Insight into "entrainment" in SRN1 reactions of 2,2-dichloro-1,1,1-trifluoroethane(HCFC-123) with thiolates initiated by Na2S2O4

Article abstract of DOI:10.1016/j.jfluchem.2014.10.006

An interesting entrainment process in S_RN1 reactions of 2,2-dichloro-1,1,1-trifluoroethane (HCFC-123) with thiolates were studied by experiments and DFT calculations. The radical-anion intermediate, generated from coupling of the fluorinated ra

Full text of DOI:10.1016/j.jfluchem.2014.10.006

Journal of Fluorine Chemistry 169 (2015) 1–5
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Insight into ‘‘entrainment’’ in SRN1 reactions of 2,2-dichloro-1,1,
1-trifluoroethane(HCFC-123) with thiolates initiated by Na2S2O4
*
Xiao-Jun Tang, Qing-Yun Chen
Key Laboratory of Organofluorine Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road,
200032 Shanghai, China
A R T I C L E I N F O
A B S T R A C T
Article history:
An interesting entrainment process in SRN1 reactions of 2,2-dichloro-1,1,1-trifluoroethane (HCFC-123)
with thiolates were studied by experiments and DFT calculations. The radical-anion intermediate,
generated from coupling of the fluorinated radical with the thiolate was considered as the key
intermediate in this reaction.
Received 3 March 2014
Received in revised form 7 October 2014
Accepted 12 October 2014
Available online
ß 2014 Elsevier B.V. All rights reserved.
Keywords:
Single electron transfer reaction
S
RN1 reaction
Radical-anion
Fluorinated radical
1. Introduction
According to these reports, depending on the number of
chlorine atom, different reaction mechanisms were involved in the
As transitional chemicals for CFCs in the field of refrigerants,
aerosol propellants and cleaning solvents, HCFCs will also be
abolished gradually [1]. In the sense of synthesis, however, some
HCFCs are indispensable for industrial applications and syntheses
of useful fluorinated building blocks. For instance, difluoromethyl
chloride (HCFC-22), one once widely used refrigerant, is not only
an efficient difluorocarbene reagent in the lab [2] but also the
starting material in manufacture of tetrafluoroethene (TFE) in the
industry. In general, compared with common alkyl chlorides,
most of HCFCs are incapable of undergoing SN1 or SN2 reactions
due to special electronic effect and steric hindrance of fluorinated
groups. In contrast, HCFCs are prone to receiving electron from
reductive reagents and undergoing single electron transfer (SET)
reactions of CFC-113a and HCFC-133a with thiolates in the same
solvent. The number of chlorine atom in CF3CHCl2 (HCFC-123) is in
between CFC-113a and HCFC-133a, so it will be attractive to check
reaction mechanism of HCFC-123 with thiolates. Herein, we
present results.
2. Results and discussion
To our surprise, no reaction was observed in the reaction of
HCFC-123 with sodium 4-methylbenzenethiolate even the reac-
tion temperature increased to 100 8C in DMSO or DMF. In other
words, this result showed that HCFC-123 undergoes neither SRN
1
reaction nor halophilic reaction when meeting the thiolate under
similar conditions. As an effective method for generating
fluorinated radicals from perfluoroalkyl iodides or bromides by
using Na2S2O4, sulfinatodehalogenation reaction has been widely
used in the organofluorine chemistry [6]. Importantly, this reaction
was extended to cleavage of inert C–Cl bonds in perfluoroalkyl
chlorides and even that in HCFC-133a when DMSO was used as the
solvent instead of MeCN/H2O [7]. As expected, HCFC-123 did react
with Na2S2O4 and resulted in the formation of corresponding
sodium sulfinate (19F NMR: ꢀ66 ppm, 88% yield) and small amount
of HCFC-133a (19F NMR:ꢀ71 ppm) at 60 8C within 3 h in DMSO. It
should be mentioned that no reaction took place when MeCN/H2O
was served as the solvent.
reactions because of low-lying energy of C–Cl
example, as shown in Scheme 1, reactions of CF3CH2Cl (HCFC-
133a) with thiolates resulting in RS-CH2CF3 were proved as SRN
s* orbitals [3]. For
1
processes [4]. In contrast to SRN1 process, it was reported by Jiang
that CF3CCl3 (CFC-113a) reacted with thiolates and gave rise to
RSCCl2CF3, which were rationalized in terms of halophilic
reactions [5].
