Dihydrogen activation by antiaromatic pentaarylboroles

Article abstract of DOI:10.1021/ja105075h

Facile metal-free splitting of molecular hydrogen (H₂) is crucial for the utilization of H₂ without the need for toxic transition-metal-based catalysts. Frustrated Lewis pairs (FLPs) are a new class of hydrogen activators wherein interactions with both a Lewis acid and a Lewis base heterolytically disrupt the hydrogen-hydrogen bond. Here we describe the activation of hydrogen exclusively by a boron-based Lewis acid, perfluoropentaphenylborole. This antiaromatic compound reacts extremely rapidly with H₂ in both solution and the solid state to yield boracyclopent-3-ene products resulting from addition of hydrogen atoms to the carbons α to boron in the starting borole. The disruption of antiaromaticity upon reaction of the borole with H₂ provides a significant thermodynamic driving force for this new metal-free hydrogen-splitting reaction.

Full text of DOI:10.1021/ja105075h

Published on Web 06/29/2010
Dihydrogen Activation by Antiaromatic Pentaarylboroles
Cheng Fan,† Lauren G. Mercier,† Warren E. Piers,*,† Heikki M. Tuononen,‡ and Masood Parvez†
Department of Chemistry, UniVersity of Calgary, 2500 UniVersity DriVe NW, Calgary, Alberta, Canada T2N 1N4,
and Department of Chemistry, UniVersity of JyVa¨skyla¨, P.O. Box 35, FI-40014 JyVa¨skyla¨, Finland
Received June 10, 2010; E-mail: wpiers@ucalgary.ca
Scheme 1
Abstract: Facile metal-free splitting of molecular hydrogen (H2)
is crucial for the utilization of H2 without the need for toxic
transition-metal-based catalysts. Frustrated Lewis pairs (FLPs)
are a new class of hydrogen activators wherein interactions with
both a Lewis acid and a Lewis base heterolytically disrupt the
hydrogen-hydrogen bond. Here we describe the activation of
hydrogen exclusively by a boron-based Lewis acid, perfluoro-
pentaphenylborole. This antiaromatic compound reacts extremely
rapidly with H2 in both solution and the solid state to yield
boracyclopent-3-ene products resulting from addition of hydrogen
atoms to the carbons R to boron in the starting borole. The
disruption of antiaromaticity upon reaction of the borole with H2
provides a significant thermodynamic driving force for this new
metal-free hydrogen-splitting reaction.
The splitting of the simplest nonpolar molecule, dihydrogen (H2),
that mechanism, the Lewis acidic borane activates the silane toward
is a critical chemical reaction that is most commonly accomplished
nucleophilic attack by the substrate by partially abstracting the silane
by a transition-metal center in homogeneous, heterogeneous, or
hydrogen via a borane-silane adduct related to II. While the
mechanism of B(C6F5)3-catalyzed hydrosilylation is well-estab-
lished, the involvement of II in FLP H2 splitting remains unproven,
biological catalysts via homolytic oxidative addition or heterolytic
processes.1 Recently, interest in more environmentally benign,
transition-metal-free systems for activation of dihydrogen2-4 has
even though computations suggest that II is energetically viable
spiked,5 primarily spurred by the development of the “frustrated
Lewis pair” (FLP) concept.6-8 In FLPs, Lewis acid/base combina-
tions that are sterically prevented from forming strong classical
adducts can heterolytically activate H2. Highly Lewis acidic
perfluoroarylboranes,9,10 such as B(C6F5)3, are typically employed
as the hydride acceptor, while bulky phosphines,11 amines/imines,12,13
or carbenes14,15 serve as the Lewis base proton acceptor.
relative to the reactants.6
Mechanistic details aside, it is clear that a high level of Lewis acidity
at the boron center is required28 in order to achieve hydrogen activation
in these systems; unfluorinated triphenylborane, B(C6H5)3, for example,
is much less effective as an FLP partner.6 Recently, we reported the
synthesis and characterization of perfluoropentaphenylborole (1),29
a
new perfluoroarylborane with exceptional Lewis acid strength as a
consequence of both fluoroaryl substitution and the antiaromaticity of
the four-π-electron borole ring.30 Its reactivity in the context of the
FLP paradigm was therefore worthy of exploration.
