Do reactions of 2,2,4,4-tetrakis(trifluoromethyl)-1,3-dithietane require fluoride anion catalysis?

Article abstract of DOI:10.1016/j.jfluchem.2015.04.003

2,2,4,4-Tetrakis(trifluoromethyl)-1,3-dithietane (cyclic dimer of hexafluorothioacetone (1)) was shown to react with a variety of organic substrates in the absence of metal fluoride catalyst at elevated temperature in DMSO or DMF solvents. While the react

Full text of DOI:10.1016/j.jfluchem.2015.04.003

Accepted Manuscript
Title: Do Reactions of
2,2,4,4-Tetrakis(trifluoromethyl)-1,3-dithietane Require
Fluoride Anion Catalysis?
Author: Viacheslav A. Petrov Will Marshall
PII:
S0022-1139(15)00099-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jfluchem.2015.04.003
FLUOR 8541
To appear in:
FLUOR
Received date:
Revised date:
Accepted date:
23-2-2015
8-4-2015
9-4-2015
Please cite this article as: V.A. Petrov, W. Marshall, Do Reactions of 2,2,4,4-
Tetrakis(trifluoromethyl)-1,3-dithietane Require Fluoride Anion Catalysis?, Journal of
Fluorine Chemistry (2015), http://dx.doi.org/10.1016/j.jfluchem.2015.04.003
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Do Reactions of 2,2,4,4-Tetrakis(trifluoromethyl)-1,3-
dithietane Require Fluoride Anion Catalysis?
Viacheslav A. Petrov and Will Marshall
DuPont Central Research and Development
Experimental Station
PO Box 80500,
Wilmington DE 19880-0500
CF3
CF3
S
CF3
CF3
1
S
7
0oC, 1-2 h
DMSO
CF3
CF3
S
S
S
F C
CF3
CF3
3
1
=
yields - >98%
F3C
Graphical
Abstract
1
Page 1 of 25

Do Reactions of 2,2,4,4-Tetrakis(trifluoromethyl)-1,3-dithietane Require Fluoride Anion Catalysis?
*
, a
b
Viacheslav A. Petrov and Will Marshall
a ≠
DuPont Central Research and Development
b
DuPont Corporate Center for Analytical Sciences
Experimental Station
PO Box 80500, Wilmington DE 19880-0500
Email: viacheslav.a.petrov@dupont.com
Abstract
2
,2,4,4-Tetrakis(trifluoromethyl)-1,3-dithietane (cyclic dimer of hexafluorothioacetone (1)) was shown to
react with a variety of organic substrates in the absence of metal fluoride catalyst at elevated temperature in DMSO
or DMF solvents. While the reactions of vinyl ethers, phenyl vinyl sulfide, dienes and anthracenes under these
conditions led to the formation of expected cycloadducts, the reaction of 1,5-cyclooctadiene resulted in the
formation of the product derived from an ene reaction with hexafluorothioacetone (HFTA). In contrast to results
reported for the CsF catalysed process, 2-vinylnaphthalene reacted with 1 to produce a compound that resulted
from reaction with two molecules of HFTA. The structure of this adduct was established using single crystal X-ray
diffraction. The formation of similar products was also observed in non-catalysed reactions of styrene and 2-
methylstyrene with 1. Interestingly, no product of this type was observed in reaction of 4-MeO-styrene with 1, which
instead led to the high yield formation of the corresponding thietane. The reaction of 1 with 1,3-cyclohexadiene was
carried out in various organic solvents: the reaction in DMSO and DMF was significantly faster than in
acetonitrile, tetrahydrofurane or hexane.
≠
Publication # 2015-1179
3
Page 2 of 25

Keywords: Reactions of cyclic dimer hexafluorothioacetone; 2,2,4,4-tetrakis(trifluoromethyl)-1,3-dithietane;
Research Highlights.



 The reaction of hexafluorothioacetone dimer does not require MF catalyst
o
 It can be carried out in DMSO or DMF solvent at 20-70 C
 These reactions are as efficient as catalyzed processes
1. Introduction.
The cyclic dimer of hexafluorothioacetone (2,2,4,4-terakis(trifluoromethyl)-1,3-dithietane (1)) is a
valuable synthon which has been used to prepare a variety of fluorine-containing martials. It has been
shown to react to give hexafluorothioacetone (HFTA) cycloaddition products with vinyl ethers [1-3] vinyl
sulfides [2], vinyl amides [4], various dienes [5-7], styrenes [5, 8], fluoroolefins [2, 9]. It has also been
shown to give the product of HFTA insertion into allylic C-H bond with certain hydrocarbon olefins [10-
12]. Most of these reactions were carried out in the presence of an alkali metal fluoride catalyst (either KF
or CsF), since in earlier publications [1, 5, 9] it was suggested that the catalyst was required to induce the
equilibrium between 1 and monomeric hexafluorothioacetone.
In contrast, a significant number of reactions of 1 have been reported that did not contain metal
fluoride catalyst, including substrates such as phosphorous, arsenic, germanium and silicon cyanides [13-
4
Page 3 of 25

