X. Wang, G. Xie, Y. Zhao et al.
Tetrahedron Letters 72 (2021) 153069
Table 1
Examination of Sm/additive on the pinacol coupling of 2-heptanone 2a.a
Entry
Sm (equiv.)
Additive (equiv.)
Reaction time
Yield %b
1
2
3
4
5
6
7
8
9
10
1.0
1.0
1.0
1.0
1.0
1.0
0.33
1.0
1.0
1.0
TMSBr (1.0)
TMSCl (1.0)
TMSI (1.0)
BBr3 (1.0)
PBr3 (1.0)
Et3N∙HBr (1.0)
TMSBr (1.0)
TMSBr (0.1)
TMSBr (0.5)
TMSBr (2.0)
0.5 h
10 h
2.5 h
Over night
Over night
Over night
6 h
0.5 h
0.5 h
78%
0
71%
20%
0%
0
63%
16%c
60%
37%
0.5 h
a
Reaction conditions: Sm 1.0 mmol (0.15 g), 2-heptanone (1.0 mmol, 0.114 g) and the additive in dry THF 3 mL under N2 at the room temperature. The reaction mixture
was quenched with Bu4NF (1.0 M in THF, 1.2 mL) and allowed to be stirred for 1 h followed by usual workup (see SI).
b
Isolated yields.
With 71% of the starting material recovered.
c
cess was monitored by GC–MS. It is satisfying that 78% of isolated
yield of the desired aliphatic diol 2a was obtained in 30 min. under
the mild conditions (Table 1, entry 1). A blue slurry was gradually
formed in the reaction indicating the formation of Sm(II) species.
For comparison, the reaction promoted by Sm/TMSCl and Sm/TMSI
were examined. No color change could be observed using Sm/
TMSCl system and GC–MS analysis showed no formation of the
desired diol even in 10 h except a small amount of the 2-heptanol
(~5% yield). The result is consistent with the report by Yu and
Zhang [9], where aliphatic ketones are sluggish substrates in Sm/
TMSCl. Slightly lower yield of the diol was obtained using Sm/TMSI
(Table 1, entries 2 and 3), and in this case, the reaction mixture also
took on blue color. Besides, BBr3, PBr3 and Et3N∙HBr were also
examined as the bromo source. The reaction with Sm/BBr3 was
sluggish as most starting ketone remained unreacted and only
20% yield of the desired pinacol was detected even by prolonging
the time overnight (Talbe 1, entries 4). No desired reaction could
be observed at all using Sm/PBr3 or Sm/ Et3N∙HBr (Talbe 1, entries
5 and 6).
From the viewpoint of atom economy, 1/3 mmol of Sm, with its
complete transformation into Sm(III) oxidation state, would be
enough to achieve the complete coupling of 1 mmol of the ketone.
Thus the substoichiometric amount of Sm was attempted and it
was found relatively good yield was afforded (Table 1, entry 7).
Screening on the loading of TMSBr showed 0.5 equiv. of TMSBr
could afford reasonable yields, while 0.1 equiv. or 2.0 equiv. of
TMSBr gave unsatisfactory yields (Table 1, entries 8–10). The
decrease in the yield of diol with excess TMSBr could be rational-
ized by the competitive formation of hexamethyl disilante, as has
been monitored by GC–MS.
trast, aromatic ones are less efficient. For acetophenone 1k 65% of
the starting material remained in 30 min. together with the diol in
22% yield (Table 2, entries 1–10 vs entry 11). Sterically hindered
ketone 1l is not a suitable substrate and only trace amount of the
diol was detected (Table 2, entry 12). As far as we know, the avail-
able reductive coupling methods always realize the pinacolization
of aromatic ketones and aldehydes better than that of their alipha-
tic counterparts [10]. Therefore, the above results can not be satis-
factorily explained by initiation via the reduction of the carbonyl to
form ketyl since the reduction of aliphatic carbonyls is more diffi-
cult than aromatic ones. The same phenomenon was observed
between aliphatic and aromatic aldehydes (Table 2, entries 13–
19 vs entries 20–21). Unfortunatley, aldehydes usually afforded
lower yields under the standard conditions, where simple reduced
products was the main side reaction together with several minor
by-products mixture. For aromatic aldehydes, 1u with the elec-
tro-withdrawing F group afforded better yield of the diol than 1t
with an electro-donating methyl group.
Measurement of cyclic voltammetry of TMSCl, TMSBr and TMSI
found the reductive potentials were ca. À2.58, À1.67 and À0.488 V
respectively (vs Ag/AgCl with Fc/Fc+ as internal standard, see sup-
porting information), which could account for the facts that Sm
reacts with TMSBr and TMSI to give a green suspension of Sm(II)
species while the reaction between Sm and TMSCl is sluggish.
In view of the more negative reductive potentials of carbonyl
compounds (-2.05 to À2.23 V vs Ag/AgCl) and aliphatic carbonyl
compounds (~3.00 V) [11], it is reasonable that Sm may reduce
TMSBr preferentially in the coexistence of ketones and aldehy-
des. Furthermore, the more negative reductive potentials of ali-
phatic carbonyls could not rationale why the reaction of
aliphatic carbonyls is more readily than aromatic ones in the
system. A mechanism rather than ketyl related coupling should
exist.
Considering the simplicity of the condition and the good yield
(entry 1, Table 1), no further optimization of the reaction was con-
ducted and the ratio of Sm/TMSBr/substrate (1:1:1) was estab-
lished as the standard condition. Subsequently, the scope of the
reaction was examined by extending the substrates to a variety
of ketones and aldehydes. The results are shown in Table 2.
It is noteworthy the Sm/TMSBr system afforded good to high
yields of the diols rapidly for aliphatic ketones (in 30 min.). In con-
According to the research on the pinacolization with Yb/TMSBr
by Takaki and Fujiwan et al. [12], where YbBr2 was proposed to be
the active species, it seems reasonable to propose the key Sm(II)
species produced here maybe SmBr2 (Scheme 1, path a). However,
treatment of 2-heptanone with SmBr2 (1.0 equv.) or SmBr2 (1.0
2