The mechanistic details of hydrogen activation by FLPs are still a
subject of debate, although computational investigations point to an
“encounter complex” (I, Scheme 1) stabilized by noncovalent interac-
tions and dispersion forces16,17 that creates an electric field in the pocket
of the FLP where a dative bond would form in a satisfied Lewis acid/
base pair. This electric field polarizes H2, leading to cleavage of the
H-H bond.18 Despite the in silico support for this picture, spectro-
scopic evidence for the encounter complex is lacking.
Borole 1 is sparingly soluble in nondonor solvents, and even
weakly Lewis basic solvents form adducts.29 Halogenated solvents
t
are most useful, but mixtures of 1 and Bu3P in CD2Cl2 exhibit
reactions that involve chloride transfer to 1, indicative of C-Cl
t
bond activation. In C6D5Br, however, 1 and Bu3P do not activate
the solvent, and no indication of conventional adduct formation is
apparent either spectroscopically or visually (the intense color of
pentaarylboroles31 is quenched upon ligation of boron). Exposure
of this mixture to H2, however, resulted in a rapid reaction.
Surprisingly, a mixture of products was observed, and the expected
phosphonium borate ion pair [(C6F5)4C4B(H)C6F5]-[HP(tBu)3]+ (2)
was a minor component (<15%) of the reaction product mixture.
This observation led us to investigate the reactivity of 1 with H2
in the absence of tBu3P. Rapid reaction (less than 1 min) in CD2Cl2,
C6D5Br, or C7D8 was indicated by the decolorization of these
solutions or suspensions; indeed, even exposure of microcrystalline
solid samples of 1 to an atmosphere of H2 resulted in conversion
to an off-white solid within 20 min.
An alternate view involves an adduct between borane and H2
(II) that is related to transition metal-H2 σ complexes.19 Intermedi-
ate II could be deprotonated directly or proceed to III, an
intermediate analogous to protonated fluorobenzenes, via heterolytic
addition of H2 across a B-C bond.20,21 This has been proposed as
the initial step in the addition of H2 to Stephan’s seminal
phosphinoborane hydrogen activation system4 and is supported
computationally.22 This mechanism is conceptually related to that
developed for the B(C6F5)3-catalyzed hydrosilylation of carbonyl23-25
and imine26 functions and the dehydrosilylation of alcohols.27 In
† University of Calgary.
‡ University of Jyva¨skyla¨.
9
9604 J. AM. CHEM. SOC. 2010, 132, 9604–9606
10.1021/ja105075h  2010 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
Figure 1. Thermal ellipsoid diagram (50%) of cis-3. Selected bond
distances (Å): B1-C1, 1.585(6); C1-C2, 1.533(5); C2-C3, 1.326(5);
C3-C4, 1.529(5); B1-C4, 1.586(6). Selected bond angles (deg): C1-B1-C5,
124.4(3); C1-B1-C4, 106.2(3); C4-B1-C5, 128.6(4); B1-C1-C11,
115.7(3); B1-C1-C2, 103.1(3); C2-C1-C11, 113.1(3); B1-C4-C29,
124.3(3); B1-C4-C3, 102.8(3); C3-C4-C29, 113.5(3).
The products are the two major species observed in the reaction
performed in the presence of Bu3P, which were identified as the
t
cis and trans isomers of the boracyclopent-3-ene heterocycles 3
that result upon formal addition of hydrogen to the carbons R to
boron in borole 1 (Scheme 2). This was deduced on the basis of
multinuclear NMR spectroscopy, derivatization to the pyridine
adducts 3-py, and X-ray crystallographic characterization of cis-3
and trans-3-py.
to each other on the C4B ring was also evident from the orientation
of the C1 and C4 C6F5 rings on opposite sides of the C4B plane.
The reaction between 1 and H2 in the absence of an external
base shows that 1 is capable of forming a reactive adduct with H2.
DFT computations showed that the LUMO of 1 is associated with
the boron center and the two R carbons (Figure S4), but the low
solubility of 1 has precluded low-temperature NMR experiments
aimed at observing an H2 adduct of 1 spectroscopically. However,
DFT computations located a minimized-energy structure for the
H2 adduct of 1 that is only 0.5 kcal mol-1 less stable than the
reactants (Figure S5). Since the H2 adduct of B(C6F5)3 itself (i.e.,
II) reacts only by dissociation of H2 (unless there is a proton
acceptor on one of the fluorinated aryl rings4), it appears that
disruption of antiaromaticity in the borole ring30 provides a driving
force for the remarkably facile reaction of 1 with H2 to give
compounds 3. The energetic destabilization of four-π-electron five-
The 11B{1H} NMR spectrum of products 3 shows a broad resonance
at 78.5 ( 1.0 ppm, consistent with a three-coordinate borane center
and distinct from the 66.0 ppm resonance associated with 1. The 1H
NMR spectrum (Figure S1 in the Supporting Information) shows two
singlets in a 2:1 ratio at 5.14 and 4.83 ppm, which were assigned to
the trans and cis isomers of 3, respectively, on the basis of the changes
in the spectrum upon addition of pyridine (Figure S1). The signal at
5.13 ppm was split into two equal-intensity peaks at 5.67 and 5.06
ppm for the now inequivalent protons of trans-3-py, while that at 4.82
ppm was transformed into two singlets at 4.09 and 4.98 ppm, the latter
barely observable initially. Over 8 h, this signal grew in until it was
present at half the intensity of the resonance at 4.09 ppm. On the basis
of NOE experiments, the kinetically favored isomer of cis-3-py is that
with pyridine oriented cis to the two R protons (Figure S2). Isomer-
ization to the thermodynamic mixture of cis-3-py isomers occurs by
reversible dissociation and recoordination of pyridine. For the reaction
of 1 with H2 in solution, trans-3 is kinetically favored, but for reactions
of solid 1 with H2, cis-3 is the dominant product by a 10:1 margin.
Density functional theory (DFT) computations showed that trans-3 is
thermodynamically favored by 6.2 kcal mol-1 (Table S1 in the
Supporting Information). Heating solutions of the two isomers to 50
°C in the dark for 12 h had no effect on the kinetic ratios. However,
irradiation of solutions enriched in cis-3 at 254 nm for 4 days resulted
in complete conversion to the more stable trans-3 isomer via an
unknown mechanism.
membered borole rings in comparison with related aromatic systems
33,34
has been estimated to be 10-20 kcal mol-1
;
thus, the
combination of antiaromaticity and high Lewis acidity in 1 leads
to rapid H-H bond activation in the absence of an external Lewis
base partner. Indeed, the extra driving force provided by antiaro-
maticity permits H2 activation in more weakly Lewis acidic
pentaarylboroles: the reaction of unfluorinated pentaphenylborole
(4)31 with H2, although slower, produces cis-5 and trans-5 in a
1.0:4.3 ratio over 2-3 h (Scheme 3). Interestingly, no H2 adduct
with 4 could be found by DFT calculations, suggesting that in this
case H2 binding to the less Lewis acidic boron center may be rate-
limiting.
The structures of cis-3 and trans-3-py were confirmed by X-ray
crystallography.32 A thermal ellipsoid diagram of the former
compound is shown in Figure 1 along with selected metrical
parameters; that of the latter is given in Figure S3. The C4B ring
in cis-3 features a trigonal-planar boron center [sum of angles )
359.2(6)°] and a CdC double bond between C2 and C3 [1.326(5)Å].