*
Corresponding author. Tel.: +86 021 54925196; fax: +86 021 54925196.
E-mail addresses: chenqy@sioc.ac.cn, chenqy@mail.sioc.ac.cn (Q.-Y. Chen).
http://dx.doi.org/10.1016/j.jfluchem.2014.10.006
0022-1139/ß 2014 Elsevier B.V. All rights reserved.

2
X.-J. Tang, Q.-Y. Chen / Journal of Fluorine Chemistry 169 (2015) 1–5
Table 2
The SRN1 reaction of HCFC-123 with RSH.
Entry
RSH
p-Me-PhSH(2a)
Base
3
Yielda
Scheme 1. Reactions of CFC-113a, HCFC-133a and HCFC-123 with thiolates.
1
Et3N
3a
3b
3c
3d
3e
3f
54%
27%
57%
25%
24%
58%
2
3
4
5
6
p-Cl-PhSH (2b)
Et3N
p-Isopropyl-PhSH (2c)
Pyridin-2-yl–SH (2d)
Furan-2-yl-CH2SH (2e)
MeOOCCH2CH2SH (2f)
Et3N
Et3N
K2CO3
NaHCO3
a
RSH:HCFC-123= 1:1.5, [RSH] = 0.5 M, the yield was determined by 19F NMR
using PhCF3 as the internal standard.
Scheme 2. Reaction of HCFC-123 with Na2S2O4.
In principle, nucleophiles themselves can transfer one electron
Furthermore, no product 3a was detectedwhen MeCN was served as
the solvent (Entry 6). The yield of 3a decreased if reaction
temperature was reduced to 50 8C (Entry 7). However, any
appreciable yield of the desired product was not observed when
reaction temperature increased to 80 8C (Entry 8). The amount of
Na2S2O4 plays an important role on the yield of 3a. The yield
decreased significantly to 38% if the amount of Na2S2O4 was reduced
to 9% (Entry 9). In contrast, the yield of 3a was not improved when
the amount of Na2S2O4 increased to 1 equivalent and more sulfinate
salt was detected (Entry 10). In addition, the yield of 3a was just
improved a little when the ratio of HCFC-123 to p-tolSH increased to
2:1 (Entry 11).
With optimized conditions in hand, we extended this reaction
to other RSH. As shown in Table 2, substituted thiophenols bearing
electron-donating groups (Entries 1 and 3) gave better yields than
those bearing electron-withdrawing groups (Entries 2 and 4). As
for alkylthiols, due to their weaker acidity compared with
thiophenols, inorganic base such as NaHCO3 or K2CO3 was more
suitable than Et3N (Entries 5 and 6).
to alkyl or aryl halides and initiate chain cycles in ‘‘thermal’’ SRN
1
reactions [8]. However, in some cases, SRN1 reaction must be
stimulated by UV light [9] or electrodes [10] in order to obtain the
radical intermediate to start a chain cycle. When taking account
into the SRN1 reactions of HCFC-123 with thiolates, the crucial
problem is that thiolates themselves are not effective to donate one
electron to HCFC-123 and generate the 1-chloro-2,2,2-trifluor-
oethyl radical through ‘‘thermal’’ SRN1 reaction. On the contrary,
the 1-chloro-2,2,2-trifluoroethyl radical can be generated under
sulfinatodehalogenation conditions as shown in Scheme 2. In line
with these results, we speculated that the SRN1 reactions of HCFC-
123 with thiolates could be initiated by the 1-chloro-2,2,2-
trifluoroethyl radical resulting from the sulfinatodehalogenation
reaction, which is in accord with the concept of ‘‘Entrainment’’ in
S
RN1 reactions [11].
