Organic Letters
Letter
molecule, such as ezetimibe derivative (1aa, 46% yield).
However, only a trace amount of desired product was observed
when tertiary alkyl trifluoroborate was applied as a substrate. In
addition, gram-scale reaction was performed under the
standard condition, and an 80% isolated yield of 3d was
gained. Furthermore, when the equivalents of silver salt and
ligand were both reduced to 10 mmol % in this transformation,
the yields decreased slightly (3e, 3f, 3h, 3p, 3q, 3t).
Scheme 3. Mechanistic Experiments and Potential Reaction
Pathway
In order to study the mechanism of this transformation, we
performed several experiments. In the first, 4.0 equiv of
TEMPO was added as the radical trapper, and the TEMPO
adduct 4 was observed (Scheme 3a). Furthermore, a radical
probe 1bb was designed and applied in the reaction, and the
ring-opening product 5 was achieved in 20% yield (Scheme
3b). These observations indicated that alkyl radicals were
generated from alkyl trifluoroborates in this reaction. In
addition, no trifluoromethoxylated product was observed when
2−3 equiv of AgF2 was used instead of AgOTf and Selectfluor,
which suggested that a higher valence state of Ag species might
be involved in this reaction. Based on the previous reports and
above results,60 a possible mechanism was proposed and
described in Scheme 3c. The Ag(I) species is oxidated by
Selectfluor to afford Ag(III)-F species,61−65 then this silver
species goes through the ligand exchange to form FAg(III)-
OCF3. The later reacts with alkyl trifluoroborates to give alkyl
radicals and Ag(II)OCF3 via single-electron oxidation.
Subsequently, the desired product was obtained by grabbing
OCF3 radical from Ag(II)OCF3,60,61,66−68 and Ag(I) was
regenerated at the same time. Further invetigations about
mechanism are needed to better understand the above
reaction.
In conclusion, we have reported, for the first time, about the
silver-catalyzed trifluoromethoxylation of alkyl trifluoroborates
with TFMS as the trifluoromethoxylation reagent. The
highlight of this method is its good tolerance for various
functional groups, compared with previous methods. Fur-
thermore, complex small molecules can also be successfully
converted to the corresponding trifluoromethoxylated com-
pounds in the late stage. In addition, preliminary mechanistic
research was investigated and suggested that the trans-
formation might undergo a free radical pathway. Our next
challenge is to achieve asymmetric trifluoromethoxylation of
sp3-hybridized carbon centers.
Me4-1,10-Phen, 3.0 equiv TFMS (2a), 3.0 equiv KF, and 3.0
equiv 18-Crown-6 in anisole under N2 atmosphere at 30 °C.
Moreover, if any of these reagents was missing, no desired
product would be obtained.
We then investigated the substrate scope for this silver-
catalyzed trifluoromethoxylation of alkyl trifluoroborates
(Scheme 2). Delightly, a broad scope of alkyl trifluoroborates
reacted smoothly and achieved 32% to 91% yields (3a−3z).
Substrates with aryl rings bearing electron-rich and electron-
deficient substituents reacted smoothly. Variously important
functional groups, such as alkene, hydroxyl, cyano, nitro,
amide, ketone, aldehyde, bromo, ether and ester moieties, were
well-tolerated, which highlight the compatibility of this
method. Notably, substrates with heteroaromatic rings were
also successfully processed to give the corresponding products
(3p, 3q). What is noteworthy is that the chemoselective
trifluoromethoxylation of alkyl trifluoroborates in the presence
of alkyl chloride or alkyl p-toluenesulfonate was observed. For
example, the alkyl trifluoroborate group in substrates 1g, 1t
was selectively converted to the corresponding “OCF3” group,
while other functional groups, such as alkyl chloride, or OTs
remained intact (3g, 3t), which could be used as good leaving
groups for further synthetic manipulation. The 40% desired
product 3w was observed, along with 20% ditrifluoromethoxy-
lated byproduct, which is generated by nucleophilic trifluor-
omethoxylation of alkyl iodide. Moreover, aryl azide was also
tolerated (3k), and the corresponding product could be further
applied in click reaction. Secondary alkyl trifluoroborates were
also tested, but lower yields were gained (3y, 3z). The main
side product observed was β-H elimination product.
Optimization was tried to improve the yield, such as changing
ligands, solvents, reaction temperature, and additives, but
failed. To our delight, this method could also be applied to the
trifluoromethoxylation of a structurally more-complicated
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
Experimental procedures and characterization of all new
1
compounds including H, 13C, and 19F NMR spectra
AUTHOR INFORMATION
■
Corresponding Author
Pingping Tang − State Key Laboratory and Institute of
Elemento-Organic Chemistry, College of Chemistry, Nankai
C
Org. Lett. XXXX, XXX, XXX−XXX