ACS Catalysis
Research Article
reaction conditions, giving monoarylated products (2x, 2y, and
2z) in good yield. These results provide a platform for further
utilization of meta-arylated products via transition metal-
catalyzed cross-coupling reactions. Diaryliodonium salts
bearing electron-releasing or electron withdrawing meta-
substituents were also effective in this protocol, giving the
corresponding products (2ab−2ad). Also, o-methoxy iodo-
nium salt reacted smoothly to give meta-arylated carbamate
(2ae) in 42% yield.
This arylation method can also be applied to get meta-
arylated carbamates in quantitative yields. After the first step,
treating the crude reaction mixture with Me2NCOCl and
K2CO3 in acetonitrile at 80 °C for 2 h, we can access the
corresponding carbamates in quantitative yields (Scheme 4).
To explore the utility of this process on a preparative scale,
we performed the reaction in gram scale with substrate 1a and
diphenyliodonium tetrafluoroborate salt. Gratifyingly, our
copper-catalyzed meta-arylation of carbamates worked well in
gram scale to give the desired product 2a in 72% isolated yield
(Scheme 5).
reaction is even more favored than the anti attack in
mechanism A, as the TS for the syn attack can have some
degree of stabilization because of the interaction between the
acetoxy oxygen and the CuIII site (Figure S1).
Based on previous literature,15 it is known that diary-
liodonium salts react with electron-rich arenes under metal-free
conditions to form arylated products. It is also known that
iodine and compounds of iodine behave in a similar manner to
the copper-catalyzed transformations.16 Hence, we speculate
that the meta-selective arylation in the absence of copper
catalysts proceeds, as shown in Figure 3. As there is no
possibility to form a very reactive Cu(III)−aryl intermediate in
the absence of the copper catalyst, we support our observation
of less yield at 70 °C for the uncatalyzed reaction. At this
moment, we are not clear about the exact mechanism for the
uncatalyzed reaction, and further investigations are needed to
support the proposed mechanism.17
In summary, we have demonstrated the first example, in the
literature, of a meta-selective arylation protocol of phenol
derivatives catalyzed by copper. The current method provides
room for further functionalization of the obtained products
and thus would find applications in pharmaceuticals and
complex molecule synthesis. Computational studies showed
that the reactions most probably proceeded via Heck-like four-
membered ring TSs.
To demonstrate the synthetic utility of this meta-arylation,
we attempted diversification of 3r using the carbamate as a
synthetic handle (Scheme 6). Carbamate 3r was converted
into possible other products in good yields by using the
reported methods such as reductive cleavage11a and cross-
coupling11b reactions.
In the literature,5a,12 there are two proposed mechanisms for
a similar meta-arylation process, as shown in Figure 1. The
mechanism (A) shown in Figure 1a proceeds via anti-
oxycupration, whereas mechanism (B) shown in Figure 1b
proceeds via a Heck-like four-membered transition state (TS).
Li, Wu, and co-workers have performed density functional
theory (DFT) calculations for the reaction between acetanilide
and PhCu(OTf)2and demonstrated that mechanism B is much
more stable and feasible.12 To probe the mechanism of the
reaction of acyl-type substrates examined in this work, we
performed DFT computational studies of the reaction between
the acetoxy-substituted substrate (2) and PhCu(OTf)2 using
the M06-2X functional.13,14 As in the case of anilides,13 our
DFT results show that the Heck-like mechanism B is much
more favored than mechanism A in the reaction of acyl
substrates, with the reaction in mechanism B proceeding with a
much lower activation barrier. Importantly, the TS for the
meta attack (TS1) is lower (11.0 kcal/mol) in energy than the
TSs for the ortho and para attacks by >6 kcal/mol. The
preference for the meta attack can be attributed to the
coordination bond existing between acetoxy oxygen and the
CuIII site (Figures 1b and 2), which the other TSs cannot form
TS1 is followed by a very stable intermediate (Int1), and the
subsequent barrier for proton abstraction from the meta
carbon via TS2 is rather small (6.8 kcal/mol). Therefore, the
selectivity of the reaction is determined by the first C−C bond
formation step, and the DFT results explain well why only the
meta product is obtained in our experiments. As shown by Li,
Wu, and co-workers for anilides, mechanism A has a
prohibitively high barrier (Figure S2). For this mechanism,
especially for the reaction of 2 and at the level of theory
employed here, a CuIII−aryl intermediate could not be
obtained, but C−C bond formation occurred directly after
the TS in mechanism A. These results suggest that mechanism
A is unlikely to operate under actual experimental conditions.
It is also interesting to note that the syn analogue of the
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
Experimental procedures and spectral data for all new
compounds (1H NMR, 13C NMR, and HRMS) (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
Hajime Hirao − School of Life and Health Sciences and
Warshel Institute for Computational Biology, The Chinese
University of Hong Kong, Shenzhen, Guangdong Province
Teck-Peng Loh − Institute of Advanced Synthesis (IAS),
School of Chemistry and Chemical Engineering, Northwestern
Polytechnical University (NPU), Xi’an 710072, China;
Yangtze River Delta Research Institute of NPU, Taicang,
Jiangsu 215400, China; Division of Chemistry and Biological
Chemistry, School of Physical and Mathematical Sciences,
Nanyang Technological University, 637371, Singapore;
Author
Manikantha Maraswami − Division of Chemistry and
Biological Chemistry, School of Physical and Mathematical
Sciences, Nanyang Technological University, 637371,
Singapore
Complete contact information is available at:
Notes
The authors declare no competing financial interest.
2307
ACS Catal. 2021, 11, 2302−2309