1
5], thiourea, dithiooxamide and cyanoformamidines[16-18], 1,1-dimethoxyethylene [3], N-vinylamides
[
4], azoles [19], N-alkyl(aryl)imidazoles [20].
In order to further elucidate the chemical behavior of 1, we undertook a separate study of its
reported reactions, but in the absence of alkali metal fluoride catalysts. The observed results demonstrate
that when using a polar nucleophilic solvent, the majority of the reactions involving 1 do not require the
catalyst.
2. Results and Discussion.
Previously [7] we reported that the reaction of 1 with quadricyclane (2) carried out in diglyme in
the presence of CsF catalyst resulted in rapid formation of the corresponding cycloadduct 3. Interestingly,
the formation of the same product, albeit at a significantly slower rate, was observed when this reaction
was carried out at ambient temperature in the absence of either catalyst or solvent [7]. Intrigued by this
result, we repeated the reaction of 1 with 2 in the absence of catalyst, but using a polar aprotic solvent -
o
19
dimethylsulfoxide (DMSO). Surprisingly, at 50 C, the reaction was complete (by F NMR) after 10 min.
to give a high yield of 3 (Eq.1).
S
S
C(CF )
3 2
o
F3C
F3C
CF3
CF3
50 C, 30 min
+
S
DMSO
1
2
3, quant. (NMR)
(1)
We also examined other previously reported KF or CsF catalyzed reactions of 1 by running the reactions
in the absence of catalyst. To our surprise, the majority of these reactions did not require metal fluoride
catalysis, as long as the reaction was carried out in a polar solvent such as DMSO or DMF.
5
Page 4 of 25

o
One example, the reaction of alkylvinyl ethers 4a,b with 1 in DMSO (70 C, 2 h) led to high
yields of previously reported [2] thietanes 5a and 5b, along with smaller amounts of 1,3-dithiolanes 6a
and 6b (Eq.2, Table 2).
CF3
DMSO
1
+
CF
3
7
0oC, 2 h
RO
S
OR
a, R=n-C H
4
4
5a,R=n-C H , 95%
3 7
5b, R=n-C H , 95%
3
7
b, R=n-C H
4 9
4
9
+
OR
S
S
F3C CF3
6a,R=n-C H , 5%
3 7
6
b, R=n-C H , 5%
4
9
(2)
o
When one equivalent of sulfur was included in this reaction (DMSO, 70 C, 24 h) the major product was
compound 5a (and trace amount of 6a), rather than the expected 1,2-dithiolane 6c [2]. However, when dry
CsF catalyst was added to the existing reaction mixture, complete conversion of 5a into 6c was observed
o
within 8 h at 70 C. (Eq.3)
OC H -n
3
7
S
F3C
F3C
5
a
7
0oC,
DMSO
CsF,
0oC,
7
8
8h
1
+ 4a + Sx
h
Sx
DMSO
Sx
OC H -n
3
7
F3C
F3C
S
S
6c, >95% (NMR)
6
Page 5 of 25

(3)
This experiment demonstrated that while the formation of thietane 5a does not require catalyst, the ring-
expansion process leading to 6c is catalyzed by CsF. It is likely that in this case CsF reacts with elemental
sulfur (rather than 5a), converting it into an anionic species that is then able to react with 5a.
o
Phenylvinyl sulfide (7), under similar reaction conditions, was completely consumed after 3 h at 70 C to
give thietane 8 in 85 % isolated yield (Eq.4). The structure of this material was established by single
crystal X-ray diffraction.
CF3
DMSO
1
+
CF
3
7
0oC, 3h
PhS
, yield 85%
S
SPh
7
8
(4)
1
,3-Cyclohexadine (9), 1,3,5-cycloheptatriene (10) and 2,3-dimethyl-1,3-butadiene (11) all readily
reacted with 1 at elevated temperature in DMSO to produce the corresponding cycloadducts 12-14 in
nearly quantitative yields.
7
Page 6 of 25