Selecting p-tolSH as the model substrate, optimizations of
conditions are listed in Table 1. In the presence of catalytic amount
of Na2S2O4, the reaction of p-tolSNa with HCFC-123 did afford the
desired product 3a with 15% (Entry 1). In the case of inorganic base
such as K2CO3 or NaHCO3, product 3a was found with 18% or 33%
yield, respectively (Entries 2 and 3). To our delight, the yield of 3a
was improved to 54% when Et3N was used as the base (Entry 4).
When DMF was used as the solvent in place of DMSO, the yield of 3a
decreased and the rate of the reaction also became slow (Entry 5).
Finally, we turned to check the detailed mechanism of the
reaction. As shown in Scheme 3, when 20% of p-DNB, a single
electron transfer (SET) inhibiter was used as the additive, no
formation of 3a was detected in this reaction. This result
demonstrated that SET process should be involved in the reaction.
On the basis of the above results, the proposed mechanism with
an ‘‘entrainment’’ process is depicted in Scheme 4. In the first step,
Table 1
Optimizations of conditions for the SRN1 reaction of HCFC-123 with 2a.
Entry
t (8C)
Na2S2O4
Solvent
Base
Yielda
Scheme 3. Inhibition reaction of HCFC-123 with 2a by p-DNB.
b
1
2
70
70
70
70
70
70
50
80
70
70
70
18%
18%
18%
18%
18%
18%
18%
18%
9%
DMSO
DMSO
DMSO
DMSO
DMF
–
15%
18%
33%
54%
50%
<1%
36%
53%
38%
52%
55%c
K2CO3
NaHCO3
Et3N
3
4
5
Et3N
6
MeCN
DMSO
DMSO
DMSO
DMSO
DMSO
Et3N
7
Et3N
8
Et3N
9
Et3N
10
11
100%
18%
Et3N
Et3N
Determined by 19F NMR using PhCF3 as internal standard, [p-tolSH] = 0.5 M, p-
a
tolSH:HCFC-123 = 1:1.5, the reaction time was 3 h.
b
No base was added, p-tolSNa was used.
c
p-tolSH:HCFC-123 = 1:2.
Scheme 4. Proposed the mechanism with an ‘‘Entrainment’’ process.

X.-J. Tang, Q.-Y. Chen / Journal of Fluorine Chemistry 169 (2015) 1–5
3
Scheme 6. Reaction of 2a with 3a.
Scheme 5. All products for the reaction of HCFC-123 with 2a.
The SET reaction of HCFC-123 with the SO2 radical-anion gives rise
to 1-chloro-2,2,2-trifluoroethyl radical. This radical couples with
the thiolate and affords the crucial radical-anion intermediate. In
the final step, The SET process of the radical-anion intermediate
with HCFC-123 gives rise to the product 3 and regenerates the 1-
chloro-2,2,2-trifluoroethyl radical.
Scheme 7. Reaction of 3a with Na2S2O4.
To elucidate the poor yield of the product 3, all byproducts were
fully investigated for the reaction of 2a with HCFC-123 in the
presence of catalytic amount of Na2S2O4. As shown in Scheme 5,
according to analysis of 19F NMR spectrum and GC/MS, besides 54%
yield of 3a, other byproducts including 13% yield of p-tolS-Stol-p 4
(M+ = 246), 3% yield of 5 (M+ = 328) and 10% yield of 6 (M+ = 206)
were also detected in the crude reaction solution.
by delocalization. Compared with corresponding radical-anions of
conventional simple alkyl halides, the delocalization effect in the
radical-anion M stabilizes it. Consequently, it is long-lived enough
for M to serve as the electron donor to HCFC-123 and finish radical
chain cycle (path 1 in Scheme 8). The more complicated case is
about fragmentation of the radical-anion M. As a rule, fragmenta-
The observation of the byproduct 4, which is generally
tion of the radical-anion usually occurs in the site of
@* in the
attributed to dimerization of corresponding thiyl radical in SRN
1
frontier orbital. As seen in Fig. 1, both C2s–S3p * and C2s–Cl3p @*
@
reactions, is an intriguing result. According to experimental results
and Scheme 4, formation of the thiyl radical from the thiolate is
impossible because the thiolate cannot act as the electron donor to
HCFC-123. Moreover, as shown in Scheme 6, no reaction of the
thiolate with 3a in the absence of Na2S2O4 indicates that
generation of the thiyl radical from the SET process of 3a should
be excluded.