The hydrogen atoms on C1 and C4 were located on the difference
map and their positions refined: the C1 and C4 carbons are clearly
pyramidalized [the sums of non-hydrogen angles about C1 and C4
are 331.9(5) and 340.6(5)°, respectively], and the R-carbon C6F5
rings lie below the C4B plane. Although the trans-3 isomer can be
produced in pure form photochemically, suitable crystals were not
obtained; instead, this isomer’s structure was confirmed via
characterization of its pyridine adduct. The hydrogen atoms on C1
and C4 were again located and refined, and their positioning trans
Scheme 3
Attempts to reverse the addition of H2 to Lewis acidic borole 1
thermally or photochemically were unsuccessful, and no deuterium
incorporation into compounds 3 was observed under any conditions
upon exposure of their solutions to 4 atm D2. Interestingly, when
mixtures of cis/trans-3 were treated with Bu3P (1 equiv per boron),
conversion to the phosphonium borates 2 and 2′ occurred (Scheme 4).
t
9
J. AM. CHEM. SOC. VOL. 132, NO. 28, 2010 9605

C O M M U N I C A T I O N S
Scheme 4
an NSERC Graduate Scholarship (to L.G.M), an Alberta Innovates
Studentship (to L.G.M.), the Academy of Finland, and the
University of Jyva¨skyla¨.
Supporting Information Available: Crystallographic data for cis-3
and trans-3-py (CIF) and additional experimental, spectroscopic, and
computational details. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
Isomer 2′ is the thermodynamic product of this reaction; pure
samples exhibit 1H NMR spectral signature resonances for the P-H
(1) Hartwig, J. F. Organotransition Metal Chemistry: From Bonding to
Catalysis; University Science Books: Mill Valley, CA, 2010.
(2) Frey, G. D.; Lavallo, V.; Donnadieu, B.; Schoeller, W. W.; Bertrand, G.
Science 2007, 316, 439–441.
(3) Zhu, Z.; Wang, X.; Peng, Y.; Lei, H.; Fettinger, J. C.; Rivard, E.; Power,
P. P. Angew. Chem., Int. Ed. 2009, 48, 2031–2034.
(4) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W. Science 2006,
314, 1124–1126.
(5) Kenward, A. L.; Piers, W. E. Angew. Chem., Int. Ed. 2008, 47, 38–41.
(6) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46–76.
(7) Stephan, D. W. Org. Biomol. Chem. 2008, 6, 1535–1539.
(8) Stephan, D. W. Dalton Trans. 2009, 3129–3136.
(9) Piers, W. E. AdV. Organomet. Chem. 2005, 52, 1–76.
(10) Piers, W. E.; Chivers, T. Chem. Soc. ReV. 1997, 26, 345–354.
(11) Welch, G. C.; Stephan, D. W. J. Am. Chem. Soc. 2007, 129, 1880–1881.
(12) Sumerin, V.; Schulz, F.; Nieger, M.; Leskela¨, M.; Repo, T.; Rieger, B.
Angew. Chem., Int. Ed. 2008, 47, 6001–6003.
1
1
(5.02 ppm, JPH ) 426 Hz) and C-H (broad, 7.16 ppm, JCH
)
149.3 Hz) protons. Furthermore, 2′ exhibits a resonance at 169.9
ppm in the 13C NMR spectrum (1:1:1:1 quartet, 1JCB ) 56 Hz) and
resonances for four inequivalent C6F5 groups in the 19F NMR
spectrum in the expected 2:1:1:1 ratio. It is likely that this reaction
is initiated by direct deprotonation of a benzylic proton in boracycles
3 by the phosphine base rather than reversible formation of the H2
adduct of 1 from 3. Nonetheless, conversion of cis/trans-3 to
hydrido borate 2 suggests a possible H2 delivery pathway via this
ion pair6 using catalytic amounts of a bulky Lewis base.
Scheme 5
(13) Chase, P. A.; Jurca, T.; Stephan, D. W. Chem. Commun. 2008, 1701–1703.
(14) Holschumacher, D.; Bannenberg, T.; Hrib, C. G.; Jones, P. G.; Tamm, M.
Angew. Chem., Int. Ed. 2008, 47, 7428–7432.
(15) Chase, P. A.; Stephan, D. W. Angew. Chem., Int. Ed. 2008, 47, 7433–
7437.
(16) Kim, H. W.; Rhee, Y. M. Chem.sEur. J. 2009, 15, 13348–13355.