CF3
S
CF3
1
2
9
1
CF3
CF3
1
S
7
0oC, 1-2 h
DMSO
0
1
3
CF3
CF3
S
11
14
12,13,14 - yields - >98% (NMR)
(5)
The reaction of anthracenes 15-17 with 1 were carried out at ambient temperature in DMF. Although
these reactions were relatively slow (2 -7 days), the corresponding cycloadducts 18-20 were isolated in
68-81 % yield.
F3C
S
R
CF3
2
5oC, 2-7 d
1
+
DMF
R
18, R=H, 81%
19, R=Me, 68%
20, R=Br, 81%
1
1
1
5, R=H
6, R=Me
7, R=Br
(6)
In the case of substituted anthracenes 16 and 17, the formation of only one regioisomer was observed.
The structures of products 18-20 were established by single crystal X-ray diffraction. Compounds 19 and
20 both were shown to exist as the isomer in which sulfur was connected to the carbon bearing the R-
group (Fig. 1).
8
Page 7 of 25

Fig.1 Single Crystal Structure for 19 with thermal ellipsoids drawn to the 30% probability level.
Ene reactions involving HFTA were first reported by Middleton [21], but later it was found, that
1
in the presence of KF catalyst, can be used for preparation of variety of allyl hexafluoro-i-propyl
sulfides [5, 11, 22]. We found that 1 produced ene- product in the reaction with 2-methoxypropene (21)
in DMSO in the absence of CsF catalyst. Indeed, the agitation of the mixture of 1 and an excess of 21
(
two layers due to limited solubility of 1 in DMSO) led to the homogenization of the reaction mixture and
formation of compound 22 as major product (Eq.7), along with smaller amounts of bis- and tris-
9
Page 8 of 25

substituted adducts (Eq.7). Compound 22 was not isolated, but it was characterized in reaction mixture
by NMR and GC/MS.
2
5oC, 10 d
DMSO
1
+ CH =C(OCH )CH
2 3 3
CH =C(OCH )CH SCH(CF )
2 3 2 3 2
21
22, 70%
(7)
The ene reaction was previously found to play an important role in the CsF catalyzed reaction of
non-conjugated dienes (such as 1,4-cyclohexadiene and 1,5-cyclooctadiene (23)) with 1. In both cases the
initial step was insertion of HFTA into the allylic C-H bond of the diene (leading to the formation of
cycloadducts with conjugated double bonds) followed by Diels-Alder reaction involving a second mole of
HFTA [6].
The outcome of the non-catalyzed reaction of 1,5-cyclooctadiene (23) with 1 was found to be
o
different from the result reported for catalyzed process [6]. At 70 C in DMSO, 23 was consumed within
1
19
3
h, but H and F NMR spectra of the reaction mixture revealed the presence of at least two isomeric
species resulting in significant broadening of the NMR signals. A longer reaction time (12 h) led to
simplification of the NMR spectra of the crude reaction mixture, from which compound 25 (the isomer of
24 reported to form in the catalyzed process [6]), was isolated in 51% yield.
1
0
Page 9 of 25

(
F C) HCS
3 2
7
0oC, 12 h
1
+
DMSO
23
(F C) HCS
3 2
24
(
F C) HCS
SCH(CF3)2
3
2
25, yield 51%
(8)
The structure of 25 was established by single crystal X-ray diffraction (Fig. 2).
Fig.2. Single Crystal Structure for 25 with thermal ellipsoids drawn to the 30% probability level.
1
1
Page 10 of 25

Obviously, the formation of 25 involved isomerization of isomer 24 [6] by the migration of both C=C
bonds under the reaction conditions.
The ene reaction was also found to play an important role in non-catalyzed reactions of 1 with
styrenes. For example, the reaction of 2-vinylnaphthalene (26) with excess 1 rapidly and selectively led to
the formation of bis-adduct 27 (NMR yield > 95%) rather than expected mono-adduct B [8] (Eq.9).
F3C
F3C
7
0oC,
1
+
S
DMSO
H
1,3-H-shift
A
26
SCH(CF3)2
F3C
F3C
F3C
F3C
S
S
B
27, yield 18.5%
(9)
Although the isolated yield of 27 was low (the isolation protocol was not optimized), its structure was
established by single crystal X-ray diffraction.
1
2
Page 11 of 25

Fig. 3 Single Crystal Structure for 27 with thermal ellipsoids drawn to the 30% probability level.
An initial Diels-Alder reaction produced compound A, which then underwent an ene reaction [8] with a
second mole of HFTA to give 27. The transient formation of A was clearly seen in NMR spectra taken
o
throughout the reaction: after 1 h at 65 C, the reaction mixture contained A and 27 in a ratio of 59:41,
while after 12 h only compound 27 was observed.
Interestingly, the formation of adducts similar to 27 was also observed in the reactions of 1 with
styrene (28) and 4-methylstyrene (30). In both cases, the reaction was not as selective, leading to the
formation of three products in each case – 29a-c and 31a-c, respectively.
1
3
Page 12 of 25

