Organic Letters
Letter
(Figure 1) as the precatalyst and 1,4-dioxane as the solvent.
Electron-rich aryl chloride 1, containing a para-morpholino
substituent, and sec-butanol (s-BuOH) were chosen as model
coupling partners. Lowering the number of equivalents of s-
BuOH from 3 to 1.2 resulted in a decline in the efficiency of
the reaction: the conversion of the starting aryl chloride 1
decreased by 50%, and the ratio of the reduction side product
4 to the coupling product 3 increased substantially (Table 1,
entries 1−3). However, when the solvent was changed to THF,
lowering the number of equivalents of s-BuOH to 1.2 had a
negligible effect on the efficiency of the reaction (Table 1,
entries 4−6). Therefore, THF was selected as an appropriate
solvent for further exploration of the substrate scope.
Scheme 3. Pd-Catalyzed C−O Cross-Coupling of
a
(Hetero)aryl Chlorides with Secondary Alcohols
Although it is widely accepted that aryl bromides exhibit
higher reactivity in cross-coupling reactions than aryl
chlorides,5b,i,10 we observed an opposite trend between
electron-rich aryl chloride 1 and aryl bromide 2. First, under
the same set of reaction conditions (Table 1, entries 6 and 9),
while the reaction of 1 provided a 90% yield of desired aryl
ether 3, along with 9% reduction product 4, that of 2 only
provided a 63% yield of 3, with a notable increase in the
amount of 4 (33%) that was formed. Although this difference
could be partially ameliorated by adjusting the quantity of s-
BuOH to 2 equiv, a further increase of the amount of alcohol
utilized did not lead to an additional improvement in the yield
of 3 (Table 1, entries 7 and 8).
A variety of (hetero)aryl chlorides and secondary alcohols
were surveyed to examine the generality of this method
(Scheme 3). The C−O cross-coupling reactions took place
under mild conditions, using only a 20% molar excess of
alcohols. Many traditionally challenging substrates, including
unactivated aryl chlorides (3, 5, 7) and five-membered
heterocycles (9, 10, 11), readily underwent C−O bond
formation at room temperature. Various heterocycles, such as
a quinoline (7), a pyridazine (8), a pyrazole (8), a thiadiazole
(9), a benzisothiazole (10), a benzimidazole (11), a pyrazine
(12), a quinazoline (13), a pyrazolopyrimidine (14), and a
pyridine (15), were tolerated as structural components in the
electrophiles. Functional groups such as an unprotected
tertiary hydroxyl group (6), a carbamate group (7), and a
lactone (14) were also compatible with the reaction
conditions. While sterically accessible alcohols proved to be
good coupling partners at room temperature, secondary
alcohols with moderate steric encumbrance at either the α-
carbon (12, 13) or the β-carbon (14, 15) required moderate
heating (40 °C) to react with activated heteroaryl chlorides
and afford corresponding heteroaryl ethers in ≥80% yields.
The coupling reactions between electron-rich aryl chlorides
and more sterically demanding nucleophiles, however,
remained challenging, as demonstrated in the reactions of
aryl chloride 1 (Table 2). As the steric congestion around the
α-carbon of the alcohol increased (Table 2, entries 1 and 2),
both the conversion and the yield of desired product decreased
by approximately 40%, while a small increase in reduction
product 4 was observed. We hypothesize that increasing the
steric bulk around the α-carbon could negatively impact the
binding tendency of alcohol nucleophiles to the oxidative-
addition complex II (Scheme 1), thus accounting for the less
efficient C−O bond formation. Although heteroaryl ethers 9,
12, and 13 were prepared and isolated in >80% yields, benzylic
alcohols proved to be more difficult coupling partners for
electron-rich aryl chloride 1 (Table 2, entries 3 and 4). The
steric environment of the benzylic carbon also played an
a
Reaction conditions: ArCl (1.0 mmol), alcohol (1.2 mmol), NaOt-
Bu (1.2 mmol), P2 (2.0−4.0 mol %), THF (1.0 mL, 1.0 M), rt−40
°C, 18 h. Isolated yields represent the average result of two runs.
b
THF (7.0 mL) and higher temperature were used due to the poor
solubility of the combination of aryl chloride, alcohol, and NaOt-Bu
under standard reaction conditions.
important role in the coupling process, as a change from a
methyl to an ethyl group led to a 60% decrease in conversion,
and only a trace amount of desired product was detected
(Table 2, entries 3 and 4).
As we continued to examine the scope of C−O cross-
coupling reactions of aryl bromides, we noticed that the
difference in reaction efficiency between the cross-coupling of
aryl bromides and aryl chlorides was greatest for highly
electron-rich substrates, such as 2 vs 1 (to prepare 3 in
Schemes 4 and 3, respectively). In contrast, for weakly
electron-rich, electron-neutral, and electron-deficient aryl
bromides (to prepare 5, 16, and 17, respectively), the cross-
coupling reactions proceeded with comparable levels of
efficiency (>80% yield, Scheme 4).
To gain an understanding of the difference in reactivity
between aryl halides 1 and 2, we performed experiments to
examine the cause of increased reduction in the cross-coupling
reaction of aryl bromide 2. In order to ascertain whether
reduction product 4 resulted solely from β-hydride elimination,
we prepared α-deutero-alcohol 18-d (98% d1, see the
bromide 2. When 2 and protio-alcohol 18 were subjected to
C
Org. Lett. XXXX, XXX, XXX−XXX