Highly selective anodic monofluorination of 4-arylthio-1,3-dioxolan-2-ones: A marked solvent effect on product selectivity

Article abstract of DOI:10.1039/b004652n

This is the first example of a solvent effect on fluorinated product selectivity; anodic fluorination of 4-arylthio-1,3-dioxolan-2-ones in CH₂Cl₂ containing a fluoride supporting electrolyte using an undivided cell provided the fluoro-desulfurization product, 4-fluoro-1,3-dioxolan-2-one preferentially while anodic fluorination in DME resulted in α-fluorination, without desulfurization, selectively.

Full text of DOI:10.1039/b004652n

Highly selective anodic monofluorination of 4-arylthio-1,3-dioxolan-2-ones: a
marked solvent effect on product selectivity
Hideki Ishii, Norihisa Yamada and Toshio Fuchigami*
Department of Electronic Chemistry, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama 226-8502,
Japan
Received (in Cambridge, UK) 12th June 2000, Accepted 17th July 2000
Published on the Web 8th August 2000
This is the first example of a solvent effect on fluorinated
product selectivity; anodic fluorination of 4-arylthio-1,3-di-
oxolan-2-ones in CH2Cl2 containing a fluoride supporting
electrolyte using an undivided cell provided the fluoro-
desulfurization product, 4-fluoro-1,3-dioxolan-2-one prefer-
entially while anodic fluorination in DME resulted in a-
fluorination, without desulfurization, selectively.
Next, we extended this anodic fluorination to the 4-(4-chloro-
phenylthio) derivative 1b. Similarly, the use of Et3N·5HF–
CH2Cl2 provided 2 exclusively in almost quantitative yield (run
7) while the use of Et4NF·4HF–DME afforded 3b6 selectively
in good yield (run 6). Thus, it was found that electrolytic
conditions, particularly electrolytic solvents, greatly affected
the fluorinated product selectivity. Such marked product
selectivity depending on electrolytic solvents has not been
previously reported for anodic fluorination.
In order to clarify the solvent effects, we investigated anodic
fluorination of 1b in a mixed solvent of DME and CH2Cl2
containing Et4NF·4HF. As shown in Fig. 1, the product ratio of
2 to 3b increased with an increase in the ratio of CH2Cl2 to
DME. Notably, addition of only 25% CH2Cl2 to DME caused a
dramatic change in the product ratio and 2 was mainly formed
in ca. 60% yield.
Fluorinated ethylene carbonates seem to be promising organic
electrolytic solvents or additives for rechargeable Li batteries
since introduction of fluorine atom(s) into ethylene carbonate is
expected to increase its electrochemical stability and decrease
its melting point. We have studied selective anodic fluorination
of various organic compounds.1,2 The selective anodic fluorina-
tion has many advantages since the fluorination can be carried
out in one step under safe conditions and the use of hazardous
reagents is not required. So far, limited examples of selective
anodic fluorination of oxygen-containing heterocycles have
been reported.3 With these facts in mind, we have attempted
anodic fluorination of ethylene carbonates having an arylthio
group 1 using various supporting fluoride salts and solvents.
At first, we investigated anodic fluorination of 4-phenylthio
derivative 1a under various conditions. Constant current
electrolysis was carried out at platinum plate electrodes in an
undivided cell at rt until 1a was completely consumed. The
results are summarized in Table 1.
As shown in Table 1, anodic fluorodesulfurization of 1a
proceeded selectively in Et4NF·4HF–CH2Cl2 and Et3N·5HF–
CH2Cl2 to provide monofluorinated product 24 in moderate to
good yields (runs 3 and 5). In sharp contrast, a-fluorination of
1a took place preferentially in Et4NF·4HF–DME to give 3a5
(run 4). In this case, difluorinated products such as 4 (Scheme 2)
were not formed. This can be explained in terms of the oxidation
potential of 3a (Eopx : 2.2 V vs. SCE) being 0.3 V higher than that
of 1a (Eopx : 1.9 V vs. SCE).
This interesting phenomenon can be explained as follows.
The fluorination can be rationalised by postulating a radical
cation intermediate A as shown in Scheme 1.
CH2Cl2 has a poor ability to solvate carbocations, therefore,
A seems to be unstable in CH2Cl2. Consequently it is reasonable
to assume that desulfurization followed by fluorination mainly
takes place prior to a-fluorination of A.7 On the other hand,
DME is known to strongly coordinate cations.8 Therefore, DME
should stabilize the intermediate A and DME also enhances the
fluoride ion nucleophilicity.9 Then, the deprotonation of A with
fluoride ions takes place prior to desulfurization followed by
further oxidation to generate cation B and this cation reacts with
a fluoride ion to provide the a-fluorinated product.
Furthermore, we examined anodic fluorodesulfurization of
3a. As shown in Scheme 2, anodic fluorodesulfurization of 3a
in Et4NF·4HF–CH2Cl2 and Et3N·5HF–CH2Cl2 proceeded to
give desired difluorinated product 410 in reasonable yield.
However, anodic fluorodesulfurization of 3a did not take place
at all in Et4NF·4HF–DME and 3a was almost recovered.
Finally, we also examined chemical fluorination of 1a.
However, treatment of 1a with various N-fluoropyridinium
Table 1 Anodic fluorination of 4-arylthio-1,3-dioxolan-2-ones
Fig. 1 Dependence of yield of 2 and 3b on the ratio of CH2Cl2 to DME.
DOI: 10.1039/b004652n
Chem. Commun., 2000, 1617–1618
This journal is © The Royal Society of Chemistry 2000
1617

