NMR and GCeMS. It is also worth noting that all reactions of
dienophile 1b with cyclic 1,3-dienes, such as cyclopentadiene,
cyclohexa-1,3-diene or silyloxy-1,3-dienes failed (conversions
<10e15% after heating at 190 ꢀC for 40e60 h).
These tandem cycloadditioneelimination reactions allow us to
obtain various aryl sulfides containing perfluoroalkyl groups in the
ortho-position to sulfur-containing substituents. It is worth noting
that perfluoroalkyl aryl sulfides are important intermediates in the
3
5 signals were easily identified by means of their JC,F couplings
(quartet, J¼5.6 Hz). One of these C-5 signals correlated with a 1H
doublet in the HSQC spectrum, thus indicating that it arose from
compound 5a, in which H-5 is coupled with H-4 (3JH,H¼8.1 Hz).
Conversely, the other H-5 signal is a broad singlet from 5b, in which
H-5 is surrounded by a methyl and a trifluoromethyl group. The full
1H and 13C assignment of both isomers was achieved through the
thorough analysis of the COSY, HSQC, and HSQC spectra.10 Signal
intensities indicated that the 5a/5b ratio was 35/65 (Fig. 3).

8666
J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
SEt
CF3
CF3
F3C
F
SEt
NMP, 190°C,
+
+
SEt
60h
SEt
1b
6a,b
40% (90:10)
Scheme 4. DielseAlder reaction of perfluoroketene dithioacetal 1b with penta-1,3-diene.
CF3
F
SEt
SEt
SEt
CF3
SEt
CF3
SEt
SEt
CF3
SEt
F
SEt
8a,b
7a,b
Fig. 4. Structure of possible cycloadducts 7a,b and intermediates 8a,b.
preparation of biologically active compounds, dyes, and chemical
reagents.11e13
Table 3
Activation energies
of the DielseAlder adducts
model compounds
D
Gact of the DielseAlder reactions, free energy of the formation
DGreact and of the aromatization process
DGarom for
The experimental data obtained lead to the following pre-
liminary conclusions: the ability of fluorinated ketene dithioacetals
1 to participate in DielseAlder reactions decreases with the length
of the perfluoroalkyl chain and depends on the nature of the sub-
stituent (SR1) on the sulfur atom. The aromatic substituent attached
to the sulfur atoms also dramatically restricts the reaction.
Entry Compds R1 RF/R2
D
Gact
DGreact
DGarom
d1 (A) d2 (A)
ꢁ
ꢁ
(kcal/mol) (kcal/mol) (kcal/mol)
1
2
3
4
5
6
7
1g
1j
1k
1c
1l
Me
Me
Me CF3
Ph CF3
Me C2F5
Ph C2F5
H
F
39.50
37.20
38.30
38.20
40.20
41.85
ꢁ16.50
ꢁ25.25
ꢁ17.00
ꢁ14.90
ꢁ14.55
ꢁ11.37
ꢁ9.20
ꢁ6.00
ꢁ23.50
ꢁ36.75
ꢁ41.40
ꢁ38.40
ꢁ53.40
ꢁ43.50
2.024 2.583
1.907 2.958
2.047 2.613
1.922 2.837
2.092 2.561
1.941 2.779
2.075 2.587
2.3. Discussion of quantum-chemical calculations
1e
1m
Me (CF2)3H 43.50
The quantum-chemical calculations of perfluoroketene
dithioacetals 1c and 1e showed that any extra conjugation raised
the energy of the HOMO and lowered the energy of the LUMO
(Table 1: entries 6 and 8). This fact probably leads to a decreased
reaction rate in the case of S-phenyl ketene dithioacetals. Moreover,
the decreased reaction rate in the case of olefin 1f with a long
perfluoroalkyl chain could be explained by steric reasons. As a net
result, the electronic effect could be counter balanced by a steric
one depending on the size and electron-withdrawing power of the
(CF2)3H substituent. For this last reason, the HOMO-
dieneeLUMOdienophile gap that varies in a small extent does not re-
flect the experimental results (Table 1: entry 9). The only way to
take into account this steric effect is to evaluate the global activa-
tion energy of both the cycloaddition and aromatization processes
and also to locate the transition states.