(17) Rokob, T. A.; Hamza, A.; Stirling, A.; Soo´s, T.; Pa´pai, I. Angew. Chem.,
Int. Ed. 2008, 47, 2435–2438.
(18) Grimme, S.; Kruse, H.; Goerigk, L.; Erker, G. Angew. Chem., Int. Ed. 2010,
49, 1402–1405.
(19) Kubas, G. J. Metal Dihydrogen and σ-Bond Complexes: Structure, Theory
and ReactiVity; Kluwer Academic/Plenum Publishers: New York, 2001.
(20) Olah, G. A.; Mo, Y. K. J. Org. Chem. 1973, 38, 3212–3221.
(21) Szulejko, J. E.; Hrusa´k, J.; McMahon, T. B. J. Mass Spectrom. 1997, 32,
494–506.
(22) Rajeev, R.; Sunoj, R. B. Chem.sEur. J. 2009, 15, 12846–12855.
(23) Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440–9441.
(24) Parks, D. J.; Blackwell, J. M.; Piers, W. E. J. Org. Chem. 2000, 65, 3090–
3098.
(25) Blackwell, J. M.; Morrison, D. J.; Piers, W. E. Tetrahedron 2002, 58, 8247–
8254.
In summary, we have reported a facile metal-free hydrogen splitting
reaction at Lewis acidic, antiaromatic pentaarylborole boron centers.
The details of the mechanism of the reaction are yet to be determined,
but the presence of the trans isomers of 3 and 5 as the major isomers
in solution suggests that the H2 adducts under go B-CR bond cleavage
followed by rapid cyclization to a mixture of boracyclopent-3-ene
products (Scheme 5). Photochemically generated cis-1,3-butadienylbo-
ranes similar to those depicted in Scheme 5 have been shown to rapidly
cyclize to boracyclopent-3-enes.35,36 That this reaction occurs so rapidly
in the absence of a frustrated Lewis base partner has implications for
the mechanism of H2 splitting by FLPs. Kinetic, thermodynamic, and
computational investigations that will address these issues in detail
are underway; the greater solubility of unfluorinated pentaphenylborole
4 and the more forgiving time scale of its reaction with H2 make it
ideal for further study.
(26) Blackwell, J. M.; Sonmor, E. R.; Scoccitti, T.; Piers, W. E. Org. Lett. 2000,
2, 3921–3923.
(27) Blackwell, J. M.; Foster, K. L.; Beck, V. H.; Piers, W. E. J. Org. Chem.
1999, 64, 4887–4892.
(28) Rokob, T. A.; Hamza, A.; Pa´pai, I. J. Am. Chem. Soc. 2009, 131, 10701–
10710.
(29) Fan, C.; Piers, W. E.; Parvez, M. Angew. Chem., Int. Ed. 2009, 48, 2955–
2958.
(30) Braunschweig, H.; Fernandez, I.; Frenking, G.; Kupfer, T. Angew. Chem.,
Int. Ed. 2008, 47, 1951–1954.
(31) Eisch, J. J.; Hota, N. K.; Kozima, S. J. Am. Chem. Soc. 1969, 91, 4575–
4577.
(32) Full crystallographic data for compounds cis-3 and trans-3-py can be
obtained free of charge from the Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif under reference nos. CCDC
777075 and 777076, respectively.
(33) Mo, Y.; Schleyer, P. v. R. Chem.sEur. J. 2006, 12, 2009–2020.
(34) De Proft, F.; Geerlings, P. Chem. ReV. 2001, 101, 1451–1464.
(35) Zweifel, G.; Clark, G. M.; Hancock, K. G. J. Am. Chem. Soc. 1971, 93,
1308–1309.
(36) Zweifel, G.; Backlund, S. J.; Leung, T. J. Am. Chem. Soc. 1977, 99,
5192–5194.
Acknowledgment. This work was supported by the Natural
Sciences and Engineering Research Council (NSERC) of Canada,
JA105075H
9
9606 J. AM. CHEM. SOC. VOL. 132, NO. 28, 2010