Scheme 1
2 (a) T. Fuchigami, Advances in Electron-Transfer Chemistry, vol. 6, ed.
P. S. Mariano, JAI Press, CT, 1999, p. 41; (b) T. Fuchigami, S.
Higashiya, Y. Hou and K. M. Dawood, Rev. Heteroatom. Chem., 1999,
19, 67.
3 (a) Y. Hou, S. Higashiya and T. Fuchigami, J. Org. Chem., 1999, 64,
3346; (b) G. P. Gambaretto, M. Napoli, C. Franccaro and L. Conte,
J. Fluorine Chem., 1982, 19, 427; (c) J. H. Meurs and W. Eilenberg,
Tetrahedron, 1991, 47, 705.
4 4-Fluoro-1,3-dioxolan-2-one (2): colorless cubes: mp 19.0–20.0 °C;
1H-NMR (CDCl3, 270 MHz) d 6.31 (ddd, J = 64, 1.3, 4.0 Hz, 1H),
4.70–4.49 (m, 2H); 13C-NMR (CDCl3, 67.8 MHz) d 152.65, 104.98 (d,
J = 237 Hz), 70.67 (d, J = 28 Hz); 19F-NMR (CDCl3, 254 MHz) d
244.42 (ddd, J = 64, 34, 21 Hz); MS (m/z) 106 (M+), 62 (M 2 CO2);
Anal. calc. for C3H3FO3: C, 33.98; H, 2.85; F, 17.91; O, 45.26. Found
C, 33.73; H, 2.91; F, 17.72%.
Scheme 2
5 4-Fluoro-4-phenylthio-1,3-dioxolan-2-one (3a): colorless oil; 1H-NMR
(CDCl3, 270 MHz) d 7.66–7.62 (m, 2H), 7.50–7.39 (m, 3H), 4.63 (dd,
J = 16.8, 10.9 Hz, 1H), 4.47 (dd, J = 26.4, 10.9 Hz, 1H); 13C-NMR
(CDCl3, 67.8 MHz) d 150.64, 136.21, 130.89, 129.70, 125.03, 119.54
(d, J = 270 Hz), 73.88 (d, J = 31 Hz); 19F-NMR (CDCl3, 254 MHz) d
0.54 (dd, J = 27, 17 Hz); MS (m/z) 214 (M+); HRMS calc. for
C9H7FO3S: 214.0100. Found: 214.0133.
6 4-Fluoro-4-(p-chlorophenylthio)-1,3-dioxolan-2-one (3b): colorless
needles; mp 60.0–61.0 °C; 1H-NMR (CDCl3, 270 MHz) d 7.60–7.57 (m,
2H), 7.43–7.40 (m, 2H), 4.66 (dd, J = 16.8, 10.9 Hz, 1H), 4.47 (dd, J
= 26.4, 10.9 Hz, 1H); 13C-NMR (CDCl3, 67.8 MHz) d 150.40, 137.72,
137.43, 129.96, 123.34, 119.21 (d, J = 271 Hz), 73.83 (d, J = 31 Hz);
19F-NMR (CDCl3, 254 MHz) d 0.93(dd, J = 27, 17 Hz); MS (m/z) 250
(M+); HRMS calc. for C9H6ClFO3S: 247.9710. Found: 247.9720.
7 In this reaction, the formation of diphenyl disulfide was detected by MS
and 1H-NMR.
8 S. E. Hill, D. Feller and E. D. Glendening, J. Phys. Chem. A, 1998, 93,
159.
9 (a) K. M. Dawood, S. Higashiya, Y. Hou and T. Fuchigami, J. Fluorine
Chem., 1999, 93, 159; (b) Y. Hou and T. Fuchigami, Electrochem.
Commun., 1999, 1, 445.
10 4,4-Difluoro-1,3-dioxolan-2-one (4): 1H-NMR (CDCl3, 270 MHz) d
4.72 (t, J = 11 Hz, 2H); 19F-NMR (CDCl3, 254 MHz) d 3.90 (t, J = 12
Hz); MS (m/z) 124 (M+); HRMS calc. for C3H2F2O3: 123.9972. Found:
123.9939.
Scheme 3
triflates in CH2Cl2 resulted in the formation of none of the
desired fluorinated products, as shown in Scheme 3. Therefore,
electrochemical fluorination is more advantageous than con-
ventional chemical methods for such heterocyclic sulfides.
In summary, we have developed a novel synthesis of
fluorinated ethylene carbonates using anodic fluorination and
we also found a unique marked solvent effect on fluorinated
product selectivity.
We thank the Kato Foundation and the Ministry of Education,
Science, Sports and Culture of Japan for Financial support
(Grant-in-Aid for Scientific Research, No. 12555252).
Notes and references
1 (a) T. Fuchigami and K. M. Dawood, J. Org. Chem., 1999, 64, 138; (b)
S. Higashiya, A. Narizuka, A. Konno, K. Momota and T. Fuchigami,
J. Org. Chem., 1999, 64, 133.
1618
Chem. Commun., 2000, 1617–1618