All of the transition states (TS) could be located and their cor-
responding activation energies were computed. A frequency cal-
culation was performed in order to verify that minima (reactants
and adducts) and TS had zero and one imaginary frequency, re-
spectively. The asynchronicity of the DielseAlder process is high-
lighted by the large difference between the length of the forming
bonds at the TS (Table 3), the shorter one corresponding, as ex-
pected, to the bond between the fluorinated carbon atom and that
of the symmetrical diene (d1, in Fig. 8). Moreover, in these TS, one
carbon atom clearly becomes sp3 while the second is close to the
sp2 hybrid state (Figs. 5e7).
Table 3 gives the free energy variations
DGreact and DGarom for
the cycloaddition process and the aromatization reactions of the
DielseAlder adducts. These calculations are in particularly good
agreement with the experimental results in the case of the
CF2CF2CF2H substituent (Table 3: entry 7). The highest value of
activation energy (DGarom) and the lowest absolute value of the
formation energy (DGreact) of the adduct confirmed the lack of
Fig. 5. Transition state corresponding to compound 1c.

J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
8667
Fig. 8. Schematic representation of the energy profiles for the cycloaddition reaction
followed by the aromatization process.
Fig. 6. Transition state corresponding to model compound 1m.
well with fluoro, trifluoromethyl or pentafluoroethyl ethylsulfanyl
ketene dithioacetals 1a,b,d with 2,3-dimethylbuta-1,3-diene,
whereas those with unsymmetrical dienes or phenylsulfanyl ke-
tene dithioacetal 1c were less efficient. Theoretical calculations
performed at the DFT level are in good agreement with the ex-
perimental results and show that the overall process seems to be
controlled by the cycloaddition step.
4. Experimental part
4.1. General methods
Reagents and solvents were used without further purification.
Perfluoroketene dithioacetals 1a,14,15 1b,15 1c, 1d,15,16 1e,17 1f,18 and
1n were prepared according to reported procedures. Reactions
were monitored by 19F NMR spectra of the crude mixture. Purifi-
cation of aromatic compounds 3, 5, and 6 were carried out using
silica gel flash column chromatography (70e200 mm). Their purity
was established by NMR spectroscopic data, and MS analysis. MS:
Thermo Finnigan, LCQ Advantage Max, Electrospray Ionization,
Source heater T¼220 ꢀC, capillary voltage¼33 V. High Resolution
Mass Spectra (HRMS) were recorded with a Q-TOF Micromass In-
strument in the positive ESI (CV¼30 V) mode. GC analyses were
performed on HewlettePackard 5890 Series II, HewlettePackard
6890 chromatograph equipped with a split/splitless injectors and
a flame ionization detectors operating both at 250 ꢀC. Helium was
used as carrier gas (flow: 1 mL minꢁ1). The split flow was adjusted
Fig. 7. Transition state corresponding to model compound 1l.
reactivity of this dienophile probably due to a steric effect. With
a difluoro group (Table 2, entry 1), the 31% yield obtained is also in
agreement with the lowest
DGact and DGreact values. Moreover, the
to 30 mL minꢁ1. GC column (HP5-ms, 27 mꢂ0.52 mm; 5
mm, Agi-
D
Garom has the lowest absolute value showing that aromatization is
lent) was used (program: 50 ꢀC (0 min)e250 ꢀC (15 min) at
not the determining step. The replacement of the SMe group by an
SPh one leads to a small increase in the activation energy and a less
favorable energy profile in nice agreement with the observed yields
for compounds 3b and 3c. In these cases, the favorable computed
Garom values confirm that the aromatization process is not able to
drive the reaction to completion and the overall process seems to
be controlled by the cycloaddition step.
25 ꢀC.minꢁ1). Data collection was performed on HP Chemstation
A.03.03. 0.2
m
L of samples diluted in CH2Cl2 was injected. 1H
(300 MHz), 19F (280 MHz), and 13C (75 MHz) NMR spectra were
recorded on a Bruker Advance DMX 300 instrument. All the ex-
periments were recorded using CDCl3 as solvent. TMS or the deu-
D
terated solvent signal was taken as internal reference for 1H and 13
C
spectra and the CFCl3 signal for 19F spectra. DFT calculations were
carried out with the Gaussian-03 suite of programs8 implemented
on a P575 IBM cluster. We used the three parameter hybrid func-
tions of Becke19 and the correlation function of Lee et al. (B3LYP)20
with the 6-31G** basis set for any element. Formation enthalpies
and Gibbs energies were calculated in vacuo after a frequency
calculation in order to be sure that the geometry found corre-
sponded to a stationary point and to achieve the zero point
3. Conclusion
In summary, we have described a new synthetic application of
perfluoroketene dithioacetals 1 in DielseAlder reactions with
electron-rich 1,3-dienes leading to various substituted aromatic
sulfides 3, 5, and 6 in moderate to good yields. The reactions work

8668
J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
correction. Transition states were fully characterized using the
keywords Opt¼(TS, CalcFC, Modredundant, NoEigenTest) and
a frequency calculation was performed to verify that the Hessian
has a single imaginary eigenvalue. In some cases, a complete IRC
calculation was performed.
