An aromatic ion platform for enantioselective Br?nsted acid catalysis

Article abstract of DOI:10.1126/science.aad0591

Chiral acid catalysts are useful for the synthesis of enantioenriched small molecules, but the standard catalysts require laborious and expensive preparations. Here, we describe a chiral Br?nsted acid prepared in one step from naturally occurring (-)-menthol and readily available 1,2,3,4,5-pentacarbomethoxycyclopentadiene. Aromatic stabilization serves as a key contributing factor to the potent acidity of the resulting compound, which is shown to catalyze both Mukaiyama-Mannich and oxocarbenium aldol reactions with high efficiency and enantioselectivity. Catalyst loadings as low as 0.01 mole percent and preparative scalability (25 grams) are demonstrated. Alternative amide catalysts are also shown to be promising platforms. In addition to proton catalysis, a chiral anion pathway is demonstrated to be viable with this catalyst system.

Full text of DOI:10.1126/science.aad0591

RESEARCH
| REPORTS
27. A. Vanossi, N. Manini, M. Urbakh, S. Zapperi, E. Tosatti,
Rev. Mod. Phys. 85, 529–552 (2013).
(European Cooperation in Science and Technology) Action
MP1303; the European Commission, under the Graphene Flagship
(award no. CNECT-ICT-604391); the U.S. Office of Naval
Research Basic Research Challenge program; and Comunidad
de Madrid, under the MAD2D-CM (S2013/MIT-3007) project.
The Partnership for Advanced Computing in Europe (project
2012071262) and the Empa high-performance computing
facility Hypatia are acknowledged for computational resources.
All data are tabulated in the main text and the supplementary
materials.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/351/6276/957/suppl/DC1
Materials and Methods
Supplementary Text
Figs. S1 to S16
References (28–44)
ACKNOWLEDGMENTS
This work was supported in part by the Japan Science and
Technology Agency PRESTO program, under the project
“Molecular technology and creation of new function;” the National
Center of Competence in Research Nanoscale Science
program; grant CRSII2 136287/1 from the Swiss National
Science Foundation; the Swiss Nanoscience Institute; COST
3 September 2015; accepted 25 January 2016
10.1126/science.aad3569
ORGANIC CHEMISTRY
cessible and effective enantioselective variants.
Herein, we demonstrate the realization of this
vision with the use of chiral PCCPs as highly ef-
ficient enantioselective Brønsted acid catalysts.
1,2,3,4,5-Pentacarbomethoxycyclopentadiene
is readily available via the reaction of dimethyl
malonate (4) and dimethylacetylene dicarbox-
ylate (5) in pyridine-acetic acid, followed by
treatment with potassium acetate (KOAc) to
yield, after acidic workup, the acid 6 (26) (Fig.
2A). Compound 6 is a stable crystalline solid,
and we have scaled this procedure to produce
more than 50 g of this material. To access chiral
derivatives, we found that refluxing 6 with excess
(–)-menthol in the presence of N-methylimidazole
in toluene furnished the pentamenthyl ester 7
in 95% yield. Because (–)-menthol is a naturally
occurring commodity chemical, we calculate that
catalyst 7 can be prepared for about US$4/g
(<$5.50/g for the unnatural enantiomer). As an
alternative means to introduce chirality, we have
also found that treatment of 6 with 1.0 equivalent
of a primary amine (e.g., sec-naphthethylamine)
in refluxing toluene resulted in production of
the corresponding monoamide product 8 in
good yield. Because in the area of Brønsted acid
catalysis, reaction rate has been shown to have
a linear correlation with catalyst acidity (27), at
least to a first approximation, we determined
the pKa (CH3CN) of both pentaester 6 (8.85
0.05) and monoamide 9 (11.7 0.1) and found
that they compared quite favorably with known
BINOL-phosphoric acids 10 (12–14) and other
derivatives (Fig. 2B).
To evaluate the effectiveness of chiral PCCPs
as enantioselective catalysts, we examined their
performance in a Mukaiyama-Mannich reaction,
which was known to be amenable to this type of
asymmetric promotion (7) (Fig. 3A). We observed
that 1 mole percent (mol %) of menthol-derived
acid 7 catalyzed the addition of silyl ketene acetal
12 to imine 11 in ethyl acetate at –78°C in 1 hour
to furnish the adduct 13 in 97% yield and 97%
enantiomeric excess (ee). This result compares
quite favorably to BINOL-phosphoric acid cata-
lyst 14, which at 10 mol % loading was reported
to catalyze the same reaction (in toluene) over
24 hours to furnish 13 in 98% yield and 89% ee.