F3C CF3
DMSO
S
H
o
1
+
70 C,
CF3
R
S
20-48h
CF3
29a, R=H
R
3
1a, R=CH3
2
3
8, R=H
0, R=CH3
SCH(CF3)2
CF3
CF3
+
F3C
F3C
+
S
Ar
S
R
29b, R=H
29c, Ar=C H
6
5
31b, R=CH3
31c, Ar=4-Me-C H4
6
Ratio 29a:29b:29c - 46:25:29
Ratio 31a:31b:31c - 26:2:72
(10)
As shown in Eq. 10, a decrease in the amount of products 29b and 31b (29% for X=H vs. 2 % for
X=CH
) and increase in the amounts of thietanes 29c and 31c (25% for X=H vs. 72% for X=CH
3
3
)
correlates well with an increase in the electron-donating properties of the substituent on the phenyl ring.
While no formation of product 29b was observed in the CsF catalyzed reaction of 1 with styrene [8], in
the reaction of monomeric HFTA with styrene reported by W. Middleton [23], the formation of two
isomeric 2:1 adducts was observed. According to Middleton, the structure of 29a was assigned to the
major isomer, while the structure of the second adduct was not established [23]. Based on the data
obtained in this present work, it is reasonable to assume that the second isomer reported by Middleton had
the structure 29b. Mechanistically, both isomers 29a and 29b formed from the same Diels-Alder
intermediate C (Eq.11) by reaction with a second mole of HFTA through either a second Diels-Alder
reaction leading to 29a, or an ene- reaction (driven by aromatization) leading to the formation of isomer
29b.
1
4
Page 13 of 25

2
(CF ) C=S
1
3 2
(
CF ) C=S
3 2
CF3
CF3
S
28
H
"C"
(
CF ) C=S
(CF ) C=S
3 2
3
2
F3C CF3
SCH(CF3)2
S
H
CF3
CF3
R
CF
3
CF3
S
S
9b
29a
2
(Eq. 11)
The absence of product 29b in the CsF catalyzed reaction of 1 with 28 [8] is likely the result of the
instability of 29b in the presence of fluoride anion. Indeed, when dry CsF was added to the reaction
mixture obtained by the non-catalyzed process (the ratio of 29a:b:c - 46:29:25), the complete
19
o
disappearance of the signals of 29b was observed in the F NMR spectrum after 8 h at 65 C, resulting in
a reaction mixture containing only 29a and 29c (ratio 63:37, see Experimental Section).
In the case of the reaction between p-methoxystyrene (32) and 1, the formation of products
analogous to 29b or 31b was not observed, and the reaction led to predominant formation of thietane 33c
along with a small amount of adduct 33a (Eq. 12). Due to its higher selectivity, this reaction to make 33c
is preferred over the CsF catalyzed reaction.
1
5
Page 14 of 25

CF3
CF3
DMSO
1
+
7
0oC, 20h
S
OCH3
32
33c
H CO
3
+
F3C CF3
S
H
CF3
H CO
3
S
CF3
33a
ratio 33c:33a - 98:2
(12)
3. Mechanistic Considerations.
The fact that 1 is able to react with a variety of organic substrates in the absence of the catalyst
strongly suggests that in polar solvents with high nucleophilicity, such as DMSO or DMF, compound 1
may exist in equilibrium with monomeric HFTA. In order to check this hypothesis we carried out a series
of small scale (0.01 mol) experiments involving the reaction of 1 (10 mol % excess) with 1,3-
cyclohexadiene (9) in different solvents, and compared the conversion of 9 to cycloadduct 10 (see Eq. 5)
o
after 45 min at 65 C. While the reaction in DMSO and DMF under these conditions resulted in 98 and
8
7% conversion of 9 to 10 (respectively), the reaction in acetonitrile (ACN) was significantly slower,
resulting in only 0.7 % conversion (see Table 1). This result was comparable to the value obtained for the
reaction in DMSO at ambient temperature (0.5%). The reaction in less polar tetrahydrofuran (THF) was
o
even slower, and gave only 0.1 % conversion (45 min, 65 C). Virtually no formation of compound 10
was observed in nonpolar hexane under the same conditions. Data summarized in Table 1 suggest that
1
6
Page 15 of 25

