Communications
DOI: 10.1002/anie.200702908
H2 Activation
Metal-Free Catalytic Hydrogenation**
Preston A. Chase, Gregory C. Welch, Titel Jurca, and Douglas W. Stephan*
Hydrogenation is the addition of hydrogen to unsaturated
organic compounds. Such reactions are used for the produc-
tion of a myriad of chemical products worldwide, from large-
scale operations including the upgrading of crude oil and the
production of bulk commodity materials to the synthesis of a
variety of fine chemicals used in the food, agricultural, and
pharmaceutical industries.[1] The process of hydrogen addition
to unsaturated precursors is mediated by either homogeneous
or heterogeneous transition-metal-based catalysts.[1] Blaser et
al.[2] place the importance of this chemistry in context, stating
that “hydrogen is the cleanest reducing agent and hydro-
genation is arguably the most important catalytic method in
synthetic organic chemistry both on the laboratory and the
production scale”. Historically, hydrogenation began with
Sabatierꢀs 1897 discovery that traces of nickel could mediate
the catalytic hydrogenation of olefins and culminated in a
share of the 1912 Nobel Prize with Grignard. In the 1960s, the
advent of organometallic chemistry gave rise to homogeneous
transition-metal-based hydrogenation catalysts for a variety
of substrates. The operation of these catalysts hinges on the
key step of oxidative addition of hydrogen.[3] More recently,
transition-metal systems that effect heterolytic cleavage of
hydrogen at a metal center have been uncovered. In these
cases, a metal hydride is formed with concurrent protonation
of an amido ligand.[4,5]
adducts.[13] Mixtures of “frustrated” phosphines and boranes
can heterolytically cleave H2, forming phosphonium borates
of the form [R3PH][BHR’3].[14] In very recent work, Bertrand
and co-workers have demonstrated that selected carbenes
exhibit transition-metal-like reactivity and cleave hydrogen
or ammonia to effect a formal oxidative addition of the
carbene C atom.[15] Last year, we reported the only nonmetal
system known to reversibly activate and liberate H2. The
phosphonium borate (2,4,6-Me3C6H2)2PH(C6F4)BH(C6F5)2
(1) is formed by reaction of the phosphine–borane species
(2,4,6-Me3C6H2)2P(C6F4)B(C6F5)2 (3) with H2 while heating of
the zwitterion 1 above 1008C liberates hydrogen and regen-
erates the phosphine–borane 3.[16] Herein, we demonstrate
that this system and a related system provide the first metal-
free hydrogenation catalysts that effect the addition of
molecular H2 to imines, nitriles, and aziridines to produce
primary and secondary amines in high yields under relatively
mild reaction conditions.
The reduction of imines and nitriles is one of the best
synthetic methods to generate secondary and primary amines,
and has found tremendous importance in the pharmaceutical
and fine chemicals industry.[17–20] The air- and moisture-stable
phosphonium borates (R2PH)(C6F4)BH(C6F5)2 (R = 2,4,6-
Me3C6H2 (1)[16] and tB u 2()[13]) are active catalysts for the
À
hydrogenation of C N multiple bonds with H2. For example,
Non-transition-metal catalysts for hydrogenation reac-
tions are all but unknown. KOtBu has been shown to act as a
catalyst effecting the addition of H2 to benzophenone under
forcing conditions of 2008C and greater than 100 bar H2.[6]
Organocatalysts have been developed for hydrogenations of
enones and imines; however, such systems do not employ H2
directly but rather a surrogate such as a Hantzsch ester as the
stoichiometric source of hydrogen.[7–11] The development of
nonmetal hydrogenation catalysts hinges on the discovery of
systems that react cleanly with H2, but few are known. Power
and co-workers reported the hydrogenation of Ge2–alkyne
analogues to give a mixture of Ge2 and primary germane
products.[12] Recently we have introduced the concept of
“frustrated Lewis pairs”, bulky Lewis acids and bases which
are sterically precluded from forming simple Lewis
imines are reduced in toluene to the corresponding amines
cleanly and in high yield at slightly elevated temperatures
(80–1408C) and H2 pressures (1–5 atm) in sealed glass bombs
(Table 1, entries 1–6). The amine products are readily sepa-
rated from residual catalyst by filtration through a plug of
silica gel; no other side products are observed in the NMR
spectra of the crude reaction mixtures. In the case of a
sterically less demanding imine (Table 1, entry 6) no catalytic
turnover was noted. Similarly, nitriles are not catalytically
reduced, as these donors intervene in the catalytic cycle by
binding strongly to the Bcenter of the catalyst. Sequestering
=
the N lone pair by coordination of Ph(H)C NCH2Ph to
B(C6F5)3 (Table 1, entry 7) allowed catalytic imine reduction
to proceed. In a similar fashion, by employing the more active
2 as the catalyst, alkyl and aryl B(C6F5)3-bound nitriles are
also successfully reduced and isolated as the corresponding
primary amine–borane adducts (Table 1, entries 8–10).[21] In
these cases, partial reduction of nitriles to the corresponding
imines can not be intercepted or observed. It is noteworthy
that the bis-borane adduct of adiponitrile is also fully reduced
under similar conditions to give the bis-borane adduct of 1,6-
diaminohexane (Table 1, entry 10). Attempts to reduce
similar B(C6F5)3–isonitrile adducts were unsuccessful. How-
ever, catalytic reductive ring opening of an unactivated N-aryl
aziridine functionality is achieved under similar conditions
(Table 1, entry 11).
[*] P. A. Chase, G. C. Welch, T. Jurca, Prof. Dr. D. W. Stephan
Department of Chemistry and Biochemistry
University of Windsor
Windsor, Ontario, N9B 3P4 (Canada)
Fax: (+1)519-973-7098
E-mail: stephan@uwindsor.ca
[**] Financial support from NSERC of Canada is acknowledged. G.C.W.
is grateful for an NSERC scholarship.
Supporting information for this article, including preparative details
for B(C6F5)3–imine and B(C6F5)3–amine adducts as well as for 6 and
from the author.
8050
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8050 –8053