284 [Mþ], 269, 236, 187. HRMS (ESIþ): calcd for C12H13F5S m/z
284.0663, found 284.0658.
4.2.4. Reaction of ketene dithioacetal 1c with 2,3-dimethylbuta-1,3-
diene (Scheme 2). (2-Trifluoromethyl-4,5-dimethylphenyl)phenyl-
sulfide (3c). This compound was purified by silica gel column
chromatography eluting with petroleum ether. Yield: 29%. Yellow
4.2. Typical procedure for DielseAlder reaction of per-
fluoroketen dithioacetal 1b with 2,3-dimethylbuta-1,3-diene
(Scheme 1, Table 2: entry 2)
oil. 1H NMR (CDCl3,
d ppm): 2.11 (s, 3H, CH3), 2.21 (s, 3H, CH3), 7.05
(s, 1H, CH, Ar), 7.15e7.25 (m, 5H, SPh), 7.39 (s, 1H, CH, Ar). 19F NMR
(CDCl3,
d
ppm): ꢁ60.5 (s). 13C NMR (CDCl3,
d ppm): 19.4 (s, CH3),
A mixture of perfluoroketene dithioacetal 1b (2.39 g, 0.01 mol)
and 2,3-dimethylbuta-1,3-diene (3.28 g, 0.04 mol) in N-methyl-
pyrrolidone (10 mL) was heated in a closed glass cylinder at 190 ꢀC
for 22 h. Reaction progress was monitored by 19F NMR spectra of
the crude mixture. After cooling to room temperature, the re-
action mixture was poured into water (150 mL) and extracted
three times with dichloromethane (3ꢂ30 mL). The combined or-
ganic phases were washed with water (50 mL), dried over mag-
nesium sulfate, and the solvent was evaporated under reduced
pressure. The residue was purified by silica gel column chroma-
tography (eluent: mixture (100/1) of petroleum ether and
dichloromethane) affording 1.47 g (yield: 63%) of aromatic
sulfide 3b.
19.5 (s, CH3), 123.8 (q, 1JC,F¼273.3 Hz, CF3), 127.1 (s, CH, Ar), 127.9 (q,
3JC,F¼5.5 Hz, CHeCCF3), 128.4 (q, 2JC,F¼27.3 Hz, CeCF3), 129.2 (s, 2ꢂ
4
CH, Ph), 131.0 (s, 2ꢂ CH, Ph), 131.2 (q, JC,F¼1.1 Hz, Cq, CeCH3),
135.76 (s, CH, Ph), 135.83 (s, Cq, CeCH3), 136.3 (s, Cq, Ph), 141.4 (q,
3JC,F¼1.1 Hz, Cq, CeSEt). MS (EI): m/z¼283 [Mþ1], 282 [Mþ], 247,
198. HRMS (ESIþ): calcd for C15H14F3S m/z 283.0768, found
283.0770.
Two trifluoromethyl derivatives were isolated from the crude as
a mixture (w60/40), which probably corresponds to the cyclo-
adduct 2c and the intermediate 4 from HF elimination. Cycloadduct
2c: 19F NMR (CDCl3,
d
ppm): ꢁ72.4 (d, 3JF,F¼7.2 Hz, 3F, CF3), ꢁ161.8
(m, 1F, CF). Intermediate 4: 19F NMR (CDCl3,
d
ppm): ꢁ54.1 (br s,
CF3). Selected 1H NMR (CDCl3,
d ppm) of the mixture 2c and 4: 1.55
(s, 3H, CH3), 1.61 (s, 3H, CH3), 1.64 (s, 3H, CH3), 1.66 (s, 3H, CH3), 2.35
(d, 2JH,H¼17.9 Hz, 2H, CH2), 2.50 (d, 2JH,H¼18.7 Hz, 2H, CH2), 2.97 (d,
2JH,H¼19.0 Hz, 1H, CH2), 3.11 (d, 2JH,H¼19.0 Hz, 1H, CH2).
