Spatially orthogonal chemical functionalization of a hierarchical pore network for catalytic cascade reactions

Article abstract of DOI:10.1038/nmat4478

The chemical functionality within porous architectures dictates their performance as heterogeneous catalysts; however, synthetic routes to control the spatial distribution of individual functions within porous solids are limited. Here we report the fabrication of spatially orthogonal bifunctional porous catalysts, through the stepwise template removal and chemical functionalization of an interconnected silica framework. Selective removal of polystyrene nanosphere templates from a lyotropic liquid crystal-templated silica sol-gel matrix, followed by extraction of the liquid crystal template, affords a hierarchical macroporous-mesoporous architecture. Decoupling of the individual template extractions allows independent functionalization of macropore and mesopore networks on the basis of chemical and/or size specificity. Spatial compartmentalization of, and directed molecular transport between, chemical functionalities affords control over the reaction sequence in catalytic cascades; herein illustrated by the Pd/Pt-catalysed oxidation of cinnamyl alcohol to cinnamic acid. We anticipate that our methodology will prompt further design of multifunctional materials comprising spatially compartmentalized functions.

Full text of DOI:10.1038/nmat4478

LETTERS
PUBLISHED ONLINE: 16 NOVEMBER 2015 | DOI: 10.1038/NMAT4478
Spatially orthogonal chemical functionalization
of a hierarchical pore network for catalytic
cascade reactions
Christopher M. A. Parlett1, Mark A. Isaacs1, Simon K. Beaumont2, Laura M. Bingham2,
Nicole S. Hondow3, Karen Wilson1 and Adam F. Lee1
*
Thechemicalfunctionalitywithinporousarchitecturesdictates true liquid crystal templating route (see Methods), adapting
their performance as heterogeneous catalysts1; however, literature methodology15–17, in which a Pluronic P123 block
synthetic routes to control the spatial distribution of indi- co-polymer surfactant-templated mesoporous silica network was
vidual functions within porous solids are limited. Here we formed through the acid hydrolysis of tetraethoxyorthosilane
report the fabrication of spatially orthogonal bifunctional around an ordered array of unfunctionalized polystyrene
porous catalysts, through the stepwise template removal colloidal nanospheres (scanning electron microscopy (SEM)
and chemical functionalization of an interconnected silica and polydispersity index <0.1 by dynamic light scattering,
framework. Selective removal of polystyrene nanosphere Supplementary Fig. 1), as illustrated in Fig. 1a. The judicious
templates from a lyotropic liquid crystal-templated silica combination of polar P123 mesopore and nonpolar polystyrene
sol–gel matrix, followed by extraction of the liquid crystal macropore templates facilitates their subsequent independent
template, aꢀords a hierarchical macroporous–mesoporous extraction by means of different polarity solvents. Exclusive
architecture. Decoupling of the individual template extrac- extraction of the polystyrene macropore template was unsuccessful
tions allows independent functionalization of macropore and through conventional calcination (or toluene reflux) protocols,
mesopore networks on the basis of chemical and/or size owing to simultaneous combustion (or dissolution) of the
specificity. Spatial compartmentalization of, and directed P123 mesopore template. However, a new sub-ambient toluene
molecular transport between, chemical functionalities aꢀords extraction protocol achieved >95% selective polystyrene removal
control over the reaction sequence in catalytic cascades2,3
;
as evidenced by TEM and porosimetry (Fig. 1b and Supplementary
herein illustrated by the Pd/Pt-catalysed oxidation of cinnamyl Fig. 2 respectively). The resulting 20 m2 g−1 material comprised
alcohol to cinnamic acid. We anticipate that our methodology empty macropores formed/abutted by surfactant template-filled
will prompt further design of multifunctional materials4–6 mesoporous silica (Fig. 1b), permitting selective hydrophobization
comprising spatially compartmentalized functions.
of the macropore network with octyl groups through reaction
Multifunctional nanomaterials are ubiquitous in diverse of triethoxyoctylsilane with exposed surface silanols (Fig. 1c),
technological applications spanning energy7, the environment confirmed by porosimetry and contact angle measurements
and health8, to information storage and communication9. (Supplementary Fig. 3). Removal of the P123 surfactant template
Synergy between chemically distinct functionalities within porous from within the mesopores was facile under methanol reflux
architectures in particular underpins new magnetic5 or optical (Supplementary Fig. 4) resulting in a fully detemplated hierarchical
devices, and heterogeneous catalysts2. Limited spatial patterning bimodal architecture (Fig. 1d) containing (interconnected)
of select materials has been achieved in three dimensions at the hydrophobic 350 nm macropores and hydrophilic 3.5 nm
submicrometre scale, for example through vapour deposition of mesopores. The total surface area of 300 m2 g−1 was consistent with
aligned carbon nanotubes10 or two-photon excitation of hydrogels reports for macroporous SBA-15 prepared using high-temperature
over planar substrates11, or through biogenic routes12. Such template decomposition, and indeed low-angle X-ray diffraction
approaches respectively incorporate only a single chemically confirmed a p6mm hexagonal arrangement of ordered mesoporous
distinct function, require optically transparent and photoresponsive channels. Conventional wet impregnation of this high-area porous
materials, or impart a poor degree of ordering and limited material with an aqueous metal salt (H2PtCl6), followed by a mild
thermochemical stability. Ordered two-dimensional arrays of (100 ◦C) reduction under molecular H2, yielded chloride-free,
monometallic or metal/oxide bifunctional nanocrystals have face-centred cubic metal Pt nanoparticles (NPs) of 2.2 nm mean
been prepared as model catalysts using electron lithography13 diameter confined within the mesopores (Fig. 1e). This spatial
and drop-casting14 respectively, but afford extremely low surface localization of nanoparticulate Pt was possible only because of
areas with co-located active sites. Hence, there are no synthetic the preceding macropore hydrophobization, which directed the
routes able to control either the nanoscale spatial distribution, water-dispersed platinum precursor away from the octyl-grafted
or the communication between, individual functions within macropores (Supplementary Fig. 5). Finally, oleylamine-capped,
three-dimensional porous solids.
5.6 ± 0.8 nm colloidal Pd NPs (ref. 18; Supplementary Fig. 6)
A hierarchically ordered macroporous–mesoporous SBA-15 were introduced exclusively into the macropores; their diameter
silica framework was synthesized through a new lyotropic preventing access to the smaller mesopores (Fig. 1f).
1European Bioenergy Research Institute, Aston University, Birmingham B4 7ET, UK. 2Department of Chemistry, University of Durham, Durham DH1 3LE,
UK. 3Institute for Materials Research, University of Leeds, Leeds LS2 9JT, UK. *e-mail: A.F.Lee@aston.ac.uk
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
1