The loading of cyclopentadienyl catalyst 7 could
be reduced to 0.01 mol % without compromis-
ing enantioselectivity. In terms of further catalyst
screening, monoamides 8 and 15 were also found
to induce appreciable enantioselectivity in this
reaction, suggesting such catalysts may also be
worthy of further development. Nevertheless,
pentaester catalysts have thus far proven optimal
An aromatic ion platform
for enantioselective Brønsted
acid catalysis
Chirag D. Gheewala, Bridget E. Collins, Tristan H. Lambert*
Chiral acid catalysts are useful for the synthesis of enantioenriched small molecules,
but the standard catalysts require laborious and expensive preparations. Here, we describe
a chiral Brønsted acid prepared in one step from naturally occurring (–)-menthol and
readily available 1,2,3,4,5-pentacarbomethoxycyclopentadiene. Aromatic stabilization
serves as a key contributing factor to the potent acidity of the resulting compound, which
is shown to catalyze both Mukaiyama-Mannich and oxocarbenium aldol reactions with
high efficiency and enantioselectivity. Catalyst loadings as low as 0.01 mole percent and
preparative scalability (25 grams) are demonstrated. Alternative amide catalysts are also
shown to be promising platforms. In addition to proton catalysis, a chiral anion pathway
is demonstrated to be viable with this catalyst system.
rotonation dramatically alters the reactiv-
ity of a molecule. As such, Brønsted acid
(proton donor) catalysts have long pro-
vided a potent strategy for the acceleration
of a diverse array of chemical transforma-
carbon acid platform for enantioselective Brønsted
acid catalysis that uses aromaticity as its central
acidifying element.
Although carbon acids typically manifest rela-
tively high pKa values (where Ka is the acid disso-
ciation constant), several structural modifications
can substantially increase the propensity of C–H
bonds to undergo ionization by stabilizing the
anionic charge of the conjugate bases (19, 20)
(Fig. 1A). The most common of these modifications
involve s-delocalization (induction) by electro-
negative elements or groups and p-delocalization
(resonance) by conjugated functionality. A third
mode of carbanion stabilization involves a spe-
cial form of resonance known as aromaticity,
epitomized by the dramatic increase (1017)
in acidity of cyclopentadiene versus its acyclic
analog (21) due to the aromatic nature of the
cyclopentadienyl anion. In the case of a 1,2,3,4,5-
pentacarboxycyclopentadiene (PCCP) 1 (Fig. 1B),
the three elements of induction-, resonance-, and
aromaticity-induced acidification conspire to pro-
duce extremely strong carbon acids that rival the
acidity of the mineral acids (22–24). In fact, the
extended conjugation of such substituted cyclo-
pentadienes is such that their acidic protons
reside not on carbon but rather on oxygen as
hydroxyfulvenes (compare 3)(25). With such strong
acidity, we reasoned that PCCPs could serve as
potent Brønsted acid catalysts and that ester or
amide derivatives incorporating simple chiral
alcohols or amines could offer synthetically ac-
P
tions (1, 2). In recent years, the invention of ef-
fective chiral Brønsted acid catalysts has enabled
the development of numerous asymmetric reac-
tions that furnish valuable chemicals in enan-
tioenriched form (3–6). By and large, this area of
enantioselective Brønsted acid catalysis has been
dominated by the binaphthol (BINOL) phosphoric
acid class of catalysts originally developed by
Akiyama (7) and Terada (8). Although the utility
of these catalysts is unquestionable, the major
drawback to substituted BINOL-based catalysts
is the lengthy, laborious, and expensive protocol
required for their synthesis (9), which complicates
catalyst optimization and limits their application
on scale. Although alternative catalysts have been
developed (10–16), the majority of these still rely
on binaphthyl or other unnatural frameworks
[e.g., VAPOL (17) or SPINOL (18)] as the basis
of their chirality. This monocultural reliance on a
single chiral scaffold is thus a major limiting
factor for this important area of catalysis, and the
identification of more readily accessible architec-
tures is an important goal. Here, we describe a
Department of Chemistry, Columbia University, New York, NY
10027, USA.
