Organic Letters
Letter
introduce the sterically congested quaternary carbons to para-
substituted phenols with the reactive OH group untouched.
Control experiments and density functional theory (DFT)
calculations show that the reaction proceeds through a phenol
alkenylation/hydroalkynylation sequence with the assistance of
the phenolic OH group. Moreover, the prominent kinetic
differentiation between the alkenylation and hydroalkynylation
processes enables us to construct unique triaryl-substituted
quaternary carbons.
We began investigating the reaction of 4-methoxyphenol 1a
and phenylacetylene 2a in CH2Cl2 using 10 mol % of B(C6F5)3
as the catalyst (see Tables S1 for optimization details).
Treatment of 1a with 2a at room temperature could only
afford 4aa in poor yields along with 3aa as a major product. It
is probably due to the deterioration of B(C6F5)3 via the
carboboration reaction17 with terminal alkyne or other
We were delighted to find that lowering the reaction
temperature to −20 °C could increase the yield of 4aa to
85% isolated yield. Other Lewis acids, including BCl3, BBr3,
(C6F5)CH2CH2B(C6F5)2, FeCl3, and ZnCl2, afford little or no
desired products 4aa or related alkenylation product 3aa
unidentifiable byproducts. Although the reaction was specu-
lated to involve the protonation of alkyne with B(C6F5)3-
phenol adduct, no desired product was detected in the
presence of Brønsted acids.
lation mechanism,18 and the existence of the OH group at the
ortho-position of phenol is important for the hydroalkynylation
step. Furthermore, deuterium labeling experiments confirmed
the involvement of the protonation of alkyne in the
alkenylation step and the deprotonation of terminal alkyne in
the hydroalkynyaltion step, respectively (see Scheme 2c and
To elucidate the mechanistic details of this reaction, DFT
calculations at the M06-2X/cc-pVTZ//M06-2X/6-311G(d,p)
level19 were performed on the model reaction of 4-
methoxyphenol 1a and phenylacetylene 2a with B(C6F5)3 as
the catalyst (see SI for computational details, and other
kinetically less favorable pathways). As shown in Figure 1,
proton transfer from the OH group of INT1 to alkyne 2a
forms a tight ion-pair INT2 with a barrier of 16.6 kcal mol−1
(via TS1). INT2 consists of a highly electrophilic vinyl cation
and a borate-phenol anion as the counteranion, which could
readily undergo electrophilic addition reaction to afford the
Wheland intermediate INT3 (via TS2, ΔG‡ = 13.9 kcal
mol−1). Subsequent rearomatization of INT3 followed by the
dissociation of B(C6F5)3 affords the alkenylated phenol 3aa
and regenerates the catalyst.
In competing with catalyst regeneration, 3aa-B(C6F5)3
complex INT5 could also undergo intramolecular protonation
reaction (via TS4, ΔG‡ = 12.7 kcal mol−1) to generate a
borane-stabilized tertiary carbenium ion INT6 (Figure 1,
right). It then undergoes electrophilic addition reaction with
another molecule 2a to form a zwitterionic intermediate INT7
(via TS5). This step has a barrier of 21.8 kcal mol−1, and the
formation of INT7 is endergonic by 10.7 kcal mol−1 relative to
2a and INT6. Then, INT7 is deprotonated to provide the
neutral complex INT8 with a barrier of 12.2 kcal mol−1.
Finally, Lewis acid−base dissociation of INT8 regenerates
B(C6F5)3 and releases the ortho-propargylation product 4aa,
which is the major product obtained experimentally. Along the
whole reaction process, the rate-determining step is the
hydroalkynylation step (3aa → 4aa) with a barrier of 21.8
kcal mol−1 (via TS5). The alkenylation step is kinetically
favored over the hydroalkynylation step by 5.2 kcal mol−1, but
the former is thermodynamically less favored than the latter.
These computational results can account for the experimental
observation that the ortho-propargyl phenols are accessible
through the reaction of phenol with a terminal alkyne under
mild conditions. Besides, the proton-initiated mechanism20 or
the direct addition of phenol toward B(C6F5)3/alkyne adduct17
could also be excluded because of the involvement of high-
energy transition states (see Figures S8−S9 in the SI for
details).
To shed light on the mechanism, we subjected 2-vinylphenol
3aa to the standard reaction conditions in the presence of 4-
chlorophenylacetylene 2b, the corresponding phenol 4aab
could be obtained in 67% yield (Scheme 2a). However, the
reaction of meta-alkenyl phenol 5 with phenylacetylene 2a did
not give the related hydroalkynylation product with 76%
recovery of the starting material 5 (Scheme 2b). These results
indicate the propargylation reaction of phenol may proceed
through a sequential phenol-alkenylation/alkene-hydroalkyny-
Scheme 2. Control Experiments
Then, the scope of phenols was examined with phenyl-
acetylene 2a (Scheme 3). Para-substituted phenols bearing
electron-donating substituents, including alkoxyl, phenoxyl,
and alkyl, could undergo ortho-propargylation reactions to give
the related phenols in good to excellent yields (4ba−4ga,
61%−88%). Para-halogenated phenols are also applicable to
the method, affording the corresponding ortho-propargylation
products in moderate to excellent yields (4ha−4ka) in the
presence of 15 mol % of B(C6F5)3 at −40 °C. It should be
noted that the compatibility of synthetically valuable C−Br and
C−I bonds provide opportunities for further synthetic
manipulations. Trisubstituted phenol 1l could also be used
as the coupling partners to afford the desired product 4la in
moderate yield. However, the current strategy is not applicable
to phenols containing groups with a strong coordinating ability
5534
Org. Lett. 2021, 23, 5533−5538