Radical Difluoromethylation of Thiols with (Difluoromethyl)triphenylphosphonium Bromide

Article abstract of DOI:10.1021/acs.orglett.7b02109

A method for facile difluoromethylation of various thiols using (difluoromethyl)triphenylphosphonium bromide under mild reaction conditions is presented. The transformation proceeds in the absence of any transition metal using a bench-stable and readily accessible phosphonium salt. Deuterium labeling experiments and cyclic voltammetry measurements reveal that the difluoromethylation occurs via a S_RN1-type mechanism. Substrate scope is broad, and various functional groups are tolerated (OH, NH₂, amide, ester).

Full text of DOI:10.1021/acs.orglett.7b02109

Letter
pubs.acs.org/OrgLett
Radical Difluoromethylation of Thiols with
(Difluoromethyl)triphenylphosphonium Bromide
Niklas B. Heine and Armido Studer*
Organisch Chemisches Institut, Westfalische Wilhelms-Universitat Munster, Corrensstraße 40, 48149 Munster, Germany
̈
̈
̈
̈
S
* Supporting Information
ABSTRACT: A method for facile difluoromethylation of
various thiols using (difluoromethyl)triphenylphosphonium
bromide under mild reaction conditions is presented. The
transformation proceeds in the absence of any transition metal
using a bench-stable and readily accessible phosphonium salt.
Deuterium labeling experiments and cyclic voltammetry
measurements reveal that the difluoromethylation occurs via
a SRN1-type mechanism. Substrate scope is broad, and various
functional groups are tolerated (OH, NH2, amide, ester).
he incorporation of tri- and difluoromethyl groups into
Light-initiated SRN1-type reactions of aryl thiols have been
investigated extensively over the past decades.13 The aromatic
thiolate anion is known to be a very efficient SRN1 nucleophile for
which mainly halobenzene derivatives have been used as
coupling partners.14 Wakselman and co-workers investigated
the halo-difluoromethylation of thiolates with CF2Br2 and
CF2Cl2 and found that these reactions can proceed via carbenes
but also via radical intermediates.15
T
organic molecules has drawn significant interest over recent
years1 because fluorine-containing compounds are of high
importance in various areas such as medicinal chemistry or
materials science.2 Compared to the trifluoromethylation where
various reagents such as the Togni reagent,1b,3 Umemoto
reagent,4 Langlois reagent,5 Ruppert−Prakash reagent6 and
fluoroalkanesulfinates7 have been introduced, the difluorome-
thylation is clearly underexplored.1e,7b,8 Along these lines, two
bench-stable phosphonium salts 1 and 2 have recently been
applied to the difluoromethylation (Scheme 1).1a These
Based on these results, we were encouraged to use 1, which can
be prepared easily on a multigram scale from commercially
available ethyl bromodifluoroacetate,12a,16 as a reagent for radical
difluoromethylation of thiols under basic conditions (Scheme
1c).
We first investigated solvent effects on the difluoromethylation
of thiophenol (3a) with phosphonium salt 1 (2 equiv) and
KOtBu (5 equiv) as a base using light initiation (365 nm) (Table
1). Reactions in acetonitrile, THF, DMSO, or dioxane provided
the desired compound 4a in moderate yields (around 30%), and
only traces of the target thioether were identified with
dichloromethane or toluene as solvent (Table 1, entries 1−6).