*Corresponding author. E-mail: tl2240@columbia.edu
SCIENCE sciencemag.org
26 FEBRUARY 2016 • VOL 351 ISSUE 6276 961

RESEARCH
| REPORTS
both in terms of reactivity and enantioselectivity,
as further exemplified by the performance of the
2-phenylcyclohexanol-derived catalyst 16.
To investigate the preparative utility of 7, we
conducted a Mannich reaction on scale to produce
25 g of product 13 using 0.1 mol % of the acid
catalyst (Fig. 2B, entry 1), which resulted in yield
and enantioselectivity comparable to the smaller-
scale reactions. In terms of additional substrate
scope, we found that a variety of aryl or vinyl
imines could be engaged efficiently and with
high selectivities (entries 2 to 7). In addition, an
aliphatic imine substrate, which is typically chal-
lenging for this type of chemistry, could also be
productively engaged (entry 8). An alternative
silyl ketene acetal bearing a cyclohexyl ring was
found to react in high yield and with high se-
lectivity (entry 9). Thus far, reactions that lead
to diastereomeric mixtures have exhibited only
modest selectivities with catalyst 7 (see the sup-
plementary materials, p. S15).
To demonstrate the potential of PCCPs to move
beyond established capabilities, we investigated
the enantioselective addition to oxocarbenium
ions. Incomparisontoiminiumionadditions, enan-
tioselective reactions with oxocarbenium ions are
notoriously difficult to achieve because they typ-
ically lack the traditional organizational elements
(e.g., hydrogen bonding groups) employed by
many asymmetric catalysts. Indeed, only recently
have successful approaches to enantioselective
oxocarbenium ion chemistry begun to appear
[see, e.g., (28–38)]. We probed the capacity of
catalyst 7 to catalyze the addition of silyl ketene
acetal 12 to oxocarbenium ion intermediates 29
generated from salicylaldehyde acetals 27, a so-
called Mukaiyama oxocarbenium aldol (Fig. 2C)
(35). First, we attempted the reaction of benzal-
dehyde dimethyl acetal, which in fact led to
product 30 in high yield but, as expected, with
essentially no enantioselectivity (entry 1). On the
other hand, the dimethyl acetal of salicylaldehyde
led to the production of adduct 31 with an en-
couraging 71% ee (entry 2). Modification to the
diethyl acetal resulted in an appreciable improve-
ment in selectivity to 85% ee for adduct 32 (entry
3), whereas further increasing the steric demand
of the acetal substituent to n-butyl or i-propyl
(entries 4 and 5) led to comparable enantiose-
lectivities for products 33 and 34. An alternative
ketone product 35 could also be obtained in 80%
ee with the use of a silyl enol ether nucleophile
(entry 6). These results demonstrate that chiral
PCCPs have the ability to address one of the
more notable challenges in asymmetric catalysis,
and it is a reasonable expectation that optimiza-
tion of catalyst structure will enable further im-
provements in this area.
Fig. 1. Tuning carbon acids for catalysis. (A) Common strategies for increasing C–H acidity, including
induction, resonance, and aromaticity, the latter epitomized by the relatively high acidity of cyclo-
pentadiene. (B) PCCPs combine all three acidifying elements to produce powerful Brønsted acid catalysts.
Me, methyl; R, generic substituent.
The mechanistic rationale for the operation
of catalyst 7 is shown in Fig. 4A, in which initial
protonation of acetal 36 produces intermediate
salt 37. Addition of the silyl ketene acetal 12
to the highly electrophilic oxocarbenium ion
(compare 38) then leads to intermediate 39. Silyl
transfer to the alcohol produced in the initial ion-
ization step would then furnish the ether product
32 and return the acid 7 to the catalytic cycle.
Fig. 2. Catalyst preparation and characterization. (A) Synthesis of 1,2,3,4,5-pentacarbomethoxycyclo-
pentadiene 6 and chiral derivatives 7 and 8. (B) pKa values measured in acetonitrile. i-Pr, isopropyl;
Ph, phenyl; NMI, N-methylimidazole.
962 26 FEBRUARY 2016 • VOL 351 ISSUE 6276
sciencemag.org SCIENCE

RESEARCH
| REPORTS
To investigate whether proton catalysis was
indeed the operative mode for catalyst 7, we
ran the Mannich reaction of 11 and 12 using
deprotonated 7 as its sodium salt (Fig. 4B). As
expected, no reaction was observed at –78°C,
lending credence to the notion that the catalyst
operates as a Brønsted acid (entry 1). Interest-
ingly, however, when the reaction was performed
at higher temperatures, product formation was
observed, with appreciable enantioenrichment
(entry 2). As there was no acidic catalyst proton
in this case, this result suggests that perhaps a
silylium-catalyzed process was operative, with
the cyclopentadienyl group serving as a chiral
anion. This notion is supported by the fact that
the reaction worked with comparable efficiency
and enantioselectivity even in the presence of
2,6-di-tert-butyl-4-methylpyridine (entry 3). It
is worth noting that chiral anion catalysis has
also proven to be an effective strategy for asym-
metric synthesis (36–38), and so the current
Fig. 3. Exploration of substrate scope. (A) Catalyst screen for enantioselective Mukaiyama-Mannich reaction. (B) Substrate scope for acid-catalyzed Mannich
reaction. (C) Enantioselective catalytic Mukaiyama oxocarbenium aldol. *5 mol % 7 was used. †Reaction was performed with the trimethylsilyl ketene acetal
derived from methyl cyclohexanecarboxylate. ‡Reaction was performed in ethyl acetate at room temperature for 72 hours. §Reaction was performed with
1-o-tolyl-1-trimethylsiloxy ethylene as the nucleophile. R, R1, and R2, generic substituent; TMS, trimethylsilyl; EtOAc, ethyl acetate.
SCIENCE sciencemag.org
26 FEBRUARY 2016 • VOL 351 ISSUE 6276 963

RESEARCH
| REPORTS
Fig. 4. Mechanistic
considerations.
(A) Mechanistic ratio-
nale for Mukaiyama
oxocarbenium addition
reaction. (B) Demon-
stration of an
alternative chiral anion
catalysis pathway.
(C and D) Stereo-
chemical rationale for
(C) the Mannich reac-
tion catalyzed by acid
7 and (D) the oxocar-
benium aldol cata-
lyzed by acid 7. *Three
of the carboxylates
(at the end of gray
bonds) are not shown
for clarity. (E) Molecu-
lar structure of
7•NMe4. In the face
view, the cation is not
shown. In the edge
view, two adjacent
cations are shown.
platform may have utility in this burgeoning area
as well.
or the cyclopentadienyl ring. In this case, p-facial
blocking as described above would lead to reac-
tion via the si prochiral face, which corresponds to
the observed stereoselectivity for the R-enantiomer.
We have obtained a single-crystal x-ray struc-
ture of catalyst 7 as its tetramethylammonium
salt (Fig. 4E). The face view (cation removed)
shows a gearing of the carboxyl substituents all
in the same direction, presumably to minimize
steric repulsion of the menthyl substituents. The
alternative edge view shows the substantial di-
hedral angle between each carbonyl and the cyclo-
pentadienyl ring ranging from 30.6° to 52.1°,
with an average of 43.8°, and the orientation of
each of the carbonyl oxygens to the same side of
the ring. In this view, two adjacent (and equiv-
alent) cations are included, which shows that one
half of each ammonium is embedded in a hy-
drophobic pocket formed by the menthyl groups,
whereas the other half is associated with the op-
posite ring face toward which the carbonyls are
oriented. The extent to which this structure is
related to the catalytic operation of 7 is not cur-
rently known and will be the subject of further
study.
A stereochemical rationale for this chemistry
is shown in Fig. 4C. The presence of multiple
rotatable bonds in catalyst 7 renders a number
of transition-state organizations plausible. For the
Mannich reaction, one possible scenario involves
association of the protonated imine and one of
the carbonyl oxygens of the catalyst via two hy-
drogen bonds (Fig. 4C). To accommodate this
interaction would require substantial torquing
of the carboxyl substituent, thereby placing
the iminium in close proximity to an adjacent
carboxymenthyl substituent. Blocking of the re
prochiral face of the iminium would then ration-
alize the observed stereochemistry.
A similar model can be invoked for the
oxocarbenium addition; however, in this case
only a single hydrogen bonding substituent is
present (Fig. 4D). We speculate that there may
be additional organizing interactions between
the C–H bonds adjacent to the oxygen of the
oxocarbenium ion and the catalyst carbonyl and/
The ability to synthesize asymmetric catalysts
rapidly, inexpensively, and on scale is an impor-
tant goal for the field of organic chemistry. The
current platform offers a useful advance in this
regard by allowing preparation of an effective
chiral Brønsted acid catalyst in a single step from
a readily available acid core and simple chiral
alcohols or amines. The central feature of this
platform, the cyclopentadienyl anion, enables the
cooperation of multiple acidifying elements to
produce powerful acid catalysts.
REFERENCES AND NOTES
1. J. N. Brønsted, Rec. Trav. Chim. 42, 718–728 (1923).
2. T. M. Lowry, Chem. Ind. 42, 43–47 (1923).
3. D. Parmar, E. Sugiono, S. Raja, M. Rueping, Chem. Rev. 114,
9047–9153 (2014).
4. T. Akiyama, J. Itoh, K. Fuchibe, Adv. Synth. Catal. 348,
999–1010 (2006).
5. T. Akiyama, in Acid Catalysis in Modern Organic Synthesis,
H. Yamamoto, K. Ishihara, Eds. (Wiley-VCH, Weinheim, 2008),
pp. 62–107.
964 26 FEBRUARY 2016 • VOL 351 ISSUE 6276
sciencemag.org SCIENCE

RESEARCH
| REPORTS
6. T. Hashimoto, in Asymmetric Organocatalysis 2: Brønsted
Base and Acid Catalysts, and Additional Topics, K. Maruoka, Ed.
(Georg Thieme Verlag, Stuttgart, New York, 2012),
pp. 279–296.
CATALYSIS
7. T. Akiyama, J. Itoh, K. Yokota, K. Fuchibe, Angew. Chem. Int. Ed.
43, 1566–1568 (2004).
8. D. Uraguchi, M. Terada, J. Am. Chem. Soc. 126, 5356–5357
(2004).
9. I. Ahmed, D. A. Clark, Org. Lett. 16, 4332–4335 (2014).
10. D. Kampen, C. M. Reisinger, B. List, Top. Curr. Chem. 291,
395–456 (2010).
11. T. James, M. van Gemmeren, B. List, Chem. Rev. 115,
9388–9409 (2015).
12. T. Akiyama, Chem. Rev. 107, 5744–5758 (2007).
13. T. Akiyama, K. Mori, Chem. Rev. 115, 9277–9306 (2015).
14. A. Berkessel, P. Christ, N. Leconte, J.-M. Neudörfl, M. Schäfer,
Eur. J. Org. Chem. 2010, 5165–5170 (2010).
15. D. Nakashima, H. Yamamoto, J. Am. Chem. Soc. 128,
9626–9627 (2006).
16. M. Treskow, J. Neudörfl, R. Giemoth, Eur. J. Org. Chem. 2009,
3693–3697 (2009).
17. G. B. Rowland et al., J. Am. Chem. Soc. 127, 15696–15697
(2005).
Palladium-tin catalysts for the direct
synthesis of H2O2 with high selectivity
Simon J. Freakley,1*† Qian He,2,3† Jonathan H. Harrhy,1 Li Lu,2 David A. Crole,1
David J. Morgan,1 Edwin N. Ntainjua,1 Jennifer K. Edwards,1 Albert F. Carley,1
Albina Y. Borisevich,3,4 Christopher J. Kiely,2 Graham J. Hutchings1*
The direct synthesis of hydrogen peroxide (H2O2) from H2 and O2 represents a potentially
atom-efficient alternative to the current industrial indirect process. We show that the
addition of tin to palladium catalysts coupled with an appropriate heat treatment cycle
switches off the sequential hydrogenation and decomposition reactions, enabling
selectivities of >95% toward H2O2. This effect arises from a tin oxide surface layer that
encapsulates small Pd-rich particles while leaving larger Pd-Sn alloy particles exposed.
We show that this effect is a general feature for oxide-supported Pd catalysts containing an
appropriate second metal oxide component, and we set out the design principles for producing
high-selectivity Pd-based catalysts for direct H2O2 production that do not contain gold.
18. F. Xu et al., J. Org. Chem. 75, 8677–8680 (2010).
19. H. Yanai, T. Taguchi, J. Fluor. Chem. 174, 108–119 (2015).
20. H. Yamamoto, D. Nakashima, in Acid Catalysis in Modern
Organic Synthesis, vol. 1, H. Yamamoto, K. Ishihara, Eds.
(Wiley-VCH, Weinheim, 2008), pp. 35–62.
21. F. G. Bordwell, G. E. Drucker, H. E. Fried, J. Org. Chem. 46,
632–635 (1981).
22. M. I. Bruce, A. H. White, Aust. J. Chem. 43, 949–995
(1990).
23. O. Diels, Ber. Dtsch. Chem. Ges. 75, 1452–1467 (1942).
24. O. Diels, U. Kock, Liebigs Ann. Chem. 556, 38–50 (1944).
25. I. E. Mikhailov, G. A. Dushenko, V. I. Minkin, L. P. Olekhnovich,
J. Org. Chem. USSR 20, 1509–1514 (1984).
26. E. Le Goff, R. LaCount, J. Org. Chem. 29, 423–427 (1964).
27. K. Kaupmees, N. Tolstoluzhsky, S. Raja, M. Rueping, I. Leito,
Angew. Chem. Int. Ed. 52, 11569–11572 (2013).
28. S. E. Reisman, A. G. Doyle, E. N. Jacobsen, J. Am. Chem. Soc.
130, 7198–7199 (2008).
29. N. Z. Burns, M. R. Witten, E. N. Jacobsen, J. Am. Chem. Soc.
133, 14578–14581 (2011).
30. M. R. Witten, E. N. Jacobsen, Angew. Chem. Int. Ed. 53,
5912–5916 (2014).
31. G. C. Tsui, L. Liu, B. List, Angew. Chem. Int. Ed. 54, 7703–7706
(2015).
32. J. H. Kim, I. Čorić, S. Vellalath, B. List, Angew. Chem. Int. Ed.
52, 4474–4477 (2013).
33. I. Čorić, B. List, Nature 483, 315–319 (2012).
34. A. Borokiva, P.-I. Tang, S. Klapman, P. Nagorny, Angew. Chem.
Int. Ed. 52, 13424–13428 (2013).
35. L. Ratjen, P. García-García, F. Lay, M. E. Beck, B. List,
Angew. Chem. Int. Ed. 50, 754–758 (2011).
36. G. L. Hamilton, E. J. Kang, M. Mba, F. D. Toste, Science 317,
496–499 (2007).
37. K. Brak, E. N. Jacobsen, Angew. Chem. Int. Ed. 52, 534–561
(2013).
38. M. Mahlau, B. List, Angew. Chem. Int. Ed. 52, 518–533 (2013).
urrently, the demand for H2O2 is met by
an indirect process, which produces H2O2
through the sequential hydrogenation and
oxidation of a substituted anthraquinone
(1). For economic reasons, the process is
Although this approach was very successful on an
activated carbon support material, the same block-
ing effect could not be fully achieved on other com-
mercial support materials such as SiO2 and TiO2.
Because O2 dissociation is undesirable in the
direct synthesis of H2O2, the reaction can be
treated as a selective hydrogenation of O2. We
explored other Pd-metal combinations that are
used for selective hydrogenation reactions as po-
tential catalysts for H2O2 synthesis, focusing on
nonprecious metals to lower costs. Tin (Sn) has
been used to modify hydrogenation catalysts in
reactions such as the selective hydrogenation of
1,3-butadiene (11). Further examples have been
reported for the liquid-phase hydrogenation of
hexa-1,3-diene and hexa-1,5-diene (12) as well as
the hydrogenation of unsaturated alcohols (13).
The addition of Sn to Pd or Pt can alter the behav-
ior of the catalyst during hydrogenation reactions
and, in particular, may have an effect on subse-
quent reactions of the products with the catalyst.
We report the development of Sn-containing
Pd catalysts on commercially available TiO2 and
SiO2 supports that can achieve >95% selectivity
toward direct H2O2 synthesis. These catalysts,
after being subjected to an appropriate heat
treatment regimen, obviate the need for pre-
treating the support with acids and contain far
less precious metal than Au-Pd catalysts. We
also present the general principles whereby
high-selectivity catalysts can be obtained with
other Pd-metal combinations.
C
operated at large scale and produces concen-
trated H2O2. In reality, many applications, such
as disinfection and water purification, require
only dilute H2O2, which means that concen-
trated H2O2 must be diluted at the point of use.
Research into the direct synthesis of H2O2 from
H2 and O2 as a more suitable solution to small-
scale, on-site H2O2 production has focused on
palladium (Pd)–based catalysts (2–4). However,
H2O2 is itself highly reactive, and the presence
of H2 favors hydrogenation and decomposition
reactions that form water. The addition of strong
acids and halides to the reaction medium can
suppress the sequential hydrogenation and deg-
radation in supported Pd catalysts (5) but can
also promote metal leaching and requires further
purification of the H2O2 before use.
Bimetallic Au-Pd alloy catalysts have been ex-
tensively studied as catalysts for the direct H2O2
synthesis reaction on a number of support mate-
rials, including TiO2, SiO2, and activated carbon
(6–9). Yields comparable to monometallic Pd cat-
alysts can be achieved without the need for acid
and halide additives in the reaction mixture, and
95% selectivity to H2O2 can be achieved with Au-
Pd alloy nanoparticles (NPs) dispersed on an acid-
pretreated activated carbon support material (10).
Hydrogen peroxide hydrogenation could be de-
coupled from H2O2 synthesis with an acid pre-
treatment that blocked sites on the carbon support
material responsible for H2O2 degradation (10).
ACKNOWLEDGMENTS
Funding for this work was provided by the National Science
Foundation under CHE-0953259. C.D.G. is grateful for a
National Science Foundation Graduate Fellowship. We thank
P. Quinlivan and the Parkin group, as well as D. Paley, for x-ray
structure determination and the National Science Foundation
(CHE-0619638) for acquisition of an x-ray diffractometer.
We are grateful to M. Vetticatt (State University of New York,
Binghamton) and V. Roytman for assistance with the
stereochemical models. Metrical parameters for the structures
of 7•NMe4 and the oxocarbenium Aldol product are available
free of charge from the Cambridge Crystallographic Data Centre
under CCDC 1450055 and 1450056, respectively.
Simple impregnation of Au and Pd metal salts
onto many catalyst supports has been shown to
generate highly active catalysts for direct H2O2
synthesis. In addition, high-temperature calcina-
tion or reduction treatments are known to be
crucial to improve the stability of the catalyst. As
a starting point, we used this simple catalyst
preparation methodology to prepare a 2.5 weight
percent (wt%) Pd–2.5 wt % Sn/TiO2 catalyst as
well as its monometallic analogs (8, 10). A syn-
ergistic effect toward higher H2O2 productivity
was observed when both metals were present
1Cardiff Catalysis Institute and School of Chemistry, Cardiff
University, Cardiff CF10 3AT, UK. 2Department of Materials
Science and Engineering, Lehigh University, Bethlehem, PA
18015, USA. 3Materials Science and Technology Division, Oak
Ridge National Laboratory, Oak Ridge, TN 37831, USA. 4Center
for Nanophase Materials Sciences, Oak Ridge National
Laboratory, Oak Ridge, TN 37831, USA.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/351/6276/961/suppl/DC1
Materials and Methods
Figs. S1 to S4
Tables S1 to S4
References (39–57)
20 July 2015; accepted 27 January 2016
10.1126/science.aad0591
*Corresponding author. E-mail: hutch@cf.ac.uk (G.J.H.); freakleys@
cf.ac.uk (S.J.F.) †These authors contributed equally to this work.
SCIENCE sciencemag.org
26 FEBRUARY 2016 • VOL 351 ISSUE 6276 965

An aromatic ion platform for enantioselective Brønsted acid catalysis
Chirag D. Gheewala et al.
Science 351, 961 (2016);
DOI: 10.1126/science.aad0591
This copy is for your personal, non-commercial use only.
If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of February 25, 2016 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
/content/351/6276/961.full.html
Supporting Online Material can be found at:
/content/suppl/2016/02/24/351.6276.961.DC1.html
This article cites 54 articles, 1 of which can be accessed free:
/content/351/6276/961.full.html#ref-list-1
This article appears in the following subject collections:
Chemistry
/cgi/collection/chemistry
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2016 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.