Eventually, we speculated that the thiyl radical is generated
from decomposition of the radical-anion intermediate. To ratio-
nalize this point, generation of the radical-anion from the SET
reaction of 3a with Na2S2O4 was studied. As shown in Scheme 7,
about 18% yield of 4, 3% yield of 5 and 19% yield of 6 were detected
according to analysis of GC/MS and 19F NMR spectrum in the crude
solutions. This result indicates that formation of the corresponding
thiyl radical from the radical-anion intermediate of 3a is
reasonable.
contribute to the SOMO orbital. In addition, as shown in Fig. 1, NBO
charges distributed on the Cl and C atom are ꢀ0.165 and ꢀ0.548,
respectively.
Taking consideration of these calculated results, it is under-
standable that fragmentation may occur with cleavage of C–S bond
or C–Cl bond in the radical-anion M. On the basis of all the above
results, as shown in Scheme 8, we proposed a comprehensive
reaction process for all the products. In the case of cleavage of C–Cl
bond, the radical-anion M is decomposed to the radical M0, which
abstracts H atom from the solvent and affords byproduct 6 (path2
in Scheme 8). At the same time, the radical-anion also can be
decomposed to thiyl radical by cleavage of C–S bond [13] (path 3 in
Scheme 8). Hence, it is reasonable that dimerization of the thiyl
radical can generate the byproduct 4. Similarly, the byproduct 5 is
attributed to coupling of the thiyl radical with the radical M0. In
addition, it can be predicted that the cleavage of C–S bond can
prevail over that of C–Cl bond because the negative charge on the
carbon atom (ꢀ0.548) is much more than that on the chlorine atom
(ꢀ0.165), and spin density on the S atom (16.2%) is also more than
that on the C atom (7.8%). To our delight, this speculation is in
To better understand the reactivity of the radical-anion
intermediate, the excited radical-anion (ERA) M corresponding
to 3a was examined by DFT calculations [12]. As shown in Fig. 1,
spin density is spread over almost the whole molecule carscation
Fig. 1. Spin density, SOMO orbital and NBO charge of the excited radical-anion calculated on UB3LYP/6-31 + G(d)//UB3LYP/6-31G(d) theory level.

4
X.-J. Tang, Q.-Y. Chen / Journal of Fluorine Chemistry 169 (2015) 1–5
4.8 mmol (0.485 g) Et3N and 6 mmol (0.912 g) HCFC-123 were
added successively. The solution was stirred at 70 8C for 3 h and
the yield was checked by 19FNMR using PhCF3 as the internal
standard. The mixture was poured into 25 mL Et2O, washed with
water and brine, and dried over MgSO4. The product of 3a was
purified by flash chromatography on a silica gel column using PE
as the eluent.
4.1.2.1. (1-Chloro-2,2,2-trifluoroethyl)(p-tolyl)sulfane (3a) [15]. 19F
NMR (282 MHz, CDCl3):
CDCl3):
d
ꢀ72.4 (d, J = 6.6 Hz); 1H NMR (300 Hz,
d
7.50 (d, J = 8 Hz, 2H), 7.20 (d, J = 8 Hz, 2H); 2.37 (s, 3H),
5.21 (q, J = 6.4 Hz, 1H).
4.1.2.2. (1-Chloro-2,2,2-trifluoroethyl)(4-chlorophen-yl)sulfane (3b)
[15]. 19F NMR (282 MHz, CDCl3):
d
ꢀ72.2 (d, J = 8 Hz); 1H NMR
(300 Hz, CDCl3):
d 7.56 (d, J = 8 Hz, 2H), 7.38 (d, J = 8 Hz, 2H); 5.22
(q, J = 6.4 Hz, 1H).
Scheme 8. Proposed reaction process for all byproducts.