Yield significantly improved upon switching to DMF (51%,
Table 1, entry 7). With the ideal solvent identified, we next
screened different bases, and yield further increased to 61% using
sodium hydride (Table 1, entry 8). DBU and Cs2CO3 provided
worse results (Table 1, entries 9 and 10). This observation is a
Scheme 1. Difluoromethylation of Thiols Using Carbene
Chemistry (a), Transition-Metal-Catalyzed Process (b), and
Our Radical Approach (c)
first hint for a SRN1-type mechanism because in many SRN
1
reactions inorganic bases, such as NaH and KOtBu, are used.17
The poor reactivity with Cs2CO3 can be explained by a known
nucleophilic attack of the carbonate to the phosphonium salt,
which probably takes place as a side reaction.18
We found that the amount of base affects the reaction outcome
(Table 1, entries 11−15), and the best result (81% yield) was
obtained using 2 equiv of NaH. Decreasing the amount of
reactions proceed via the CF2-carbene that is readily generated
from these phosphonium salts, and different functional groups
such as aliphatic thiols,9 alcohols,10 benzylic halides,11 or
carboxylic acids9 can be difluoromethylated with such salts or
related reagents. Surprisingly, only a few reports on the use of 1
as a C-radical precursor have appeared to date.12
Received: July 11, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b02109
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 1. Reaction Optimization Using Thiophenol as
Substrate
Scheme 2. Scope of Aryl Thiols
a
entry
base (equiv)
solvent
1 (equiv)
yield (%)
1
KOtBu (5)
KOtBu (5)
KOtBu (5)
KOtBu (5)
KOtBu (5)
KOtBu (5)
KOtBu (5)
NaH (5)
MeCN
THF
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
1
1.5
2
2
2
33
2
31
3
DMSO
dioxane
toluene
CH2Cl2
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
31
4
27
5
traces
traces
51
6
7
8
61
9
Cs2CO3 (5)
DBU (5)
NaH (1)
27
10
11
12
13
14
15
16
17
13
3
NaH (1)
5
NaH (1.5)
NaH (2)
NaH (3)
18
d
81 (70)
66
13
61
78
75
73
NaH (2)
NaH (2)
b
18
NaH (2)
a
All reactions were carried out with thiols 3b−m (0.2 mmol), NaH
(0.4 mmol) and 1 (0.4 mmol) in DMF (1 mL) under irradiation for
c
19
NaH (2)
bc
,
20
NaH (2)
b
16 h. All yields are isolated yields. Reaction was carried out with 0.8
mmol of NaH and 0.8 mmol of 1. Reaction in parentheses was carried
a
c
All reactions were carried out with 3a (0.2 mmol) under irradiation in
the given solvent (1 mL) for 16 h. Benzotrifluoride was used as an
out with thiol 3i (1 mmol), NaH (2 mmol), and 1 (2 mmol) in DMF
(5 mL).
internal standard to determine the yield by 19F NMR spectroscopy.
b
c
Reaction was carried out in the dark. Reaction time was decreased to
d
4 h. Isolated yield.
Scheme 3. Difluoromethylation of Heteroaryl and Benzylic
a
Thiols
phosphonium salt 1 to 1.5 and 1 equiv led to worse results (Table
1, entries 11, 16, and 17). A slightly decreased yield was also
noted in the absence of light or by shortening of the reaction time
(Table 1, entries 17−19).
With optimized conditions in hand (Table 1, entry 14), the
scope of the thiol difluoromethylation was studied. Various aryl
thiols were tested to explore arene substituent effects (Scheme
2). Compared to the parent thiophenol, electron-donating alkyl
groups at the ortho or para position did not influence the reaction
outcome, and the difluoromethylated products 4b−d were
isolated in 66−71% yield. However, with the ortho-methyl/
ortho′-tert-butyl congener 3e, the yield slightly decreased (4e).
Double difluoromethylation of 3f yielded 4f in 75%. High yields
were obtained for 2-naphthalenethiol (3i) and methyl benzoate
3g to give 4i and 4g in 86 and 88% yields, respectively. Reaction
with 3i was repeated at a larger scale, and the yield of 4i further
improved to 91% (0.19 g prepared). Importantly, in contrast to
carbene-mediated difluoromethylations, where free OH and NH
groups lead to side reactions (see below), the process introduced
herein is highly chemoselective. Hence, thiophenols 3h and 3j−
m bearing additional amide, amine, or phenol XH moieties
underwent SH-difluoromethylation with perfect chemoselectiv-
ity. The desired products 4h and 4j−m were isolated in moderate
to good yields. These results convincingly demonstrate the
additional value of the radical-mediated difluoromethylation as
compared to the difluoromethyl carbene processes.
