Organometallics
Communication
RESULTS AND DISCUSSION
Scheme 2. Control Experiments Performed to Understand
■
First, the catalytic capability of Ca[N(SiMe ) ] (THF) for the
3
2 2
2
hydroborative reduction with HBpin was examined with
tertiary amides. The reduction was performed under the
catalyst loading of 5 mol % Ca[N(SiMe ) ] (THF) in C D at
3
2 2
2
6
6
2
5 °C. Although only 2 equiv of HBpin (relative to amide) is
required by stoichiometry to complete amide reduction, 5
equiv of HBpin was included into this transformation to
compensate for the possible formation of transient amide−
34,40
borane and amine−borane adducts during the reaction.
The amide reduction proceeded via deoxygenation to give the
corresponding tertiary amine. As shown in Table 1,
quantitative conversion was obtained in 4 h at 25 °C using
DMF as the substrate (run 1). Due to steric encumbrance,
bulky aliphatic amides showed relatively low activity (run 2
and 3). Notably, N-allyl-N-methylbenzamide was selectively
reduced into the corresponding amine without the contami-
nation of an intramolecular alkene hydroboration product,
indicative of an excellent chemoselectivity (run 3). This
strategy was applicable for secondary amide reduction;
however, it required an elevated temperature and longer
reaction time to achieve good yields. Preliminary results
summarized in Table 1 show that Ca[N(SiMe ) ] (THF)
2
(
3
2
3 2
along with a white precipitate (reaction a). The white powder
was supposed to be a Ca-hydrido species, despite the fact that
3
2 2
worked excellently for the reduction of N-alkyl substituted
amides, including aliphatic, aromatic amides and lactams,
giving the corresponding amines in quantitative yields at 120
commercially available CaH was inert to this reduction. In
2
addition, the stoichiometric reaction of Ca[N-
(
SiMe ) ] (THF) with (p-CH )C H CONH in a 1:2
3 2 2 2 3 6 4 2
°
C for 24 h (>99%) (runs 4−6).
molar ratio quantitatively gave HN(SiMe ) , and an insoluble
3
2
Since Ca[N(SiMe ) ] (THF) /HBpin was effective for the
3
2 2
2
catalytic reduction of tertiary and secondary amides, it
motivated us to extend this catalytic system to a more
challenging task of deoxygenating primary amides. Encourag-
ingly, it was revealed that Ca[N(SiMe ) ] (THF) could serve
40
when isolated, showed no activity toward amide reduction.
These results indicated that the active species should be a
calcium-hydrido species, like [BpinCaH] ([Ca−H], Figure S4
3
2 2
2
as an efficient catalyst for the selective reduction of primary
amides with HBpin. The hydroborylated products were
isolated as their hydrochloride salts. As illustrated in Table 1,
both aromatic and aliphatic primary amides were reduced by
HBpin in the presence of 5 mol % Ca[N(SiMe ) ] (THF) at
in the SI), which was similar to the active species in the
47,48
hydroboration employing metal-based catalysts.
Besides,
the reaction of (p-CH )C H CONH with excessive HBpin at
3
6
4
2
room temperature gave the pinacolborane-substituted secon-
3
2 2
2
dary amide (p-CH )C H CONH(Bpin) as the major product
3
6
4
1
20 °C in toluene. Aliphatic primary amides such as
(
(
92%) and the N-borylated imine (p-CH )C H CHN-
3
formamide, hexanamide, and cyclohexanecarboxamide were
subjected to reduction to afford the corresponding amine salts
in 72−90% yields (runs 7 and 8). Benzamide and its
derivatives bearing electron-donating (−Me and −OMe) and
Bpin) as the minor product (8%) (reaction b, Figure S7 in
49
the SI). In this respect, it gave the choice to mimic the
reaction mechanism of the reduction of primary amides with
that of second amides. Furthermore, the addition of
39
electron-withdrawing (−F, −Cl, and −NO ) functional groups
2
Ca[N(SiMe ) ] (THF) into the reaction mixture of (p-
3
2 2
2
at the para position were also reduced into their corresponding
amine hydrochlorides in moderate to excellent yields (68−
CH )C H CONH(Bpin) and HBpin at room temperature
3
6
4
accelerated the formation of (p-CH )C H CHN(Bpin),
3
6
4
9
3%) (runs 10−15). The dependence of the electronic effect
of substituents on the reactivity was not significant. It was
worth noting that a highly reducible functional group nitro
group on the benzamide framework remained intact, indicative
of a good chemoselectivity (run 14). Steric hindrance would
affect the reduction activity. For example, introduction of
cycle toward the final product (reaction c, Figure S12 in the
38,39
1
SI).
Notably, H NMR monitoring showed the disappear-
ance of the imine intermediate and the generation of (p-
CH )C H CH N(Bpin) when (p-CH )C H CONH(Bpin)
3
6
4
2
2
3
6
4
was treated with HBpin in the presence of Ca[N-
−
CH in the aromatic ring ortho to the amide moiety resulted
3
2
in a strikingly low activity (run 10 vs 16). Besides, this
reduction reaction showed a good tolerance of heteroatoms,
such as the thienyl group (run 17).
d, Figure S14 in the SI). This finding showed that the
enhanced reaction temperature favored the transformation of
the imine to (p-CH )C H CH N(Bpin) . On the basis of these
3
6
4
2
2
28,50
To date, there are only limited examples describing the
mechanistic pathway for the catalytic reduction of primary
preliminary results and the previous literature,
a possible
reaction mechanism proposed for primary amide reduction is
outlined in Scheme 3. Activation of Ca[N(SiMe ) ] (THF)
3
8,42,43
amides via hydroboration.
To provide a plausible
3
2 2
2
mechanism for the deoxygenation of primary amides into
amines with the present catalytic system, we conducted some
control experiments. As shown in Scheme 2, treatment of
Ca[N(SiMe ) ] (THF) with 2 equiv of HBpin in C D
6
with HBpin generated Ca(II)-hydride species A. Treatment of
a primary amide with excessive HBpin gave a N-monobory-
lated second amide B, which reacted with A via insertion of an
amide into the CaH bond to give intermediate C. Further
3
2 2
2
6
1
203
Organometallics 2021, 40, 1201−1206
Gong, Mingliang