4.1.2.3. (1-Chloro-2,2,2-trifluoroethyl)(4-isopropyl-phenyl)sulfane
(3c) [3b]. 19F NMR (282 MHz, CDCl3):
(300 Hz, CDCl3):
d
ꢀ72.5 (d, J = 7 Hz); 1HNMR
d
7.53 (d, J = 8.4 Hz, 2H), 7.25 (d, J = 8.4 Hz, 2H),
accord well with experimental results about yields of byproducts 4,
5 and 6 shown in Scheme 5.
2.37 (s, 3H), 5.22 (q, J = 6.3 Hz, 1H), 2.87–2.97 (m, 1H), 1.25 (d,
J = 7 Hz, 6H).
3. Conclusion
4.1.2.4. 2-((1-Chloro-2,2,2-trifluoroethyl)thio)
pyri-dine
(3d)
[15]. 19F NMR (282 MHz, CDCl3):
d
ꢀ72.5 (d, J = 6 Hz); 1H NMR
In summary, we have reported an interesting example of
‘‘entrainment’’ process in a SRN1 reaction. The SRN1 reaction of
HCFC-123 with thiolates cannot occur directly because the SET
reaction of HCFC-123 with thiolates is unfavored, whereas the
sulfinatodehalogenation of HCFC-123 is favored. With catalytic
amount Na2S2O4, the 1-chloro-2,2,2-trifluoroethyl radical gener-
ated from the sulfinatodehalogenation reaction can couple with
thioates, afford the radical-anion intermediate and initiate the
(300 Hz, CDCl3): d 8.52 (d, J = 5 Hz, 1H), 7.59–7.68 (m, 1H), 7.15–
7.27 (m, 2H), 6.41 (q, J = 7 Hz, 1H).
4.1.2.5. 2-(((1-Chloro-2,2,2-trifluoroethyl)thio)meth-yl)furan
(3e)
[15]. 19F NMR (282MHz, CDCl3):
d
ꢀ72.4 (d, J = 6 Hz); 1H NMR
(300 Hz, CDCl3):
d 7.41–7.42 (m, 1H), 6.31–6.32 (m, 2H), 5.14 (q,
J = 6.3 Hz, 1H), 4.10 (d, J = 14.7 Hz), 4.01 (d, J = 14.7 Hz).
S
RN1 reaction of HCFC-123 with thiolates.
4.1.2.6. Methyl
(3f) [15]. 19F NMR (282 MHz, CDCl3):
NMR (300 Hz, CDCl3): d 5.28 (q, J = 6.5 Hz, 1H), 3.71 (s, 3H), 3.09–
3-((1-chloro-2,2,2-trifluoroethyl)thio)-propanoate
d
ꢀ72.7 (d, J = 8 Hz); 1H
4. Experimental
3.15 (m, 2H), 2.73 (t, J = 7 Hz, 2H).
4.1. General information
Acknowledgements
All reactions were carried out under the N2 atmosphere. DMF,
MeCN and DMSO were distilled prior to use. Reagents were
purchased at the commercial quality and used without further
purification. 1H NMR spectra were recorded at Varian 300 MHz
spectrometer. Chemical shifts are reported in ppm relative to TMS
as the reference. 19F NMR spectra were recorded at Varian 282 MHz
spectrometer without proton decoupling. Chemical shifts are
reported in ppm relative to CFCl3 as the external standard.
We thank the Chinese Academy of Science, the National Natural
Science Foundation of China (21032006) and the 973 Program of
China (2012CBA01200) for the financial support.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jfluchem.2014.
10.006.
4.1.1. Preparation of sodium sulfinate
A 20 mL schlenk was charged with 7.5 mmol (1.305 g) Na2S2O4
and 7.5 mmol (0.630 g) NaHCO3, then filled with N2. 10 mL DMSO,
5 mmol (0.760 g) HCFC-123 were added successively. The solution
was stirred at 60 8C for 3 h. DMSO was removed by vacuum
distillation. The residue was extracted using bointing EtOAc for
several times. After removing EtOAc, sodium sulfinate was
afforded as a white solid.