4.2.1. Reaction of ketene dithioacetal 1a with 2,3-dimethylbuta-1,3-
diene (Scheme 1, Table 2: entry 1). (2-Fluoro-4,5-dimethylphenyl)
ethylsulfide (3a). This compound was purified by silica gel column
chromatography, eluting with a mixture (100/1) of petroleum ether
and dichloromethane. Yield: 31%. Yellow oil. 1H NMR (CDCl3,
4.2.5. Reaction of ketene dithioacetal 1b with isoprene (Scheme
3). (2-Trifluoromethyl-5-(and 4-)methylphenyl)ethylsulfide (5a)
and (5b). These compounds were purified by silica gel column
chromatography, eluting with petroleum ether. Yield: 35%. Mixture
of regioisomers 5a/5b (35/65). Yellow oil. MS (EI): m/z¼220 [Mþ],
205 [MꢁMe]þ, 172, 127. HRMS (EI): calcd for C10H11F3S m/z
220.0534, found 220.0535. For thorough assignment of
regioisomers 5a,b, see also Supplementary data.
d
ppm): 1.25 (t, 3JH,H¼7.3 Hz, 3H, SCH2CH3), 2.20 (s, 3H, CH3), 2.22 (s,
3H, CH3), 2.87 (q, 3JH,H¼7.3 Hz, 2H, SCH2CH3), 6.85 (d, 3JH,F¼10.2 Hz,
1H, Ar), 7.14 (d, JH,F¼7.5 Hz, 1H, Ar). 19F NMR (CDCl3,
d
d
ppm):
ppm):
4
ꢁ115.4 (dd, JF,H¼10.2 Hz, JF,H¼7.5 Hz). 13C NMR (CDCl3,
3
4
4
14.6 (s, CH2CH3), 18.9 (s, CH3), 19.5 (d, JC,F¼1.7 Hz, CH3), 28.1 (d,
4JC,F¼2.2 Hz, CH2CH3), 116.6 (d, JC,F¼22.5 Hz, CHeCF), 118.5 (d,
2
2JC,F¼17.6 Hz, Cq, CeSEt), 132.4 (d, 4JC,F¼3.8 Hz, Cq, CeCH3), 134.0 (d,
Minor regioisomer 5a: 1H NMR (CDCl3,
d ppm): 1.33 (t,
3JC,F¼2.2 Hz, CHeCS), 137.7 (d, JC,F¼7.1 Hz, Cq, CeCH3), 160.6 (d,
3JH,H¼7.4 Hz, 3H, SCH2CH3), 2.38 (s, 3H, CH3), 2.99 (q, 3JH,H¼7.4 Hz,
3
3
1JC,F¼240.5 Hz, CeF). MS (EI): m/z¼184 [Mþ], 169, 156, 123. HRMS
(ESIþ): calcd for C10H14FS m/z 185.0800, found 185.0803.
2H, SCH2CH3), 7.05 (d, JH,H¼8.1 Hz, 1H, CH, Ar), 7.27 (br s, 1H, CH,
Ar), 7.52 (d, 3JH,H¼8.1 Hz,1H, CH, Ar). 19F NMR (CDCl3,
ppm): ꢁ60.5
d
(s).
4.2.2. Reaction of ketene dithioacetal 1b with 2,3-dimethylbuta-1,3-
diene (Scheme 1, Table 2: entry 2). (2-Trifluoromethyl-4,5-
dimethylphenyl)ethylsulfide (3b). Yield: 63%. Yellow oil. 1H NMR
Major regioisomer 5b: 1H NMR (CDCl3,
d ppm): 1.30 (t,
3JH,H¼7.4 Hz, 3H, SCH2CH3), 2.37 (s, 3H, CH3), 2.95 (q, 3JH,H¼7.4 Hz,
3
2H, SCH2CH3), 7.27 (d, JH,H¼8.1 Hz, 1H, CH, Ar), 7.39 (d,
(CDCl3,
d
ppm): 1.30 (t, 3JH,H¼7.4 Hz, 3H, SCH2CH3), 2.27 (s, 3H, CH3),
3JH,H¼8.1 Hz, 1H, CH, Ar), 7.46 (s, 1H, CH, Ar). 19F NMR (CDCl3,
3
2.29 (s, 3H, CH3), 2.95 (q, JH,H¼7.4 Hz, 2H, SCH2CH3), 7.26 (m, 1H,
d
ppm): ꢁ60.7 (s).