although the polarity of the solvent strongly influences the rate of the reaction, the obtained results cannot
be explained only by solvent polarity since the value of the dipole moment (as a reflection of polarity
of the reaction media) of ACN (3.92 D) is close to that of both DMF (3.82 D) and DMSO (3.96 D, Table
1
) [24].
6
5oC, 45 min
C(CF3)2
1
+
Solvent
S
10
9
a
Table 1. Reaction of 9 with 1 in different solvents
Solvent
Dipole moment Conversion
(
D)
of 9 (%)
b,c
DMSO
DMF
3.96
3.82
98
87
d
e
ACN
3.92
0.7
f
THF
1.75
<0.1
no rxn
g
Hexane
0
a
b
c
d
e
f
o
0
.01 mol scale, 15 mL solvent, 10 mol % excess of 1, 65 C, 45 min.
ref. [24], p.193
o
same reaction in DMSO at 25 C; conversion of 9 = 0.5 % after 45 min and 2.8% after 5h
ref. [24], p.178
ref. [24], p.7
ref. [24], p.434
ref. [24], p. 262
g
1
7
Page 16 of 25

Such difference in reaction rates can, however, be explained based on the nucleophilicity of these
solvents. Indeed, while DMSO and DMF are both relatively nucleophilic, ACN has significantly lower
nucleophilicity, which is reflected in the much higher resistance of ACN towards action of electrophiles.
As an example, DMF was reported to react at ambient temperature with relatively weak electrophile as
3
(CF )
2
C=CF
2
, resulting in formation of (CF
3
)
2
C=CHN(CH
3
)
2
[25, 26] and the formation of same product
o
was observed in the reaction of 1 and DMF at elevated temperature (70 C) [8], while such reactions are
not reported for ACN. The proposed mechanism for both reactions involves the initial attack of DMF on
the electron-poor double bond of either (CF
3
)
2
=CF [25, 26] or HFTA [8].
2
It is thus reasonable to propose that the mechanism of dissociation of 1 in polar solvents, such as
DMF, which involves nucleophilic attack of DMF on sulfur leading to the final formation of two moles of
HFTA monomer (Eq. 13).
S
F3C
F3C
CF3
CF3
+
O=C(H)NMe2
S
1
S
F3C
F3C
CF3
CF3
S
O-C(H)=NMe2
2
(CF ) C=S + O=C(H)NMe
3 2 2
(13)
A similar mechanism can be used for the dissociation of 1 in DMSO as well, since in this molecule the
S=O bond is strongly polarized and oxygen atom bears substantial negative charge.
1
8
Page 17 of 25

Acknowledgement.
Authors are grateful to Dr. C. Junk for helpful discussion and suggestions and Joe Buriak for technical
support.
4
. Experimental
1
19
H NMR and F NMR spectra were recorded on a Bruker DRX-500 (499.87 MHz) instrument using CFCl or TMS
3
as an internal standards. Unless stated otherwise, CDCl was used as a lock solvent. GC and GC/MS analyses were
3
carried out on a HP-6890 instrument, using HP FFAP capillary column and either TCD (GC) and mass selective
(GS/MS) detectors, respectively. Dry DMF, THF, ACN and DMSO (Aldrich) and all hydrocarbon starting materials
were obtained from commercial sources and used without further purification. Compound 1 was prepared according
modified procedure using CsF as a catalyst [2]. For most experiments 1 having purity ~96-98% (reminder DMF)
was used. For experiments involving the reaction of 1 with 9 in different solvents (see below), compound 1 with
purity 99.3 % (containing 0.55% of 3,3,5,5-tetrakis(trifluoromethyl)-1,2,4-trithiolane and 0.15 % of DMF) was used.
Due to a high ratio of sulfur to fluorine, elemental analysis were not attempted for new materials, and the purity of
all isolated compounds was established by GC and NMR spectroscopy to be at least 98%.
4
.1.Crystallography:
X-ray data for 19, 25, 27, 20, 18 and 8 were collected at -100˚C using a Bruker 1K CCD system equipped with a
sealed tube molybdenum source and a graphite monochromator. The structures were solved and refined using the
2
Shelxtl [27] software package, refinement by full-matrix least squares on F , scattering factors from Int. Tab. Vol. C
Tables 4.2.6.8 and 6.1.1.4. Crystallographic data (excluding structure factors) for the structures in this paper have
been deposited with the Cambridge Crystallographic Data Centre as supplementary publication nos. CCDC
#
1043351 to #1043353 and #1043357 to #1043359 . Copies of the data can be obtained, free of charge, on
1
9
Page 18 of 25

