.
Angewandte
Communications
Table 1: Optimization of the reaction conditions.
and ketone groups (Scheme 3 and Table 2, product 2w), were
tolerated under the reaction conditions. Notably, only the
desired product was isolated when other potential directing
groups, such as a methoxy or acetoxy group, was present
(Table 2, products 2e,f,p,q).[12] When the silanol precursor
À
contained two possible reactive sites, C H carboxylation
occurred preferentially at the sterically less demanding
position (products 2l–p). Bis(silanol) derivative 1v was
transformed into monoacid 2v in excellent yield. Evidently,
the newly installed carboxylic group electronically deactivates
Entry
Ligand[a]
Oxidant
Solvent
Yield [%][b]
1
2
3
4
5
L3
/
AgOAc
AgOAc
AgOAc
AgOAc
AgOAc
AgOAc
AgOAc
AgOAc
Cu(OAc)2
BQ
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
dioxane
EtCN
xylene
17
<1
37
33
42
90
59
0
5
3
7
<1
4
À
the aromatic ring, thus preventing a second C H carboxyla-
L1
L2
L4
L5
L5
L5
L5
L5
L5
L5
L5
L5
tion event. Importantly, substrates possessing a sensitive
cinnamyl group at the para or meta position (substrates 1w,x)
were also competent reactants: The corresponding salicylic
acids 2w,x were produced in reasonable yields. Likewise, the
carbazole derivative 1y reacted smoothly to give 2y in 64%
yield. In contrast, the ester-substituted phenol 1z was much
less reactive under these conditions (Table 2, entry 26), and
strongly electron deficient 3-NO2- and 4-CN-substituted
phenols did not react at all. Notably, this method tolerates
substituents at the ortho position, as demonstrated by the
carboxylation of 1h (Me), 1i (Ph), 1j (naphthalene), and 1k
(chloro). This result is in sharp contrast with our previous
6
7[c]
8[d]
8
9
10[e]
11
12
13
O2
AgOAc
AgOAc
AgOAc
20
[a] L1=Ac-Val-OH; L2=Boc-Val-OH; L3=(+)-menthyl(O2C)-Leu-OH;
L4=Ac-Leu-OH; L5=Boc-Leu-OH. [b] The yield was determined by GC
with nC15H32 as the internal standard. [c] The reaction was carried out
with 5 mol% of Pd(OAc)2 and 10 mol% of the ligand. [d] No Pd(OAc)2
was added. [e] The reaction was carried out under an O2 atmosphere (O2
balloon). Boc=tert-butoxycarbonyl, DCE=1,2-dichloroethane.
À
results for the silanol-directed C H olefination reaction, in
which case substrates possessing substituents at the ortho
position showed no reactivity.[10a]
Next, we were eager to clarify the reaction pathway. In our
À
previously developed silanol-directed C H oxygenation of
phenols,[10b] with the aid of a 18O-labeling study, we estab-
lished that the silanol oxygen atom was not incorporated into
the oxygenated product C (Scheme 2a). Accordingly, we
under the carboxylation conditions, along with the cleanness
of the reaction, encouraged us to focus on optimization
studies. Only a trace amount of the product was detected in
a control experiment without a ligand (Table 1, entry 2).
Accordingly, we screened a range of monoprotected amino
acid (MPAA) ligands in this reaction. Notably, the product
yield increased to 37% with Ac-Val-OH as the ligand
(Table 1, entry 3). Among the other ligands tested (Table 1,
entries 4–6; see the Supporting Information for the complete
range of ligands screened), Boc-Leu-OH exhibited the best
reactivity with the production of 2b’ in 90% yield (entry 6). A
decrease in the amount of palladium used to 5 mol% had
a deleterious effect on the yield (Table 1, entry 7), and no
carboxylation occurred in the absence of palladium (entry 8).
Other oxidants and solvents were also examined. Cu(OAc)2,
benzoquinone (BQ), and O2, which are commonly used
Scheme 2. 18O-Labeling studies. Piv=pivaloyl.
À
oxidants in C H functionalization, were ineffective (Table 1,
entries 9–11), and solvents other than DCE gave poorer
results (entries 12–14).
18
À
prepared 1b- O and subjected it to the conditions of our C
H carboxylation reaction. In contrast to the previous study,
this experiment revealed complete retention of the 18O label
in the silacycle 2b’ (Scheme 2b). Therefore, we propose that
Under the optimized conditions for carboxylation, fol-
lowed by a routine desilylation step, silanol 1a was converted
into salicylic acid (2a) in 76% yield (Table 2, entry 1).
Substrates with electron-donating groups were transformed
into the corresponding SA derivatives (2b,c,e,h) in good to
excellent yields, whereas those possessing electron-neutral
and electron-deficient substituents were converted into the
desired products (2d,i,j) in slightly diminished yields.
Remarkably, a number of useful functionalities, including
ester (product 2s), nitrile (product 2t), aryl chloride (products
2g,k), alkyl chloride (product 2u), cyclopropyl (product 2r),
À
the key intermediate D generated upon C H activation of 1b
and migratory insertion of CO undergoes reductive elimina-
tion to produce the 18O-containing product 2b’-18O.[13]
Finally, this novel carboxylation method was tested on
more complex phenols. Thus, estrone was first converted into
its silanol derivative 1aa in almost quantitative yield. The
latter underwent
sequence to produce 2aa as a single isomer in 89% yield
(Scheme 3a). Considering the importance of multisubstituted
a
smooth carboxylation/desilylation
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
These are not the final page numbers!