a
All reactions were carried out with thiols 3n−t (0.2 mmol), NaH (0.4
mmol), and 1 (0.4 mmol) in DMF (1 mL) under irradiation for 16 h.
b
Product contains 13% of a side product (based on 1H NMR),
probably difluoromethylation at the nitrogen atom. Yields provided
correspond to isolated yields. Benzeneselenol was used instead of
thiol. N-Difluoromethylation instead of S-difluoromethylation
occurred.
c
d
thiazole 3r (see 4n, 4s, and 4p). Notably, for 2-mercaptopyridine
(3n), a side product with the same mass as 4n (probably
difluoromethylation at the nitrogen atom) was formed in 13%
(see the Supporting Information). Benzylic thiols 3p and 3q
were transformed successfully to the target thioethers 4p and 4q,
showing that our method is not restricted to aromatic thiols.
We next tested the difluoromethylation of various heteroaryl
thiols, benzylic thiols, and benzeneselenol (Scheme 3). High
yields were obtained with pyridine 3n, pyrimidine 3s, and
B
DOI: 10.1021/acs.orglett.7b02109
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
However, aliphatic thiols could not be difluoromethylated under
the optimized conditions. Benzeneselenol also engaged in this
transformation, and 4o was isolated in 61% yield. In contrast to
other examples where exclusive S-difluoromethylation was
achieved, reaction with benzo[d]oxazole-2-thiol (3t) occurred
with complete chemoselectivity at the nitrogen atom to provide
oxazolethione 4t in an excellent yield (83%). The reason for the
switch in chemoselectivity is currently not understood.
Scheme 5. Proposed Mechanism
To confirm our hypothesis that the difluoromethylation
proceeds via a radical-type process, we ran additional mechanistic
experiments (Scheme 4). Reaction of thiophenol with the
a
Scheme 4. Mechanistic Studies
and electron transfer from the thiolate is only important in the
initiation step.
Based on these studies, we propose the following mechanism
for the difluoromethylation reaction (Scheme 5). Initiation of the
chain reaction occurs by electron transfer from the thiolate anion
to the phosphonium salt 1 upon irradiation21 to generate the
difluoromethyl radical A. Direct electron transfer from the
thiolate to the phosphonium salt is also feasible as an initiation
step because, albeit less efficiently, the chain is also initiated in the
dark. Radical A then reacts with the thiolate anion to form the
radical anion B in a propagation step. Single electron transfer
from B to the phosphonium salt 1 affords the corresponding
difluoromethylation product along with the radical A, NaBr, and
PPh3. To our knowledge, this is the first report on an SRN1-type
reaction using a phosphonium salt.
a
All reactions were carried out with thiol (0.2 mmol), NaH (0.4
mmol), and 6 or 7 (0.4 mmol) in DMF (1 mL) under irradiation for
b
1
16 h. D/H ratio of the starting material was 97:3 as analyzed by H
c
and 19F NMR spectroscopy. D/H ratio of the isolated product as
analyzed by H and 19F NMR spectroscopy was 90:10 for 2 equiv of
1
d
NaH (0.4 mmol) and 95:5 for 1.5 equiv of NaH (0.3 mmol). Yield
determined using benzotrifluoride as an internal standard by 19F NMR
spectroscopy: S (8%), O (2%), S and O (2%) (see the Supporting
Information, section 7).
Finally, we used reagent 1 for the difluoromethylation of both
isocyanide 8 and benzofuran 10 to show the versatility of the
phosphonium salt in radical transformations (Scheme 6). Note
deuterated phosphonium salt 6 (D/H = 97:3) gave 5a in 68%
yield with 90% deuterium content, which was determined by
both 1H and 19F NMR spectroscopy. We assumed that some loss
of deuterium content might have been caused after the
difluoromethylation by a deprotonation/reprotonation se-
quence. Indeed, decreasing the amount of base from 2 to 1.5
equiv afforded 6 with 95% D content on the basis of 1H and 19F
NMR spectra. Based on these results, it is highly unlikely that
reactions proceed via the difluorocarbene as an intermediate.
