lectivity, we conducted the hydrogenolysis of
diphenyl ether and 2-methoxynaphthalene in the
presence of a 300-fold excess of mercury with
respect to the catalyst. No decrease in conversion
or product yields was observed (23). Although
such data are suggestive, rather than fully con-
clusive, of the homogeneity versus heterogeneity
of the catalyst, the absence of poisoning by mer-
cury is consistent with our proposal that the
distinct selectivities result from the action of a
homogeneous catalyst (22).
Fig. 2. (A) Relative reactivity of aryl and benzyl ethers toward hydrogenolysis of C-O bonds catalyzed by
Ni(COD)2 and SIPr. Selective hydrogenolysis of (B) diphenyl ether in the presence of 4-tert-butylbenzyl
methyl ether; (C) 2-methoxynaphthalene in the presence of 4-tert-butylbenzyl methyl ether; and (D)
diphenyl ether in the presence of 4-tert-butylanisole.
ysis of an oligomeric aryl ether. Hydrogenolysis
of the oligomeric phenylene oxide containing
eight aromatic C-O bonds (Fig. 3A) in the presence of 10 mol of the catalyst per C-O bond
under 1 bar of hydrogen at 120°C in m-xylene
resulted in complete depolymerization to give
benzene, phenol, and resorcinol in good total
Lastly, we studied the hydrogenolysis of
various aromatic and benzylic C-O bonds that
constitute roughly 75% of all the intermonomer
linkages in lignin, one of the most stable biopolymers in nature (Fig. 3, B to D) (2, 5). Cleavage of each of these ether linkages with a single
catalyst would illustrate the potential to conduct
the catalytic depolymerization and valorization of
lignin to form aromatic products (4). The diaryl
ether models of one of the most recalcitrant lignin
linkages, the 4-O-5 linkage, were cleaved under
1 bar of hydrogen at 120°C in m-xylene to yield
anisole, benzene, and phenols in moderate yields
in the presence of 20 mol of the catalyst generated from Ni(COD)2 and SIPr⋅HCl (Fig. 3B).
Hydrogenolysis of the a-O-4 lignin model compound proceeded with just 5 mol of the Ni-SIPr
catalyst at 80°C under 1 bar of hydrogen to afford
3,4-dimethoxytoluene and 2-methoxyphenol in
nearly quantitative yields (Fig. 3C). In agreement
with previous results, cleavage of the b-O-4 model under the basic conditions of our system proceeded without catalyst to give guaiacol in 89%
yield (Fig. 3D) (30).
Available data do not allow one to distinguish
between several possible mechanisms for the
nickel-catalyzed hydrogenolysis of aryl ethers,
but several mechanisms can be envisioned. The
nickel center that cleaves the aromatic C-O bond
could be a nickel hydride, a neutral nickel(0)
complex, or an anionic nickelate species. The favorable effect of the strong base in the hydrogenolysis could lead to the formation of anionic
nickel complexes that are more active for the
cleavage of the C-O bonds or for activation of
coordinated dihydrogen. Although these mechanistic issues are yet to be resolved, and turnover
numbers of the catalyst must be improved, the
results from this work have demonstrated that
the selective cleavage of aromatic C-O bonds in
the presence of other C-O bonds can be conducted
without reduction of the arene units by using a
widely available metal and the cheap, mild, and
atom-economical reductant hydrogen.
Fig. 3. Hydrogenolysis of (A) bis(m-phenoxyphenyl)benzene. (B) A model of the 4-O-5 linkage in lignin.
(C) A model of the a–O-4 linkage in lignin. (D) A model of the b–O-4 linkage in lignin.
References and Notes
1. E. Furimsky, Appl. Catal. A. 199, 147 (2000).
2. J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius,
B. M. Weckhuysen, Chem. Rev. 110, 3552 (2010).
3. G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 106,
4. A. L. Marshall, P. J. Alaimo, Chem. Eur. J. 16, 4970
5. Y. Pu, D. Zhang, P. M. Singh, A. J. Ragauskas, Biofuels
Bioprod. Bioref. 2, 58 (2008).
6. D. G. Levine, R. H. Schlosenberg, B. G. Silbernagel,
Proc. Natl. Acad. Sci. U.S.A. 79, 3365 (1982).
7. We define a directing group as a substituent, usually
containing a basic heteroatom, attached to the aromatic
ring that enables functionalization of the arene by