Journal of the American Chemical Society
Communication
a
a b
,
Table 1. Optimization of Reaction Conditions
Table 2. Synthesis of B(3)-amino-o-carboranes
entry
ligand (mol %)
K2CO3 (equiv)
3a (%)
1
2
3
4
P(o-tol)3 (20)
P(Cy)3 (20)
XPhos (20)
P(tBu)3 (20)
P(Cy)3 (20)
P(Cy)3 (20)
P(Cy)3 (20)
P(Cy)3 (20)
P(Cy)3 (20)
P(Cy)3 (20)
P(Cy)3 (30)
P(Cy)3 (30)
P(Cy)3 (40)
P(Cy)3 (30)
P(Cy)3 (30)
P(Cy)3 (30)
P(Cy)3 (15)
-
-
-
-
-
trace
40
trace
trace
N. R.
19
N. R.
70
71
63
69
76
71
b
5
c
6
-
-
d
7
8
9
10
11
12
13
2
3
4
2
3
3
3
3
3
3
e
14
15
80
59
66
56
f
g
16
h
17
a
Reactions were conducted on a 0.1 mmol scale in 6 mL of THF in a
200 mL closed flask with 1.5 atm of NH3, XPhos = 2-
dicyclohexylphosphino-2,4,6′-triisopropylbiphenyl, yield of isolated
b
c
d
products. Toluene instead of THF. DME instead of THF. 1,4-
e
f
Dioxane instead of THF. 18 h instead of 12 h. 24 h instead of 12 h;
the lower yield was due to deboronation. [Ir(cod)Cl]2 instead of
[Ir(cod)OMe]2. [Ir(cod)OMe]2 (2.5 mol %) was employed.
g
h
SI). It was later found that the addition of base promoted such
B-H/N-H dehydrocoupling, and the use of 3.0 equiv of K2CO3
increased the yield of 3a to 71% (entries 8−10, Table 1 and
Table S3 in the SI). Further optimization of the phosphine
loading indicated 30 mol % was the best, resulting in 3a in 76%
isolated yield (entries 9, 12, and 13, Table 1). A longer
reaction time (18 h) further improved the yield of 3a to 80%
(entry 14, Table 1). However, a prolonged reaction time (24
h) led to a lower yield of 3a due to the deboronation (entry 15,
Table 1). The employment of [Ir(cod)Cl]2 resulted in a drop
of yield to 66% (entry 16, Table 1). In view of the yield of 3a,
entry 14 in Table 1 was chosen as the optimal reaction
conditions.
Various cage C- or B-substituted o-carboranes were
subsequently examined for this Ir-catalyzed B-H/N-H
dehydrocoupling, and results were compiled in Table 2. A
wide range of substituents including alkyl, aryl, and benzyl at
the B(9) position were compatible with this reaction, leading
to the corresponding B(3)-amino-o-carborane derivatives in
moderate to high isolated yields (3a-3m). Electron-donating
groups generally offered higher yields than the electron-
deficient ones (3f-3j). The thiophene-containing substrate
worked well, affording the aminated product 3m in 77%
isolated yield. 4-Ph-o-carborane afforded two regioisomers 6-
NH2-4-Ph-o-carborane (3na) and 3-NH2-4-Ph-o-carborane
(3nb) in 51 and 27% isolated yields, respectively. Various
substituents at cage C were also tolerated, and the
corresponding products 3o-3s were isolated in 46−82% yields.
The more steric demanding substituents at the cage C atom
provided the relatively lower yields (3a vs 3o vs 3p). The use
of 1,2-Me2-o-carborane led to no reaction (3t). Cage B(3)-
a
b
Reactions were conducted on a 0.1 mmol scale. Yield of isolated
c
d
e
products. 24 h instead of 18 h. 40 h instead of 18 h. 8 h instead of
18 h. 12 h instead of 18 h.
f
substituted carboranes reacted with NH3 smoothly to generate
the target 6-amino-o-carboranes in 66−72% isolated yields
(3u-3w).
Compounds 3 were fully characterized by H, 13C, 19F, and
1
11B NMR spectroscopy as well as high-resolution mass
spectrometry. The molecular structures of 3e, 3na, 3nb, 3s,
and 3u were further confirmed by single-crystal X-ray analyses.
To gain some insights into the reaction mechanism, several
control experiments were performed (Scheme 2). It was
expected that 1 equiv of H2 should be generated as the
byproduct during the reaction. Indeed, a significant peak at δ =
1
4.47 ppm assignable to H2 was observed in the H NMR
spectroscopy of the reaction mixture (see Figure S6 in the SI),
which was further confirmed by gas chromatography thermal
conductivity detector (GC-TCD) analyses (see Figure S7 in
the SI).15 The crucial role of the Ir(I) catalyst was proved by
the control experiment shown in Scheme 2a, as no reaction
was observed in the absence of [Ir(cod)OMe]2. On the other
hand, the yields for 3a were gradually enhanced by increasing
the loading of the Ir(I) catalyst in the absence of K2CO3
(Scheme 2b). Such a phenomenon may suggest the presence
of catalyst poisoning by the product 3-amino-o-carborane. To
test this hypothesis, product 3a was added to the reaction
4149
J. Am. Chem. Soc. 2021, 143, 4148−4153