Furthermore, the reaction of 3l with the typical carbene
precursor 7 under standard conditions provided a mixture of
products, including the S, O, and double trifluoromethylation
product in low overall yield. This result showed that
difluorocarbene reacts with 3l non-chemoselectively, further
supporting that the transformation with 1 is radical in nature.
Difluorocarbene can be generated from 7 by phenolate or
thiolate attack at the Br atom in 7.
a
Scheme 6. Difluoromethylation at Carbon Atoms
a
All reactions were carried out with 8 or 10 (0.2 mmol), Na2HPO4
(0.4 mmol), and 1 (0.4 mmol) in DMF (2 mL).
We also measured the reduction potential of 1 by cyclic
voltammetry (see the Supporting Information). Salt 1 is reduced
at a potential of −1.67 V against the ferrocene/ferrocenium ion
pair, and this shows the ability of 1 to act as a weak single electron
oxidant. As expected, single electron reduction of 1 is not
reversible. For comparison, the oxidation potential of sodium
thiophenolate has been previously investigated by Persson and
Nygard and lies between −0.4 and −0.5 V (vs SCE) depending
on the concentration.19 However, more important for the radical
chain is the electron transfer from radical anion B to A (Scheme
5). The potential of Ar−S−Ph/Ar−S−Ph•− has been inves-
tigated by Saveant et al. and lies between −1.5 and −2.0 V (vs
Ag/Ag+) depending on the substituents.20 Therefore, it is likely
that the radical anion B acts as a reductant in the chain reaction,
that the phenanthridine synthesis starting from ortho-isocyano-
biphenyls has been intensively investigated recently.1d In this
case, the photoredox catalyst fac-Ir(ppy)3 was necessary to
mediate the cascade, and the difluoromethylated phenanthridine
9 was obtained in 67% yield. However, with carbene precursor 7
under identical conditions, product 9 was not formed. In
addition, benzofuran derivative 10 was successfully difluorome-
thylated with 1 in moderate yield (see 11) under these
conditions.
In conclusion, we have presented a facile, transition-metal-free
method for the difluoromethylation of various thiols under mild
conditions. The method shows high functional group tolerance,
C
DOI: 10.1021/acs.orglett.7b02109
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(8) (a) Hu, J.; Zhang, W.; Wang, F. Chem. Commun. 2009, 7465−7478.
(b) Zhang, W.; Wang, F.; Hu, J. Org. Lett. 2009, 11, 2109−2112.
(c) Zafrani, Y.; Sod-Moriah, G.; Segall, Y. Tetrahedron 2009, 65, 5278−
5283. (d) Wang, F.; Huang, W.; Hu, J. Chin. J. Chem. 2011, 29, 2717−
2721. (e) Yang, J.; Jiang, M.; Jin, Y.; Yang, H.; Fu, H. Org. Lett. 2017, 19,
2758−2761. (f) Yerien, D. E.; Barata-Vallejo, S.; Postigo, A. Chem. - Eur.
J. 2017, DOI: 10.1002/chem.201702311. For electrophilic pathways,
see: (g) Zhu, D.; Gu, Y.; Lu, L.; Shen, Q. J. Am. Chem. Soc. 2015, 137,
10547−10553. (h) Huang, Z.; Matsubara, O.; Jia, S.; Tokunaga, E.;
Shibata, N. Org. Lett. 2017, 19, 934−937. (i) Prakash, G. K. S.;
Krishnamoorthy, S.; Kar, S.; Olah, G. A. J. Fluorine Chem. 2015, 180,
186−191. (j) Prakash, G. K. S.; Zhang, Z.; Wang, F.; Ni, C.; Olah, G. A. J.
Fluorine Chem. 2011, 132, 792−798.
and the phosphonium reagent is readily prepared in large scale.
Mechanistic studies reveal that the process is radical in nature.
The introduced method is complementary to reported thiol
difluoromethylations that proceed via the difluorocarbene
intermediate. Clearly, the radical process offers advantages over
the carbene routes in terms of functional group tolerance. This
letter shows that phosphonium salts can engage in SRN1-type
reactions, and considering the diverse SRN1 chemistry, such salts
should have a future as precursors in radical chemistry.
ASSOCIATED CONTENT
* Supporting Information
■
S
(9) Deng, X.-Y.; Lin, J.-H.; Zheng, J.; Xiao, J.-C. Chem. Commun. 2015,
51, 8805−8808.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b02109.
(10) Hua, M.-Q.; Wang, W.; Liu, W.-H.; Wang, T.; Zhang, Q.; Huang,
Y.; Zhu, W.-H. J. Fluorine Chem. 2016, 181, 22−29.
(11) Zheng, J.; Wang, L.; Lin, J.-H.; Xiao, J.-C.; Liang, S. H. Angew.
Chem., Int. Ed. 2015, 54, 13236−13240; Angew. Chem. 2015, 127,
13434−13438.
1
Experimental procedures; characterization data; H, 13C,
19F, and 31P spectra; cyclovoltammetry measurements
(PDF)
(12) (a) Lin, Q.-Y.; Ran, Y.; Xu, X.-H.; Qing, F.-L. Org. Lett. 2016, 18,
2419−2422. (b) Lin, Q.-Y.; Xu, X.-H.; Zhang, K.; Qing, F.-L. Angew.
Chem., Int. Ed. 2016, 55, 1479−1483; Angew. Chem. 2016, 128, 1501−
1505. (c) Ran, Y.; Lin, Q.-Y.; Xu, X.-H.; Qing, F.-L. J. Org. Chem. 2016,
81, 7001−7007. (d) Orsi, D. L.; Easley, B. J.; Lick, A. M.; Altman, R. A.
Org. Lett. 2017, 19, 1570−1573.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: studer@uni-muenster.de.
ORCID
(13) (a) Bardagi, J. I.; Rossi, R. A. Encyclopedia of Radicals in Chemistry,
Biology and Materials; Chatgilialoglu, C., Studer, A., Eds.; Wiley-Verlag:
Chichester, UK, 2012; pp 333−364. (b) Rossi, R. A.; Pierini, A. B.;
Penenory, A. B. Chem. Rev. 2003, 103, 71−168. (c) Zhang, N.; Samanta,
S. R.; Rosen, B. M.; Percec, V. Chem. Rev. 2014, 114, 5848−5958.
(14) For recent examples on thiol SRN1 chemistry, see: (a) Heine, N.
B.; Studer, A. Macromol. Rapid Commun. 2016, 37, 1494−1498.
(b) Janhsen, B.; Daniliuc, C. G.; Studer, A. Chem. Sci. 2017, 8, 3547−
Armido Studer: 0000-0002-1706-513X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Deutsche Forschungsgemeinschaft (DFG) for
financial support, and Dr. Dirk Leifert (WWU Munster) for
providing isocyanide 8.
́ ̀
3553. (c) For a review on thiyl radical chemistry, see: Denes, F.;
̈
Pichowicz, M.; Povie, G.; Renaud, P. Chem. Rev. 2014, 114, 2587−2693.
(15) (a) Rico, I.; Cantacuzene, D.; Wakselman, C. J. Org. Chem. 1983,
48, 1979−1982. (b) Wakselman, C.; Tordeux, M. J. Org. Chem. 1985, 50,
4047−4051.
REFERENCES
■
(16) Zheng, J.; Cai, J.; Lin, J.-H.; Guo, Y.; Xiao, J.-C. Chem. Commun.
