Journal of the American Chemical Society
Communication
difluoromethyl reactions with aromatic rings; therefore, a
transition metal complex with bulky ligands may enable para-
selective C−H difluoromethylations.
was not obtained using iron chloride or Fe(OTs)3 as the
catalyst (Table 1, entries 12, 13). The control experiment
showed that Fe(TPP)Cl is indispensable for this trans-
formation (Table 1, entry 14).
With these conditions in mind, a sterically hindered
porphyrin was selected as the ligand and ruthenium as the
catalyst precursor. Initially, 9-benzyl-6-phenylpurine (1a) was
treated with bromodifluoroacetate (2) in the presence of
[RuCl2(p-cymene)]2 (5 mol %), L1 (10 mol %), and K2CO3
(3 equiv) in DCE at 140 °C for 24 h. Mixed meta- and para-
difluoromethylated products were obtained in 51% yield
(para:meta = 2.7:1; Table 1, entry 1). When PivOH was
a
Scheme 1. Scope of N-Substituted Purines
a
Table 1. Optimized Reaction Conditions
(3a+3a′) yield (3a/3a′) (p/
a
b
c
entry
1
catalyst (x mol %)
ligand
(%)
51
m)
[RuCl2(p-cymene)]2
(5)
[RuCl2(p-cymene)]2
(5)
[RuCl2(p-cymene)]2
(5)
[RuCl2(p-cymene)]2
(5)
L1
2.7:1
1:1.5
1:1.5
1:3.9
d
2
PivOH
L2
22
38
42
d
3
a
4
5
L3
Reactions were performed in 0.2 mmol scale with 2 (1.2 mmol) in a
b
sealed tube; isolated yields. K2CO3 (2 equiv) was added for 48 h.
[RuCl2(p-cymene)]2
(5)
L4
trace
With the optimized reaction conditions established, we next
explored the substrate scope of this reaction. Various N-
substituted purines were subjected to the standard reaction
conditions; phenemyl, 4-methylbenzyl, methyl, isopropyl, and
n-pentyl substrates were all well tolerated and afforded para-
selective difluoromethylated products in good to excellent
yields (3b−h). More appealingly, we found that highly reactive
nucleosides were all well-tolerated, even those bearing sensitive
protecting groups, which led to the para-difluoromethylated
products in moderate to good yields (3i−k). These results are
complementary to the ruthenium-catalyzed meta-difluoroalky-
lation reaction reported by the Ackermann12 group and also
have potential use for antiviral or anticancer drug screenings.
We expanded this strategy to substrates with strongly
coordinating heteroaromatic compounds. For example, various
5-position-substituted 2-phenylpyridines were well tolerated
and yielded para-difluoromethylated products in moderate to
good yields (5a and 5g). Functional groups (methyl, methoxy,
fluoro, chloro, and cyano) were fully compatible with this
transformation (5b−f). Several 2-arylpyrimidines provided
para-selective difluoromethylated products in moderate to
good yields, highlighting the importance of this synthetic
method (5i−n). Interestingly, substrate 4h yielded product 5h
in good yield with high para selectivity, while 2-fluoro-6-
phenylpyridine and 2-methoxy-6-phenylpyridine all gave
difluoromethylated products with poor site selectivity.
Although the reason is unclear, it might be due to a weak
interaction between the chlorine atom in 4h with the iron
6
7
8
9
10
11
12
13
14
hemin (5)
Fe(TPP)Cl (2.5)
CoTPP (5)
NiTPP (5)
CuTPP (5)
Mn(TPP)Cl (2.5)
FeCl3 (10)
Fe(OTs)3 (10)
41
78
25:1
20:1
trace
trace
trace
50
trace
trace
nr
17:1
a
Reactions were performed with 1a (0.2 mmol), 2 (1.2 mmol, 160
uL), cat. (x mol %), and K2CO3 (0.6 mmol, 3 equiv) in DCE (0.5
mL), Ar, 140 °C, 24 h. Yields of product after silica gel
chromatography. The ratio of selectivity determined by H NMR
b
c
1
d
spectroscopy. K2CO3 (0.4 mmol, 2 equiv) was added at 120 °C.
used instead of L1, the meta-difluoromethylated product 3a′
was obtained as the major product (Table 1, entry 2). Various
porphyrins were tested; however, neither the yield nor
selectivity was improved (Table 1, entries 4, 5). To our great
delight, when the hemin was used instead of the ruthenium
catalyst, C−H difluoromethylated product 3a was obtained in
41% yield with high para selectivity (Table 1, entry 6).
Encouraged by this result, several metal porphyrin complexes,
such as Fe(TPP)Cl, CoTPP, NiTPP, CuTPP, and Mn(TPP)-
Cl, were examined (Table 1, entries 7−11). Those results
clearly showed that Fe(TPP)Cl was the most effective catalyst
and afforded product 3a in a 78% yield with high para-
selectivity (Table 1, entry 7). The difluoromethylated product
B
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX