news & views
may expand the potential of these
systems for applications in technologies
such as molecular machinery, chemical
biology, supramolecular catalysis and
information storage.
Guido H.Clever is at the Institute for Inorganic
Chemistry, Georg-August University Gttingen,
Tammannstr. 4, 37077 Gttingen, Germany.
e-mail: gclever@gwdg.de
FRUSTRATED LEWIS PAIRS
A metal-free landmark
The synthesis and isolation of a silane adduct of an electrophilic boron species provides insight into the mechanism
of metal-free catalytic reductions based on frustrated Lewis pairs.
Douglas W. Stephan
T
he production of many essential
materials, from polymers and
foodstuffs to pharmaceuticals and
agrochemicals, relies on the hydrogenation
of C=C, C=O or C=N double bonds.
Alternatively, hydrosilylation of such double
bonds provides essential components for
sealants, adhesives, lubricants, cookware and
insulation. Indeed, the reduction of multiple
bonds is the most commonly used chemical
reaction worldwide.
The technology to effect hydrogenations
was initially uncovered almost a century ago,
when Sabatier discovered that amorphous
heterogeneous metal catalysts could mediate
the reduction of organic substrates1. In the
past 50years, organometallic chemists have
explored numerous discrete homogeneous
transition metal-based catalysts2. While
these advances have provided highly
specific and selective catalyst systems, the
frequent dependence of current technologies
on expensive and toxic precious metals
has spawned efforts targeting catalysts
derived from Earth-abundant elements.
To this end, the dogma that links small-
molecule activation to transition metals
has led to recent discoveries of Fe, Co and
Ni-based catalysts3.
A unique strategy to achieve catalysis
with Earth-abundant elements is based on
frustrated Lewis pairs (FLPs)4, which are
combinations of main group Lewis acids
and Lewis bases that, by virtue of their
sterics and/or electronics, do not form
classical Lewis acidbase adducts. These
reagents act in concert to effect both the
activation of small molecules (such as
hydrogen) and the catalytic reduction of
organic substrates5.
952 NATURE CHEMISTRY | VOL 6 | NOVEMBER 2014 | www.nature.com/naturechemistry
References
1. Ayme, J.-F., Beves, J.E., Campbell, C.J. & Leigh, D.A. Chem. Soc.
Rev. 42, 17001712 (2012).
2. Forgan, R.S., Sauvage, J.-P. & Stoddart, J.F. Chem. Rev.
111, 54345464 (2011).
3. Fujita, D., Takahashi, A., Sato, S. & Fujita, M. J.Am. Chem. Soc.
133, 1331713319 (2011).
4. Yamashita, K.-I., Kawano, M. & Fujita, M. J.Am. Chem. Soc.
129, 18501851 (2007).
5. Leigh, D.A., Pritchard, R.G. & Stephens, A.J. Nature Chem.
6, 978982 (2014).
6. Hiraoka, S., Yamauchi, Y., Arakane, R. & Shionoya, M.
J.Am. Chem. Soc. 131, 1164611647 (2009).
C6F5 F C6F5 F
F EtSiH
C6F5 C6F5
B F B F
C6F5
C6F5 F H F
Et3Si
1 2
C6F5 F
[Ph3PNPPh3]Cl -Et3SiCl
F
C6F5 [Ph3PNPPh3]
B F
C6F5
H F
3
Figure 1 | The long-awaited isolation and crystallographic characterization of a boranesilane adduct 2
provides insight into the mechanism of reductions catalysed by frustrated Lewis pairs. Right: crystal
structure of 2. Atom colours: C, dark grey; F, lime green; B, pink; Si, light green; H, light grey.
While the notion of FLPs was articulated
following the 2006 discovery of their ability
to activate H2 (ref.6), it is important to
recognize that several pertinent conceptual
precedents predate this work. For example,
in 1942, Brown7 recognized that sterically
demanding combinations of Lewis acids and
bases do not form classical Lewis acidbase
adducts, whereas in the 1950s and 1960s
Wittig 8 and Tochtermann9 observed several
examples of unexpected reactivity arising
from the combination of Lewis acids and
bases with organic substrates. Perhaps most
pertinent to the concept of FLP catalysis
was work from the Piers group10 in 1996
describing the hydrosilylation of ketones
and imines using tris(pentafluorophenyl)
borane as a catalyst. Importantly, Piers
showed that while classic Lewis acidbase
adducts between the ketone and borane
2014 Macmillan Publishers Limited. All rights reserved
7. Mohr, B., Sauvage, J.-P., Grubbs, R.H. & Weck, M. Angew. Chem.
Int. Ed. 36, 13081310 (1997).
8. Ayme, J.-F. etal. Nature Chem. 4, 1520 (2011).
9. Hasenknopf, B. etal. J.Am. Chem. Soc. 119, 1095610962 (1997).
10. Oleksi, A. etal. Angew. Chem. Int. Ed. 45, 12271231 (2006).
11. Nguyen, T.D. etal. Proc. Natl Acad. Sci. USA
102, 1002910034 (2005).
12. Neal, E.A. & Goldup, S.M. Chem. Commun.
50, 51285142 (2014).
13. Hardy, J.G. Chem. Soc. Rev. 42, 78817899 (2013).
Published online 21 September 2014
F
are thermodynamically stable, it is the
dissociation of these adducts that makes the
borane available to activate the SiH bond
of the silane. This then prompts backside
attack of the silane by the carbonyl fragment
leading to hydrosilylation. At the time,
Piers supported this mechanistic postulate
with kinetic and isotopic labelling data10.
Subsequently, ingenious experiments by
Oestreich11 exploited the stereochemistry at
silicon to provide further support for what is
now referred to as an FLP-type activation of
silane, as both the Lewis acid (borane) and
Lewis base (ketone) act on the SiH bond.
Now, writing in Nature Chemistry,
Piers, Tuononen and co-workers12 report
a landmark finding they have prepared,
isolated and characterized the first silane
adduct of an electrophilic boron species,
illuminating and affirming the details of this

metal-free mechanism. Using the previously
reported electrophilic boraindene 1 (ref.13;
Fig.1), Piers, Tuononen and co-workers
demonstrate its ability to form an adduct
with triethylsilane. While the adduct is not
observed spectroscopically in solution at
room temperature, variable-temperature
nuclear magnetic resonance experiments
following a fluorine in the boraindene reagent
and a proton in the silane clearly indicate
the formation of an adduct, 2, on cooling
to 195K. The thermodynamic parameters
governing the equilibrium were found to be
consistent with a thermoneutral situation at
room temperature. The boraindenesilane
adduct 2was also isolated in the solid state
and a crystallographic study affirmed its
formulation and revealed the metric details
of the SiHB contact (SiH: 1.51(2) ,
BH: 1.46(2) , SiHB angle: 157).
Computational studies were used to
compare the stability of 2 to the silane
adducts of tris(pentafluorophenyl)borane
and a perfluoroarylborole (C6F5)4C4B(C6F5).
Based on an analysis of the geometries of
these adducts and the computed change in
Gibbs free energy and enthalpy for bond
METALORGANIC FRAMEWORKS
3D frameworks from 3D printers
High-throughput screening of solvothermal crystallization conditions for MOFs and other solids may receive a
boost from the application of 3D printing techniques to low-cost, disposable pressure vessels.
Ian D. Williams
I
n MOF (metalorganic framework)
chemistry, solvothermal synthesis methods
are widely applied1. Often crystallization
at elevated temperatures is the only means
of generating robust frameworks from a
given metal and ligand system. Traditionally,
solvothermal synthesis is carried out, on
scales of tens of milligrams to a few grams,
in Teflon cups. These are then inserted
into stainless steel pressure vessels. Soon
after the field of MOF synthesis emerged
around two decades ago, it was quickly
realized that variation of reaction conditions
stoichiometry, solvent, concentration,
temperature, time and pH can lead to
numerous different product phases.
The application of high-throughput
screening of conditions for MOF synthesis
was advocated by Cheetham and others
around a decade ago2. However, the cost
of individual steel pressure vessels (about
$500 each) is not insubstantial, especially
when numerous conditions need to be used
NATURE CHEMISTRY | VOL 6 | NOVEMBER 2014 | www.nature.com/naturechemistry 953
formation, it appears that the Lewis acidity
of 1 lies between that of the other two
boron species considered, and it is a subtle
balance of steric and electronic features that
accounts for the stability of 2.
Piers, Tuononen and co-workers go on
to show that addition of the nucleophile
(Ph3PNPPh3)Cl to mixtures of 1and silane
results in the formation of silyl chloride and
(Ph3PNPPh3)((C6F4)C2(C6F5)2BH(C6F5)) 3
(Fig.1). A direct analogy can be drawn
between this reaction and the immediate
reaction of silaneboron species with
ketones that leads to hydrosilylation via this
FLP-type mechanism.
In summary, exploiting the high Lewis
acidity and tempered steric environment
about the boron centre in 1 has enabled
interception of 2 a species that
illuminates the intimate details of the
first step in boronLewis acid-mediated
hydrosilylation of ketones. This finding
has significant implications for the role of
other Lewis acids in hydrosilylation catalysis
and more generally for the mechanism of
activation of a range of small molecules by
FLPs. Understanding the molecular details
for such a screen. Construction of reactor
arrays for the purpose has already been
demonstrated to be effective, but these
bespoke systems are even more costly.
A recent report in Angewandte Chemie
International Edition by Leroy Cronin,
Ross Forgan and colleagues3 is a potentially
exciting development for many
researchers, including those working on
high-temperature crystallizations of MOFs.
The paper describes how an inexpensive
array of hydrothermal reaction chambers
can be 3D printed in polypropylene (reactor
R3 in Fig.1a). Loading these chambers
with reagents allows a number of individual
reactions to be set up simultaneously for
exploration of crystallization conditions. The
Cronin group has previously championed 3D
printing for use in custom labware4,5, and for
these reactors, the economics of production
seem attractive after an initial outlay on
the 3D printer itself. Polypropylene is a
cheap thermoplastic ($25 per kg) and each
2014 Macmillan Publishers Limited. All rights reserved
news & views
of the mechanism of action is essential to
the design and discovery of new and more
efficient metal-free catalysts.
Douglas W.Stephan is in the Department of
Chemistry, University of Toronto, 80St George
Street, Toronto, Ontario M5S 3H6, Canada.
e-mail: dstephan@chem.utoronto.ca
References
1. Sabatier, P. Ind. Eng. Chem. 18, 10051008 (1926).
2. Hudlick, M. Reductions in Organic Chemistry Vol. 2 (American
Chemical Society, 1996).
3. Bullock, R.M. (ed.) Catalysis without Precious Metals (Wiley-
VCH, 2010).
4. Stephan, D.W. & Erker, G. Angew. Chem. Int. Ed.
49, 4676 (2010).
5. Stephan, D.W. & Erker, G. Top. Curr. Chem. 332, 85110 (2013).
6. Welch, G.C., Juan, R.R. S., Masuda, J.D. & Stephan, D.W.
Science 314, 11241126 (2006).
7. Brown, H.C., Schlesinger, H.I. & Cardon, S.Z. J.Am. Chem. Soc.
64, 325329 (1942).
8. Wittig, G. & Benz, E. Chem. Ber. 92, 19992013 (1959).
9. Tochtermann, W. Angew. Chem. Int. Ed. Engl. 5, 351371 (1966).
10. Parks, D.J. & Piers, W.E. J.Am. Chem. Soc.
118, 94409441 (1996).
11. Rendler, S. & Oestreich, M. Angew. Chem. Int. Ed.
47, 59976000 (2008).
12. Houghton, A.Y., Hurmalainen, J., Mansikkamki, A., Piers, W.E.
& Tuononen, H.M. Nature Chem. 6, 983988 (2014).
13. Houghton, A.Y., Karttunen, V.A., Piers, W.E. &
Tuononen,H.M. Chem. Commun. 50, 12951298 (2014).
individual reactor uses only around 20g
of material, making the cost of production
much more economically reasonable than
traditional steel pressure vessels.
Choosing a thermoplastic with suitable
characteristics for solvothermal reaction
vessels, but that can still be 3D printed,
presents an engineering challenge. Many
contemporary 3D printers use either
hydrolytically biodegradable polylactic
acid, or acrylonitrile-butadiene which is
soluble in many common organic solvents
both of which would be unsuitable
for solvothermal crystallizations. The
team behind the research has shown
that the polypropylene vessels can,
perhaps surprisingly, withstand the
rigors of conditions such as aqueous
dimethylformamide up to 140C for
several days, as well as the presence of
acids and bases.
Synthesis of MOFs using the reactors
described by Cronin, Forgan and co-workers