2013, 49, 7513−7515.
(17) Wolfe, J. F.; Carver, D. R. Org. Prep. Proced. Int. 1978, 10, 225−
253.
(1) For some reviews, see: (a) Zhang, C. Adv. Synth. Catal. 2017, 359,
372−383. (b) Charpentier, J.; Fruh, N.; Togni, A. Chem. Rev. 2015, 115,
̈
650−682. (c) Barata-Vallejo, S.; Lantano, B.; Postigo, A. Chem. - Eur. J.
̃
2014, 20, 16806−16829. (d) Zhang, B.; Studer, A. Chem. Soc. Rev. 2015,
44, 3505−3521. (e) Lu, Y.; Liu, C.; Chen, Q.-Y. Curr. Org. Chem. 2015,
19, 1638−1650. (f) Hu, J. J. Fluorine Chem. 2009, 130, 1130−1139.
(g) Zhang, C.-P.; Chen, Q.-Y.; Guo, Y.; Xiao, J.-C.; Gu, Y.-C. Coord.
Chem. Rev. 2014, 261, 28−72. (h) Studer, A. Angew. Chem., Int. Ed. 2012,
51, 8950−8958; Angew. Chem. 2012, 124, 9082−9090.
(18) Deng, Z.; Lin, J.-H.; Xiao, J.-C. Nat. Commun. 2016, 7, 10337.
(19) Persson, B.; Nygard, B. J. Electroanal. Chem. Interfacial
Electrochem. 1974, 56, 373−383.
(20) Amatore, C.; Oturan, M. A.; Pinson, J.; Saveant, J. M.; Thiebault,
A. J. Am. Chem. Soc. 1985, 107, 3451−3459.
(21) Studer, A.; Curran, D. P. Nat. Chem. 2014, 6, 765−773.
(2) For some reviews, see: (a) Hagmann, W. K. J. Med. Chem. 2008, 51,
4359−4369. (b) Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V.
Chem. Soc. Rev. 2008, 37, 320−330. (c) Wang, J.; San
́
chez-Rosello,
́
M.;
Acena, J. L.; del Pozo, C.; Sorochinsky, A. E.; Fustero, S.; Soloshonok, V.
̃
A.; Liu, H. Chem. Rev. 2014, 114, 2432−2506. (d) Berger, R.; Resnati,
G.; Metrangolo, P.; Weber, E.; Hulliger, J. Chem. Soc. Rev. 2011, 40,
3496−3508.
(3) Eisenberger, P.; Gischig, S.; Togni, A. Chem. - Eur. J. 2006, 12,
2579−2586.
(4) (a) Umemoto, T.; Ishihara, S. J. Am. Chem. Soc. 1993, 115, 2156−
2164. (b) Umemoto, T. Chem. Rev. 1996, 96, 1757−1778.
(5) (a) Tordeux, M.; Langlois, B.; Wakselman, C. J. Org. Chem. 1989,
54, 2452−2453. (b) Zhang, C. Adv. Synth. Catal. 2014, 356, 2895−2906.
(6) (a) Ruppert, I.; Schlich, K.; Volbach, W. Tetrahedron Lett. 1984, 25,
2195−2198. (b) Prakash, G. K. S.; Yudin, A. K. Chem. Rev. 1997, 97,
757−786.
(7) (a) Ma, J. J.; Yi, W.-B.; Lu, G.-P.; Cai, C. Catal. Sci. Technol. 2016, 6,
417−421. (b) Ma, J.; Liu, Q.; Lu, G.; Yi, W. J. Fluorine Chem. 2017, 193,
113−117. (c) Zhu, D.; Shao, X.; Hong, X.; Lu, L.; Shen, Q. Angew.
Chem., Int. Ed. 2016, 55, 15807−15811; Angew. Chem. 2016, 128,
16039−16043.
D
DOI: 10.1021/acs.orglett.7b02109
Org. Lett. XXXX, XXX, XXX−XXX