application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK, (fax: +44 1223 336033 or e-mail:
deposit@ccdc.cam.ac.uk).
4
.2. Reaction of 1 with substrates 2,4,7,9-11,21,23,28,30,32 (typical procedure).
A mixture of 12-35 mmol of 1 (~10-20% excess), 10-30 mmol of the corresponding substrate , in 15 mL dry DMSO
o
or DMF was agitated at 65-70 C for period of time specified in Table 2. The reaction mixtures were analyzed by
GC and NMR. Compounds 3 [7], 5a,5b,6a-c [2], 12-14 [6], 29a, 29c, 31a, 31c, 33a, 33c [8] were identified in
reaction mixture by comparison of NMR data with values reported in the literature. Reaction conditions, ratio of
1
13
19
reagents, and conversions are given in Table 2. H, C, F NMR and mass-spectrometry data are given in Table 3.
4.3. Preparation of compounds 8 and 25.
A mixture of 25-35 mmol of 1 (~10-15 % excess), 20-30 mmol of the corresponding substrate in 15 mL dry DMSO
o
was agitated at 70 C for 3-12 h (see Table 2). The reaction mixtures were monitored by GC. The reaction mixture
was diluted with 200-300 mL of water, extracted by hexane (50 mL x 3), and combined organics washed by water
(
100 mL x 2), dried over MgSO and hexane was removed under reduced pressure to give crude product of 96-98 %
4
purity. Pure samples (~99%) of 8 and 25 were obtained by recrystallization from hexane. Reaction conditions, ratio
1
13
of reagents, yields, melting (boiling) points and data of mass-spectrometry for products are given in Table 2. H, C,
19
F NMR and mass-spectroscopy data are given in Table 3.
4.4. Preparation of compounds 18-20.
A mixture of 30 mmol of 1 (~10-15 % excess), 20 mmol of the corresponding anthracene in 15 mL dry DMF was
o
agitated at 25 C for 2-7 days (see Table 2). The reaction mixtures were monitored by GC. The reaction mixture was
diluted with 200-300 mL of water, extracted by hexane (50 mL x 3), and combined organics washed by water (100
mL x 2), dried over MgSO and hexane was removed under reduced pressure to give crude product of 96-98 %
4
purity. Pure samples (~99%) of 8 and 25 were obtained by recrystallization from hexane. Reaction conditions, ratio
1
13
19
of reagents, yields, melting points and data of mass-spectrometry for products are given in Table 2. H, C,
NMR and mass-spectroscopy data are given in Table 3.
F
2
0
Page 19 of 25

4.5 Isomerization of the mixture 29a-c.
o
An agitated mixture of 1.1 g (0.01 mmol) 28, 3.8 g (0.052 mol) of 1 and 12 mL of DMSO was kept at 65 C for 24 h
19
in a closed 20 mL glass vial. According to F NMR at this point the reaction mixture contained products 2a-c in
1
19
ratio 46:25:29, respectively ( H, F NMR). The reaction mixture was placed inside a dry box and 0.4 g (0.0026 mol)
of dry CsF was added and the heating was resumed. An aliquot of the reaction mixture was analyzed by NMR after
o
8
h at 65 C which showed the reaction mixture contained compounds 29a and 29c in the ratio 75:25.
4
.6 Relative rates of the reaction of 1 and 9 in different solvents.
To a 20 mL glass sample vial equipped with magnetic stir bar was added dry solvent (15 mL, Table 1), compound 9
(0.8 g, 0.01 mol, 98%, Aldrich), and compound 1 (2.0 g, 0.055 mol, 99.3%). Sealed sample vials were kept in a
o
heating block (65 C) with agitation. An aliquot of the reaction mixture was taken after 45 min. and analyzed by
1
19
GC/MS and H and F NMR. Conversions and values of dipole moments of solvents are given in Table 1.
a
Table 2. Reaction of 1 with Different Substrates in the Absence of Metal Fluoride Catalyst.
c
o
c
Entry No
Reagent
mols)
Solv. (mL) Temp.
(Time, h)
Conv.
Selectivity (%)
Product(s) (ratio, m.p C)
b
(
(%)
o
1
2 (0.01)
DMSO
15)
DMSO
15)
50 C (30 Quan 100
3
(
min)
t.
2
4a (0.01)
70 (2h)
100
95
5a, 6a (95:5)
(
DMSO
15)
3
4b (0.01)
70 (2h)
100
95
5b, 6b (95:5)
(
2
1
Page 20 of 25

