Page 145 - Read Online

Page 2 of 9 Liu et al. Chem Synth 2023;3:22 https://dx.doi.org/10.20517/cs.2023.18

the increased entropy of the reaction system as well as the principle of microscopic reversibility under
thermal conditions [9,10] . To overcome this hurdle, various strategies have been devised to achieve successful
[1-8]
deracemization , including the use of excited states (via photochemical condition) [1,2,11-14] , reversal of
thermodynamics by extrusion of small gas molecules [15,16] , and the design of multistep reaction sequence
(e.g., kinetic resolution or dynamic kinetic resolution) [17-21] . However, there are limited examples of
successful implementation of these strategies, and more efficient methods for this purpose remain in high
demand.

Enantioenriched organic molecules with benzylic chirality show broad applications in various fields,
including organic synthesis, medicinal chemistry, and materials science [22,23] . In particular, a stereogenic
carbon center attached to multiple aryl groups represents an important substructure widely observed in
natural products and biologically active molecules [24-28] . In contrast to the well-documented diverse strategies
to construct benzylic stereogenic centers, the exploitation of the deracemization approach for this purpose
has been underdeveloped in general. Among these limited examples, the majority have dealt with those
bearing one aryl group at the benzylic position [Scheme 1a] [29-33] . Instead, only very few deracemization
protocols have been developed for access to enantioenriched 1,1-diarylalkanes with a diaryl-substituted
stereogenic center [34-36] . More disappointingly, to the best of our knowledge, there has been no
demonstration of deracemization of triaryl-substituted stereogenic centers, despite the fact that 1,1,1-
triarylalkanes are versatile structures in medicinal chemistry. In this context, here we report the first
example of this type employing para-quinone methides as the key intermediate.

Recently, Liu and co-workers have reported a series of elegant organocatalytic redox racemization examples
with outstanding performance for the access to enantioenriched chiral molecules bearing benzylic
stereogenic centers [34-39] . Inspired by this strategy as well as our previous efforts in the study of asymmetric
processes involving para-quinone methides (p-QMs) [40-56] , we envisioned that the deracemization of
triarylmethane 1 could be potentially achieved by a similar strategy. Specifically, initial oxidation is expected
to form the p-QM intermediate. Next, in the same pot, a reductant, as well as a chiral catalyst, would affect
the asymmetric reduction of this key intermediate, thereby representing a formal deracemization
[Scheme 1d]. The challenges associated with this strategy include not only stereo control which requires
3
2
discrimination between two aryl groups (Ar and Ar ), but also the compatibility of the two steps which
involve mutually destructive oxidant and reductant.

EXPERIMENTAL
At room temperature, a solution of the triarylmethane 1 (0.4 mmol) and DDQ (99.0 mg, 0.44 mmol, 1.1
equiv) in CHCl (1.44 mL) was charged into an oven dried 4 mL vial. The mixture was stirred for 5 h and
3
then cooled. The catalyst (R)-A3 and the hydrogen source (0.6 mmol, 1.5 equiv) were added to a lower
temperature as specified in each case. The mixture was stirred for 96 h. Upon completion, as monitored by
TLC, it was concentrated under reduced pressure. The residue was purified by silica gel column
chromatography to afford the desired product 2.

RESULTS AND DISCUSSION
The racemic triarylmethane 1a was chosen as the model substrate for the initial study [Figure 1]. The phenol
ring serves as the precursor to the p-QM structure. To distinguish the remaining two aryl groups, one of
them was substituted with an ortho-methoxy group to provide additional interaction with the catalyst [51-56] .
DDQ was used as an oxidant for the first step. Based on TLC analysis, this step could be achieved cleanly in
DCE at room temperature within 4 h. Notably, other oxidants, including Ag O, TEMPO, Mn(acac) , and
2
3
O , could not work as effectively as DDQ. Next, the search for a suitable reductant and a chiral catalyst
2