References
[1] (a) R.E. Banks, J. Fluorine Chem. 67 (1994) 193–203;
(b) J.S. Francisc, M.M. Maricq, Acc. Chem. Res. 29 (1996) 391–397.
[2] (a) J. Hine, J.J. Porter, J. Am. Chem. Soc. 79 (1957) 5493–5496;
(b) B. Soborowskii, Russ. J. Org. Chem. 29 (1959) 1142.
[3] (a) X.J. Tang, Q.Y. Chen, Chem. Sci. 3 (2012) 1694–1697;
(b) X.J. Tang, Q.Y. Chen, Org. Lett. 14 (2012) 6214–6217.
[4] Z.Y. Long, Q.Y. Chen, J. Fluorine Chem. 91 (1998) 95–98.
[5] X.Y. Li, X.K. Jiang, H.Q. Pan, J.S. Hu, W.M. Fu, Pure Appl. Chem. 59 (1987) 1015–
1020.
[6] C.P. Zhang, Q.Y. Chen, Y. Guo, J.C. Xiao, Y.C. Gu, Chem. Soc. Rev. 41 (2012) 4536–
4559;
For a review about Sulfinatodehalogenation reaction, see:
[7] Z.Y. Long, Q.Y. Chen, J. Org. Chem. 64 (1999) 4775–4782.
4.1.1.1. Sodium 1-chloro-2,2,2-trifluoroethanesulfinate(2) [14]. 19F
NMR (282 MHz, DMSO):
DMSO):
d
ꢀ66.0 (d, J = 8 Hz); 1H NMR (300 Hz,
d
4.14 (d, J = 9 Hz, 1H); MS–ESI (M–Na)ꢀ: 181; HRMS–
ESI(M–Na)ꢀ calcd. for C2HClF3O2S: 180.93434, found 180.93428.
´
´
[8] C. Costentin, P. Hapiot, M. Medebielle, J.M. Saveant, J. Am. Chem. Soc. 121 (1999)
4.1.2. Reactions of HCFC-123 with RSH
4451–4460;
A 20 mL schlenk was charged with 0.72 mmol (12 mg) Na2S2O4
and filled with N2, 10 mL DMSO, 4 mmol (0.480 g) p-tolSH,
Actually, ‘‘thermal’’ SRN1 reactions are very common in the field of organofluorine
chemistry, for some accounts on ‘‘thermal’’ SRN1 reactions, see:

X.-J. Tang, Q.-Y. Chen / Journal of Fluorine Chemistry 169 (2015) 1–5
5
[9] J.F. Bunnett, Acc. Chem. Res. 11 (1978) 413–420;
For a review, see:
[12] A.A. Isse, A. Gennaro, C.Y. Lin, J.L. Hodgson, M.L. Coote, T. Guliashvili, J. Am. Chem.
Soc. 133 (2011) 6254–6264;
[10] J.M. Save´ant, Acc. Chem. Res. 13 (1980) 323–329;
For a review, see:
[11] (a) H.P. Albrecht, G.H. Jones, J.G. Moffatt, J. Am. Chem. Soc. 92 (1970)
5513–5514;
For mechanism study on fraction of radical-anion in the alkyl halides, see:
[13] S. Telitel, A.L. Vallet, S. Schweizer, B. Delpech, N. Blanchard, F. Morlet-Savary, B.
Graff, D.P. Curran, M. Robert, E. Lacoˆte, J. Laleve´e, J. Am. Chem. Soc. 135 (2013)
16938–16947;
(b) W.R. Bowman, H. Heaney, P.H.G. Smith, Tetrahedron Lett. 23 (1982)
5093–5096;
(c) K. Uneyama, K. Kitagawa, Tetrahedron Lett. 32 (1991) 3385–3386;
For selected reports on ‘‘Entrainment’’, see:
For one recent report on fragment of radical-ion into thiyl radical, see:
[14] W. Dmowski, K. Piasecka-Maciejewska, J. Fluorine Chem. 126 (2005) 877–882.
[15] Y. Pustovit, A. Alexeenko, S. Trofymchuk, O. Lukin, A.A. Tolmachev, Synthesis
(2010) 1159–1165.