DMSO
15)
4
5c (0.01)
70 (8h)
100
95
6c
(
Sx (0.015)
CsF (0.001)
d
e
DMSO
30)
5
6
7
8
9
1
7 (0.037)
9 (0.01)
10
70 (3h)
70 (2h)
70 (2h)
70 (2h)
25C,
100
100
100
100
100
100
>95
>99
>99
>99
>99
>99
8 (85%, m.p.=35.5-36.5C)
(
DMSO
15)
12
13
14
(
DMSO
15)
(
DMSO
15)
11
(
e
o
DMF (15)
DMF (15)
15(0.02)
16(0.02)
18 (80.6%, m.p. 123-124
(
48h)
25C,
48h)
C)
e
o
0
(67.5%, m.p. 142-143 C)
(
e
o
DMF (15)
DMSO(15)
1
1
1
1
2
3
17(0.02)
25C, (1 100
week)
>99
18(80.6%, m.p. 163-165 C)
f
g
2
1 (0.06)
22 (major)
25C,
100
~70
(10days)
e, h
DMSO
23(0.02)
70 (12h) 100
>99
25 (51% )
(15)
2
2
Page 21 of 25

e
DMSO
15)
1
1
4
5
26(0.02)
28 (0.01)
65 (12h) 100
70 (48h) 100
>95
27(18.5% , m.p. 53-55)
(
i
c
DMSO
15)
30
29a, 29b, 29c (50:30:20)
(
j
c
DMSO
15)
1
1
6
7
30 (0.01)
32(0.01)
70 (20h) 100
70 (20h) 100
3.5
31a,31b,31c (70:3.5:26.5)
33a, 33c (2:98)
(
DMSO
15)
(
a
reactions were carried out using 0.01-0.02 mol of organic substrate and 10 mol % excess of 1 in 15 ml of DMSO or DMF
solvent
b
conversion of organic substrate, NMR
NMR data
crude product contained ~ 5 % of unknown material (likely to be the corresponding 1,3-dithiolane), structure of which was not
c
d
established.
e
isolated yield
f
three mol excess of 21
g
reaction mixture also contained ~ 30 % of unsaturated products containing two and three –SCH(CF ) groups
3
2
h
i
purity 98%
selectivity for 29b
selectivity for 31b
j
a
Table 3. NMR and Mas-spectroscopy Data for New Materials
1
19
13
b
Compound No
H NMR
F NMR
C NMR
(δ, ppm, J, Hz)
35.23, 43.71,
MS
+
(
δ, ppm, J, Hz)
(δ, ppm, J, Hz)
(m/z, M )
+
+
8
3.28(2H, m),
-73.43(3F, q, 9.6),
318 (M ,C11
H
8
F
6
S
2
)
2
3
Page 22 of 25

5
7
.08(1H, t, 7.1),
-74.26(3F, q, 9.6)
47.80(sept., 32.2),
122.88(q, 278.0),
.40(3H,m), 7.45(2H, m)
1
1
24.46(q, 282.2),
29.07, 129.34, 130.64
+
1
8
5.12(1H,s), 5.23(1H,s),
-64.56(s)
46.58, 47.53,
362( M ,
+
7
7
.20 (4H,m), 7.34(2H,d),
.39(2H,d)
66.89(sept., 25.6),
122.46, 124.16(q,
C H F S )
17 12 6
2
1
86.0), 126.52, 126.84,
27.39, 138.40, 142.39
+
+
1
9
2.25(3H,s), 5.04(1H,s),
-64.86(s)
16.35, 47.69, 50.31
68.30(sept., 25.3),
119.49,
376 (M ,C18
H
14
F
6
S )
7
7
.12 (2H, t), 7.18(2H,t),
.26(2H,d), 7.33(2H,d)
1
1
1
24.20(q, 288.0),
26.35, 126.52,
27.17, 138.76, 144.91
+
2
0
5.16(1H,s), 7.27 (4H, m), -64.64(s)
.39(2H,d), 7.73(2H,d)
47.82, 62.84
441 (M ,
+
7
70.22(sept., 25.5),
C H BrF S )
1
7
11
6
1
1
1
1
22.95
23.80 (q, 288.0),
26.02, 127.74,
36.89, 142.76
2
2
2
7
3.41(3H,s), 3.55(2H, s),
-66.90(d, 7.7)
254
+
+
3
4
4
.90(1H, sept, 7.7),
.05(1H,d, 2.4),
.15(1H, d, 2.4)
(M ,C H F OS )
7 8 6
+
2.33(1H,dd, 14.8, 10.5),
-65.99(3F, sept, 8.9),
-66.54(3F, sept., 8.9),
-71.09(3F,q, 11.0),
-71.43(3F, q, 11.0)
35.74,46.02,
518 (M ,
+
3
3
4
7
8
.04(1H,dd, 14.7, 3.8),
.75(1H, sept., 8.0),
.47(1H,dd, 10.5, 3.5),
.61(2H,m), 7.87(3H,m),
.34(1H, d, 8.3)
51.68(sept. 31.2),
58.28(sept. 28.3),
122.67(q, 2777.5),
122.94(q, 278.5)’
123.28, 123.64,
C H F S )
1
8
10 12 2
2
4
Page 23 of 25

