Organic Letters
Letter
having either electron-rich or electron-deficient aryl substitu-
ents were found suitable under the optimized reaction
conditions to furnish the α-allyl products (3) in moderate to
good yield and with a moderate to good dr and uniformly high
enantioselectivity (entries 1−15). In general, allylic alcohols
containing electron-rich aryl substituents afforded the products
with excellent enantioselectivity. Allylic alcohols bearing
heterocyclic (2p−q) as well as alkenyl (2r) substituents also
gave the products in moderate to good yield with excellent
enantioselectivity (entries 16−18). In the latter case, the
reaction had to be carried out at 25 °C. Unfortunately, no
reaction was observed in the case of allylic alcohols with
aliphatic substituents, which marks a limitation of this protocol.
Having demonstrated the scope and limitations of allylic
alcohol, we examined the effect of deconjugated butyrolactam
substituents on the outcome of this reaction. Deviation from
1a by replacing the N-protecting group with p-methoxybenzyl
(PMB; 1b) led to a significant drop in the enantioselectivity
when reacted with allylic alcohol 2a (Scheme 2, entry 1).
room for improvement. However, it must be mentioned that in
nearly all of the cases, the minor diastereomer of the products
was formed with either the same or a very similar level of er.14
In line with our previous studies,11,12a,b no α-allyl conjugated
butyrolactam or γ-allyl deconjugated butyrolactam product was
detected in any of our reactions, and exclusive regioselectivity
was observed.
During product isolation through column chromatographic
purification, we observed a significant erosion in the dr in some
cases. For example, the α-allyl-deconjugated butyrolactam 3ah,
which was formed as a 7:1 mixture of two diastereomers
(Table 2, entry 8), when subjected to purification through
silica gel column chromatography, was isolated as a 2:1 dr. The
presence of an enolizable α-proton in 3ah made us wonder
whether the dr of the isolated product reflects the
thermodynamic ratio. To verify this possibility, we carefully
isolated a fraction of 3ah with a 3.4:1 dr and treated it with 10
mol % of Na2CO3 in EtOH at 25 °C for 24 h (Scheme 3). The
Scheme 3. Product Epimerization in Silica Gel
a
Scheme 2. Scope of Deconjugated Butyrolactams
recovered 3ah showed a dr of 2:1, which remained unchanged
even after chromatographic purification. This experiment
indicates that the diastereomeric ratios determined from the
crude catalytic reaction mixture (e.g., 7:1 for 3ah) reflect the
catalyst-controlled kinetic ratios, whereas those obtained from
the purified products (e.g., 2:1 for 3ah) reveal the equilibrium
ratios of the two product diastereomers. Considering the
higher acidity of the α-proton (ring proton) compared with
that of the allylic proton of 3, the diastereomers are likely to
arise from the epimerization of the ring proton (vide infra).
The scalability of this Ir-catalyzed AAA protocol was
examined on a 10-fold scale compared with that used for
displaying the substrate scope. Using a slightly lower loading
(4 mol %) of the Ir/(S)-L complex, the reaction between 1a
and 2a, while taking somewhat longer, furnished the product
3aa in similar yield and with similar enantioselectivity but with
a slight increase in the diastereoselectivity (Scheme 4A).
The densely functionalized α-allyl-substituted deconjugated
butyrolactams obtained in this reaction were deemed suitable
for elaboration to potentially useful building blocks. For
example, the catalytic hydrogenation of 3aa led to the isolation
of the α-(1-phenylpropyl)-substituted deconjugated butyro-
lactam 4 in 93% yield with a 7:1 dr and a 99:1 er (Scheme 4B).
Olefin cross metathesis with methyl acrylate in the presence of
Grubbs second-generation catalyst produced 5 in modest yield
but with a diminished dr. Ir-catalyzed hydroboration16 of 3aa
followed by oxidation with sodium perborate led to an overall
anti-Markovnikov hydration. The resulting alcohol 6 is the
product isolated from our previously reported catalytic
enantioselective Michael reaction,12b and it helped in
establishing its relative and absolute configurations as well as
those of the AA product 3aa. The stereochemistry of the other
a
1
Yields refer to the isolated product. dr is determined by H NMR
analysis of the crude reaction mixture. er corresponds to that of the
major diastereomer and was determined by HPLC analysis.
However, an excellent level of enantioselectivity was preserved
when butyrolactams with other electron-rich N-protecting
groups such as 3,4-dimethoxybenzyl (1c) and 2,4-dimethox-
ybenzyl (1d) were employed (entries 2 and 3). Our protocol
was found to be tolerant to bulkier alkyl substituents at the C5
position of butyrolactam, and the products (3ea and 3fa) were
formed with an outstanding er (entries 4 and 5). Similarly, 4-
benzyloxycarbonyl-substituted butyrolactam (1g), even though
N-protected with PMB, furnished the product (3ga) with a
99:1 er (entry 6).
Notwithstanding the excellent level of enantioselectivity
observed in most of the examples shown above, the yield and
diastereoselectivity generally remained modest, thereby leaving
3023
Org. Lett. 2021, 23, 3021−3026