Ar), 7.40 (m, 1H, Ar). 19F NMR (CDCl3,
(CDCl3,
d
ppm): ꢁ60.6 (s). 13C NMR
Three other trifluoromethyl derivatives were observed in the 19F
d
ppm): 14.0 (s, CH2CH3), 19.2 (s, CH3), 19.6 (s, CH3), 28.9 (s,
NMR spectra of the crude mixture, which probably correspond to
the mixture of cycloadducts or intermediates coming from HF or
EtSH elimination; nevertheless, more precise assignment was not
1
2
CH2CH3), 124.0 (q, JC,F¼273.5 Hz, CF3), 127.8 (q, JC,F¼29.6 Hz, Cq,
CeCF3), 127.9 (q, 3JC,F¼5.4 Hz, CHeCCF3), 132.3 (s, Cq, CeCH3), 132.9
(s, CHeCSEt), 134.7 (s, Cq, CeCH3), 140.8 (s, Cq, CeSEt). MS (EI): m/
z¼234 [Mþ], 219, 186. HRMS (ESIþ): calcd for C11H13F3S m/z
234.0646, found 234.0652.
possible. Selected data of 19F NMR (CDCl3,
d
ppm): ꢁ73.6 (dm,
J¼6.1 Hz, 3F, CF3), ꢁ73.8 (dm, J¼6.1 Hz, 3F, CF3), ꢁ76.3 (dd, J¼11.3,
6.1 Hz, 3F, CF3). GCeMS (EI): m/z¼241 [Mþ1] (C10H12F4Sþ1).
4.2.3. Reaction of ketene dithioacetal 1d with 2,3-dimethylbuta-1,3-
diene (Scheme 1, Table 2: entry 3). (2-Pentafluoroethyl-4,5-
dimethylphenyl)ethylsulfide (3d). This compound was purified by
silica gel column chromatography, eluting with a mixture (100/1) of
petroleum ether and dichloromethane. Yield 60%. Yellow oil. 1H
4.2.6. Reaction of ketene dithioacetal 1b with penta-1,3-diene
(Scheme 4). (2-Trifluoromethyl-6(or 3)-methylphenyl)ethylsulfide
(6a) or (6b). The crude reaction mixture was complex, containing
several fluorinated compounds. Among them, compounds 6a,b
were isolated by silica gel chromatography, eluting with petroleum
ether. Yield: 40%. Mixture of regioisomers (w90/10). Yellow oil. MS
(EI): m/z¼220 [Mþ], 205 [MꢁMe]þ, 191, 172, 127.
NMR (CDCl3,
d
ppm): 1.30 (t, 3JH,H¼7.3 Hz, 3H, SCH2CH3), 2.28 (s, 3H,
CH3), 2.30 (s, 3H, CH3), 2.97 (q, 3JH,H¼7.3 Hz, 2H, SCH2CH3), 7.25 (m,
1H, Ar), 7.30 (m, 1H, Ar). 19F NMR (CDCl3,
d
ppm): ꢁ84.0 (m, 3F, CF3),
Major regioisomer: 1H NMR (CDCl3,
d
ppm): 1.21 (t, 3JH,H¼7.3 Hz,
ꢁ108.2 (m, 2F, CF2). 13C NMR (CDCl3,
d
ppm): 13.8 (s, CH2CH3), 19.2
3H, SCH2CH3), 2.62 (s, 3H, CH3), 2.75 (q, 3JH,H¼7.3 Hz, 2H, SCH2CH3),
3
3
3
(s, CH3), 19.7 (s, CH3), 28.5 (s, CH2CH3), 110e120 (m, CF2), 119.4 (qm,
7.31 (dd, JH,H¼7.8, JH,H¼7.5 Hz, 1H, CH, Ar), 7.45 (d, JH,H¼7.5 Hz,
2
3
1JC,F¼286.9 Hz, CF3), 125.0 (t, JC,F¼21.8 Hz, Cq, CeCF2), 129.8 (t,
1H, CH, Ar), 7.57 (d, JH,H¼7.8 Hz, 1H, CH, Ar). 19F NMR (CDCl3,
3JC,F¼8.6 Hz, CHeCCF2), 131.9 (s, CHeCSEt), 134.2 (s, Cq, CeCH3),
134.5 (s, Cq, CeCH3), 141.0 (s, Cq, CeSEt). MS (EI): m/z¼285 [Mþ1],
d
ppm): ꢁ59.5 (s).