1
1
1
1
1
23.91(q,280.0),
26.52, 127.15,
27.67, 128.23,
28.65, 131.83,
32.20, 133.60
2
9b
3.29(2H,m),
-66.11(3F, sept., 8.8),
-66.73(3F, sept., 8.8),
-71.35(3F, q, 10.2),
3
4
7
7
7
.86(1H, sept., 7.5),
.22(1H, dd, 10.5, 3.5)
.31(1H,m), 7.39(1H,m), -71.62(3F, q, 10.2)
.46(1H,d, 7.2),
.67(1H, d, 7.2)
3
1b
-66.11(3F, sept., 9.2),
-66.66(3F, sept., 9.2),
-71.26(3F, q, 10.1),
-71.58(3F, q, 10.1)
a
in CDCl3
b
13
{
H} C NMR spectrum
[
[
[
[
1] T. Kitazume, T. Otaka, R. Takei, N. Ishikawa, Bull. Chem. Soc. Jpn. 49 (1976) 2491-2494.
2] V.A. Petrov, W. Marshall, J. Fluorine Chem. 131 (2010) 1144-1155.
3] V.A. Petrov, W. Marshall, J. Fluorine Chem. 143 (2012) 220-225.
4] V.A. Petrov, W. Marshall, Beilstein J. Org. Chem. 9 (2013) 2615-2626-2619.
[
[
[
5] T. Kitazume, N. Ishikawa, Lett. No. 3 (1973) 267-268.
6] V. Petrov, W. Marshall, J. Fluorine Chem. 128 (2007) 729-735.
7] V.A. Petrov, C.G. Krespan, W. Marshall, J. Fluorine Chem. 126 (2005) 1332-1341.
2
5
Page 24 of 25

[
[
8] V.A. Petrov, W. Marshall, J. Fluorine Chem. 132 (2011) 783-791.
9] S.A. Postovoi, I.M. Vol'pin, N.I. Delyagina, M.V. Galakhov, Y.V. Zeifman, L.S. German, Izv. Akad. Nauk
SSSR, Ser. Khim. (1989) 1380-1383.
[
[
[
[
[
[
10] B.B. Snider, L. Fuzesi, Tetrahedron Lett. (1978) 877-880.
11] D.J. Burton, Y. Inouye, J. Fluorine Chem. 18 (1981) 459-473.
12] D.J. Burton, Y. Inouye, Tetrahedron Lett. (1979) 3397-3400.
13] K. Burger, E. Huber, Chem.Ztg. 110 (1986) 211-213.
14] H.W. Roesky, K.S. Dhathathreyan, J. Chem. Soc., Chem. Commun. (1984) 1053-1054.
15] H.W. Roesky, K.S. Dhathathreyan, M. Noltemeyer, G.M. Sheldrick, Z. Naturforsch., B Anorg. Chem.,
Org. Chem. 40B (1985) 240-246.
[
(
[
[
16] H. Gruetzmacher, N. Keweloh, H.W. Roesky, M. Noltemeyer, G.M. Sheldrick, J. Fluorine Chem. 37
1987) 279-287.
17] G. Rabe, H.W. Roesky, R. Bohra, H.G. Schmidt, M. Noltemeyer, J. Fluorine Chem. 52 (1991) 235-244.
18] G. Rabe, J. Sundermeyer, H.W. Roesky, H.G. Schmidt, M. Noltemeyer, Chem. Ber. 123 (1990) 691-
6
96.
[
[
[
[
19] V.A. Petrov, W. Marshall, J. Fluorine Chem. 156 (2013) 262-267.
20] V.A. Petrov, W. Marshall, R. Dooley, J. Fluorine Chem. (2014) Ahead of Print.
21] W.J. Middleton, J. Org. Chem. (1965) 1395-1398.
22] Y. Inouye, Burton D., J. Fluorine Chem. 14 (1979) 89-92.
[
[
[
[
[
23] W.J. Middleton, J. Org. Chem. (1965) 1390-1394.
24] R.D. Lide, Handbook of Organic Solvents, CRC Press, London, 1995.
25] K. Cheburkov Yu.A., I.L., Izv. AN USSR. Ser. Khim. (1967) 346.
26] M.N. Cheburkov Yu.A., Aronov Y.E., Knunyants, I.L., Izv. AN USSR. Ser. Khim. (1965) 1478.
27] G. Sheldrick, (SHELXTL Sofware Suite, v. 6.0, Bruker Axes Corp., Madison WI, 1996.
2
6
Page 25 of 25