.
Angewandte
Communications
We then investigated the substrate scope of the thioethers.
Firstly, we found that the substituents on the S atom played
a vital role in enhancing the efficiency. Alkyl groups gave very
good yields (1a,b) while aryl groups (1c,d) showed very low
efficiency (Table 3). With the methyl group as a substituent on
carboxylic acids might occur, thus reducing the yield.[24]
However, by either replacing the thioether with benzalde-
hyde or running the reaction in the absence of thioether, the
reaction did not occur, thus indicating the aldehyde was not
the key intermediate and the thioether was essential (Sche-
me 4a,b). The desired product was not obtained from 4 under
Table 3: The effect of substituents on sulfur atom.
S, we further investigated the reactivity of substituted
thioethers (Table 4). Different functional groups, such as F,
Cl, Br, COOMe, OMe, CF3, etc., all tolerated the reaction
conditions and gave the desired products in moderate to good
yields (3ej–gj, 3ha–ja, 3lj). Again, electron-neutral sub-
strates showed the best efficiency. In addition, substrates
having ortho and meta substituents also reacted well (3kj–lj,
3ma).
Scheme 4. Control experiments and intermediate study.
the standard reaction conditions (see Scheme 2b; also see the
Supporting Information). To further investigate the inter-
mediate, we synthesized 9. Indeed, it was transformed into the
desired product in high yield under the standard reaction
conditions (Scheme 4c), thus indicating it might be a key
intermediate in the catalytic cycle. This cyclization step can be
promoted by either the rhodium catalyst or silver salt.[25]
Further efforts showed that the kinetic isotope effect (KIE)
of the carboxylic acid and thioether were 2.0 and 4.9,
Table 4: Substrate scope for thioether derivatives.[a]
À
respectively, thus indicating that C H activation for both
arenes was possibly involved in the rate-determining step (see
the Supporting Information).
Based on these mechanistic studies, we proposed the
mechanism shown in Scheme 5. The first thioether-directed
À
rate-determining first C H bond activation forms the inter-
mediate A. Carboxylic acid coordination and ligand exchange
then forms the intermediate C, from which a relatively facile
À
second C H bond activation occurrs to form the intermediate
À
D. Reductive elimination forms the C C bond coupling
intermediate E and rhodium(I) species. With the assistance of
the rhodium catalyst or silver salt, E is cyclized to the final
product. Reoxidation of rhodium(I) into rhodium(III) by
AgNO3 completes the catalytic cycle. The intermediates A
and D were detected by ESI-HRMS (see the Supporting
[a] Reaction conditions: [{Cp*RhCl2}2] (3.1 mg, 0.005 mmol), AgSbF6
(27.6 mg, 0.08 mmol), AgNO3 (136.0 mg, 0.8 mmol), carboxylic acid (2,
0.2 mmol), and thioether 1 (0.6 mmol) were added to a 50 mL Wattecs
reaction tube before adding solvent (toluene, 2 mL). Then the tube was
sealed and heated to 1608C for 8–12 h. [b] Used 2.4 equiv of 1 f.
[c] Benzyl(ethyl)sulfane (1b).
À
Information). At this stage the carboxylic acid C H activa-
À
tion/thioether C H activation sequence cannot be ruled out.
With 1h as a substrate, we observed small amounts of both
the aldehyde 7 (derived from 1h) and ester 8 (derived from
1h and 2; Scheme 3). As a result, the relatively low efficiency
arises in part from esterification. In addition, in the presence
of the silver salt at high temperature, the decarboxylation of
In summary, we have developed a rhodium-catalyzed
double C H bond coupling reaction between benzyl thio-
À
ethers and aryl carboxylic acids to synthesize dibenzooxepi-
nones. Here, two directing groups were used and they were
either incoorporated into the product or removed in situ after
the reaction. The substrate scope is good and several func-
tional groups could be tolerated. The synthetic application of
this method in useful natural products and further investiga-
tion of the mechanism are underway.
Experimental Section
General procedures: [{Cp*RhCl2}2] (3.1 mg, 0.005 mmol), AgSbF6
(27.6 mg, 0.08 mmol), AgNO3 (136.0 mg, 0.8 mmol), and carboxylic
acid (2a; 24.4 mg, 0.2 mmol) were added to a 50 mLWattecs reaction
tube (sealed tube) under an atmosphere of air. Then the thioether
Scheme 3. Byproduct analysis.
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5478 –5482