Minor regioisomer: 19F NMR (CDCl3,
d
ppm): ꢁ60.5 (s).

J.-P. Bouillon et al. / Tetrahedron 68 (2012) 8663e8669
8669
(d) Ruzicka, L. Angew. Chem., Int. Ed. Engl. 1990, 29, 1320e1357; (e) Silvester, M.
J. Aldrichimica Acta 1995, 28, 45e54; (f) Resnati, G. Il Farmaco 1990, 45,
1137e1167; (g) Begue, J. P.; Bonnet-Delpon, D. Bioorganic and Medicinal
Another fraction was obtained from silica gel column chroma-
tography and was analyzed by GCeMS, containing cycloadducts
7a,b, and intermediates 8a,b. Cycloadducts 7a,b: mixture of
regioisomers, both of them being a mixture of stereomers (ratio not
determined): GC: rt¼19.43 min, 19.52 min, 19.71 min, and
21.34 min; MS (EI): m/z¼302 [Mþ], 273, 241, 179, 159. Intermediates
8a,b: mixture of regioisomers (ratio not determined): GC:
rt¼21.55 min (minor), 24.12 min (major); MS (EI): m/z¼282 [Mþ],
253, 192, 171.
ꢀ
ꢀ
Chemistry of Fluorine; John Wiley & Sons: New York, NY, 2008.
6. (a) De Cock, C.; Piettre, S.; Lahousse, F.; Janousek, Z.; Merenyi, R.; Viehe, H. G.
Tetrahedron 1985, 41, 4183e4193; (b) Piettre, S.; De Cock, C.; Merenyi, R.; Viehe,
H. G. Tetrahedron 1987, 43, 4309e4319; (c) Purrington, S. T.; Sheu, K. W.
Tetrahedron Lett. 1992, 33, 3289e3292.
7. Cheng, D.; Zhou, J.; Saiah, E.; Beaton, G. Org. Lett. 2002, 4, 4411e4414.
8. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant,
J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone Mennucci, V. B.;
Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;
Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Och-
terski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dan-
nenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M.
C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Ste-
fanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02; Gaussian: Wall-
Acknowledgements
The authors thank V. Datinska for her help in the synthesis of
compounds 5a,b, Dr. J.-M. Nuzillard for NMR discussion, Pr. G.
Dupas for theoretical calculations discussion, and ‘Centre de
Ressources Informatiques de Haute-Normandie’ (CRIHAN) for
software optimization and technical support. Financial support by
CNRS, Conseil Regional Champagne Ardenne, Conseil General de la
Marne, Ministry of Higher Education and Research (MESR), and EU-
programme FEDER to the PlAneT CPER project is also gratefully
acknowledged.
ingford CT, 2004; This package has been implemented on
IBM cluster.
a power-575
9. (a) Mueller, K.; Sauer, J. Tetrahedron Lett. 1984, 25, 2541e2544; (b) Dvorak,
D.; Arnold, Z. Tetrahedron Lett. 1982, 23, 4401e4402; (c) Ichikawa, Y.;
Narita, A.; Shiozawa, A.; Hayashi, Y.; Narasaka, K. J. Chem. Soc., Chem.
Commun. 1989, 1919e1921; (d) Brueckner, R.; Huisgen, R. Tetrahedron Lett.
1990, 31, 2561e2564; (e) Narasaka, K.; Shibata, T.; Hayashi, Y. Bull. Chem.
Soc. Jpn. 1992, 65, 1392e1396; (f) Bates, R. W.; Pratt, A. J.; Rendle, P. M.;
Robinson, W. T. Aust. J. Chem. 1998, 51, 383e387; (g) Tietze, L. F.; Schuf-
fenhauer, A. Eur. J. Org. Chem. 1998, 1629e1637; (h) Vogt, K.; Schmidt, R. R.
Tetrahedron 1988, 44, 3271e3280; (i) Narasaka, K.; Kusama, H.; Hayashi, Y.
