Journal of the American Chemical Society
Communication
ring could be incorporated, with trivial differences in catalytic
efficiencies. Furthermore, aryl bromides installed on hetero-
cycles such as indole (48), furan (49), thiophene (50), and
pyridines (51−54) could also be employed as effective
coupling partners. A broad range of functionalities could be
well-tolerated, including halogen atoms (30, 31, 34−37) and
silane (41), ether (33, 38, 47, 53), and ketal groups (43).
Interestingly, the boronic ester functional handle in 44 was
kept intact during this synergistic cross-coupling. Moreover, β-
bromostyrene could be employed for dehydroxymethylative
vinylation under the same set of reaction conditions, although
alkene isomerization was observed in the products.
Encouraged by the generality of this dehydroxymethylative
cross-coupling, we next carried out mechanistic investigations
to elucidate the enabling effect of the benzoates in promoting
cerium/nickel synergistic catalysis. Control experiments have
demonstrated the critical role of benzoate salts. During
optimization, a 3.5:1 benzoate/cerium loading was found to
be optimal, and increased loadings of benzoate would shut
down the dehydroxymethylative arylation. Moreover, the ortho
substitution effect revealed in the optimization study clearly
precludes the benzoate being a mild or soluble base. We were
intrigued by the fact that aromatic carboxylic acids were
previously demonstrated to be viable LMCT substrates and
converted into lactones in the oxidative decarboxylation
developed by the Mashima group.5 Although in their seminal
report, cerium(IV) tert-butoxide was employed as a precatalyst
and an oxo-cerium(IV) cluster was demonstrated as the active
intermediate, this promoted us to investigate whether the
benzoate was consumed in this dehydroxymethylation
reaction. Even after 36 h of irradiation under our conditions,
benzoic acid derived from L4 could be almost completely
recovered (93%) after acidification of the reaction mixture, and
no corresponding lactone could be detected using gas
chromatography−mass spectrometry (GC−MS). (See Figure
different patterns of photoactivity, further demonstrating the
intriguing potential and underexploited properties of cerium
catalysis. Importantly, both steady-state and time-resolved
emission quenching experiments revealed the linear photo-
luminescence quenching of the cocatalyst DPA by cerium(III)
complexes.14 As shown in the Stern−Volmer plots, CeL3
alkoxide complexes exhibit higher quenching efficiency
compared with CeL3 complexes at the same concentration,
which indicates a favorable SET between photoexcited DPA
(E* = 1.19 V vs SCE in DMSO) and CeL3 alkoxide for the in
situ generation of Ce(IV)L3(OR) complexes.
In light of these experiments, a synergistic catalytic cycle was
proposed. The coordination of free alcohol with a cerium
benzoate complex followed by photoinduced electron transfer
with DPA generates a photoactive Ce(IV) alkoxide species.
Under LED irradiation, one electron of the higher lying
alkoxide ligand orbital will be promoted to the empty 4f orbital
of cerium, resulting in the homolysis of the Ce−O bond. The
bond homolysis leads to the generation of Ce(III) benzoate
and alkoxy radicals, which would undergo β-scission to form
alkyl radicals. Meanwhile, in the nickel catalytic cycle, low-
valent Ni complexes undergo oxidative addition to generate
Ni(II) aryl complexes. The radical interception with the Ni(II)
aryl complex forms a Ni(III) species that undergoes rapid
reductive elimination to deliver the desired dehydroxymethy-
lative arylation product. The SET events enabled and
coordinated by the DPA cycle, including the reduction of
Ni(I) by a radical anion of DPA and the oxidation of Ce(III)
by excited DPA, have furnished cerium and nickel catalytic
cycles and, more importantly, have established a synergy
between radical generation and conversion.
In summary, the photocatalytic dehydroxymethylative
arylation of free alcohols with aryl halides has been achieved
under mild and practical conditions. Enabled by the synergistic
utilization of cerium and nickel catalysts, the α-C(sp3)−C(sp3)
bond of free alcohols can be selectively cleaved and harnessed
as an unconventional synthon in cross-couplings. Sterically
encumbered benzoate ligands have enabled the incorporation
of cerium catalysts in metallaphotoredox catalysis, providing
intriguing opportunities for the exploitation of abundant
cerium catalysts in sustainable synthesis.
The crystallizations of cerium trichloride with benzoate
ligand L7 resulted in a crystallized compound 56 (Figure 3).
X-ray analysis of 56 revealed that three benzoate ligands were
ligated to the cerium center in the κ2 chelating mode with the
Ce−O bond length ranging from 2.501 to 2.613 Å. Three
solvent molecules of DMSO were also found ligated to the
cerium(III) center. The steric shielding around the cerium
center caused by ortho-substitution groups of the benzoates
probably helped to prevent the formation of dimeric or
polymeric structures that typically prevail in lanthanide
carboxylate complexes,13 rendering multiple open coordination
sites for the in situ ligation of alkoxide. In cyclic voltammogram
measurements, complex 56 showed an irreversible redox wave
in CH3CN/DMSO solution (Ep/2 = 0.46 V vs SCE). Because
of the difficulties in isolating Ce(IV) compounds,7d we were
only able to conduct 1H NMR spectroscopic analysis on the in
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
■
sı
Experimental procedures and compound characteriza-
Accession Codes
CCDC 2056187 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
1
situ formed Ce(IV) carboxylate complexes. (The H NMR
spectrum reveals an approximate carboxylate/Ce ratio of 3:1;
Under the standard reaction conditions, cerium complex 56
demonstrated identical catalytic efficiency compared to the
addition of cerium trichloride with L7. This finding together
with the optimal 3.5:1 ligand/cerium loading that we identified
during the optimization lends support for the in situ formation
of CeL3 alkoxide complexes. The distinct coordination pattern
we observed, in comparison with the oxo-cerium cluster
discovered by the Mashima group, would have resulted in
AUTHOR INFORMATION
Corresponding Author
■
Zhiwei Zuo − State Key Laboratory of Organometallic
Chemistry, Shanghai Institute of Organic Chemistry, Chinese
Academy of Sciences, Shanghai 200032, China; orcid.org/
4900
J. Am. Chem. Soc. 2021, 143, 4896−4902