Bull. Chem. Soc. Jpn. 1991, 64, 1471e1478; (j) Pariza, R. J.; Hannick, S. M.;
Sowin, T. J.; Doherty, E. M. Eur. Pat. Appl. EP 452729 A1 19911023, 1991. (k)
Gallagher, T. C.; Hales, N. J.; Bradbury, R. H.; Wisedale, R. M., Sarah R. PCT
Int. Appl. WO 9747621 A1 19971218, 1997. (l) Sauer, J.; Heldmann, D. K.;
Hetzenegger, J.; Krauthan, J.; Sichert, H.; Schuster, J. Eur. J. Org. Chem. 1998,
12, 2885e2896.
Supplementary data
Supplementary data related to this article can be found online at
http://dx.doi.org/10.1016/j.tet.2012.07.054.
References and notes
1. (a) Smith, M. B. Organic Synthesis, 2nd ed.; McGraw-Hill: New York, NY, 2002,
pp 844e849; (b) Cycloaddition Reactions in Organic Synthesis; Kobayashi, S.,
Jorgensen, K. A., Eds.; Wiley-VCH: Weinheim, 2002; (c) Nicolaou, K. C.; Snyder,
S. A.; Montagnon, T.; Vassilikoyiannakis, G. Angew. Chem., Int. Ed. 2002, 41,
1650e1667; (d) Kagan, H. B.; Riant, O. Chem. Rev. 1992, 92, 1007e1019.
2. (a) Samusenko, Y. V.; Aleksandrov, A. M.; Yagupol’skii, L. M. Zh. Org. Khim. 1975,
11, 626e632; (b) Al’bekov, V. A.; Benda, A. F.; Gontar’, A. F.; Sokolskii, G. A.;
Knunyants, I. L. Bull. Acad. Sci. USSR (English Translation) 1988, 37, 765e769; (c)
Chanteau, F.; Plantier-Royon, R.; Haufe, G.; Portella, C. Tetrahedron 2006, 62,
9049e9053; (d) Hajduch, J.; Paleta, O.; Kvicala, J.; Haufe, G. Eur. J. Org. Chem.
2007, 30, 5101e5111; (e) Qiu, W.; Peng, S. U.S. 7,078,575 B1, 2006.
10. The full 1H and 13C assignment of both regioisomers 5a,b was achieved through
the thorough analysis of the COSY, HSQC, and HSQC spectra. See Supplementary
data.
11. Quiclet-Sire, B.; Saicic, R. N.; Zard, S. Z. Tetrahedron Lett. 1996, 37, 9057e9058.
12. Cabiddu, M. G.; Cabiddu, S.; Cadoni, R.; Fattuoni, C.; Floris, C.; Melis, S. J. Or-
ganomet. Chem. 1997, 531, 125e140.
13. Murata, M.; Buchwald, S. L. Tetrahedron 2004, 60, 7397e7403.
14. Tanaka, K.; Nakai, T.; Ishikawa, N. Chem. Lett. 1979, 175e178.
15. Muzard, M.; Portella, C. J. Org. Chem. 1993, 58, 29e31.
3. Portella, C.; Bouillon, J. P. In Perfluoroketene Dithioacetals: Versatile Building
Blocks Towards Trifluoromethyl Heterocycles; Soloshonok, V., Ed.; ACS, Sympo-
sium Series # 911; Oxford University Press: Washington, DC, 2005, pp 232e247.
ꢀ
16. Bouillon, J. P.; Henin, B.; Huot, J. F.; Portella, C. Eur. J. Org. Chem. 2002,
ꢀ
1556e1561.
4. Portella, C.; Muzard, M.; Bouillon, J. P.; Brule, C.; Grellepois, F.; Shermolovich,
17. Bouillon, J. P.; Kikelj, V.; Tinant, B.; Harakat, D.; Portella, C. Synthesis 2006,
1050e1056.
Yu. G.; Timoshenko, V. M.; Chernega, A. N.; Sotoca-Usina, E.; Parra, M.; Gil, S.
Heteroat. Chem. 2007, 18, 500e508.
18. Bouillon, J. P.; Tinant, B.; Nuzillard, J. M.; Portella, C. Synthesis 2004, 711e721.
19. Becke, A. D. J. Chem. Phys. 1993, 98, 5648e5652.
20. Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785e789.
5. (a) Wilkinson, J. A. Chem. Rev. 1992, 92, 505e519; (b) Hiyama, T. Organofluorine
Compounds. Chemistry and Applications; Springer: Berlin, 2000; (c) Welch, J. T.;
Eswarakrishnan, S. Fluorine in Bioorganic Chemistry; Wiley: New York, NY, 1991;