NATURE MATERIALS DOI: 10.1038/NMAT4478
LETTERS
a
b
c
Surfactant micelle
Hydrophobic
macropores
Nonpolar
polymer
nanosphere
250 nm
250 nm
f
e
d
Hydrophobic
macropores
100 nm
30 nm
20 nm
200 nm
Figure 1 | Synthetic strategy for spatially orthogonal functionalization of hierarchical architectures. a,b, Illustrations and bright-field TEM images of
parent polystyrene colloidal nanospheres encapsulated within a P123-templated SBA-15 silica network (a), and the ordered macroporous framework after
selective removal of the polystyrene macropore template by means of toluene reflux (b). c, Illustration of selective hydrophobization of the macropore
framework by triethoxyoctylsilane. d, Illustration and bright-field TEM image of ordered hydrophobized macropores and (inset) ordered mesopores
following P123 extraction under methanol reflux. e, Illustration and HAADF-STEM image of mesopore channels selectively functionalized with Pt NPs
through aqueous impregnation and reduction. f, Illustration and bright-field TEM image of hydrophobic macropores selectively functionalized with
monodispersed 5.6±0.8 nm colloidal oleylamine-stabilized Pd NPs (highlighted in yellow).
Spatial compartmentalization of small Pt NPs within the and 5.5 nm, superimposable with those observed for monometallic
hydrophilic mesopores, and larger Pd NPs within the hydrophobic Pt or Pd analogues prepared according to the synthetic route in
macropores, was verified by high-angle annular dark-field scanning Fig. 1, and consistent with Pd particle sizes from X-ray diffraction
transmission electron microscopy (HAADF-STEM) imaging (Supplementary Fig. 9). Area-averaged EDX compositions of
and energy-dispersive spectroscopy (EDX) analysis (Fig. 2) and mesopores and macropores in Fig. 2e affirmed the presence of Pt
additional SEM/DF-STEM imaging (Supplementary Figs 7 and 8). solely within the former and Pd within the latter. Detailed textural
Z-contrast imaging and line profile analysis through cross- analysis (Supplementary Table 1) confirmed that colloidal Pd
sections of the ordered mesoporous silica framework spanning incorporation had minimal impact on BET (Brunauer–Emmett–
two macropores (Fig. 2a,b) provide compelling evidence for Teller)/mesopore surface areas or mesopore volume, congruent
partitioning of the two metals, with Pt observed only within the with selective functionalization of macropores; in contrast, Pt
mesopores and Pd at the macropore perimeter, and NPs localized wet impregnation significantly lowered these same properties
within the mesopores affording stronger contrast as anticipated congruent with selective functionalization of mesopores.
for heavier Pt scatterers (Fig. 2c). Particle size distributions for
Cascade reactions are sequential chemical transformations
the spatially orthogonal bimetallic material in Fig. 2d also reveal in which the starting substrate undergoes a reaction whose
a bimodal distribution of NPs, with well-defined maxima at 2.4 product becomes the substrate for the next step, and so on, until
2
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT4478
LETTERS
a
b
c
Line profile 1
Mesopore
Line profile 2
Mesopore
30
25
20
15
10
5
1,200
1,000
800
600
400
200
0
Si
1
O
Macropore
2
Pd
Pt
Pd Z = 46
Macropore
Pt Z = 78
0
50 nm
0
25 50 75 10 50 90 130 170
Distance (nm) Distance (nm)
10 nm
d
e
f
60
50
40
30
20
10
0
5.7 wt% Pt
0.0 wt% Pd
0.0 wt% Pt
2.8 wt% Pd
Mesoporous NPs
Monometallic Pt NPs
Macroporous NPs
Monometallic Pd NPs
Macropore
Macropore
Macropore
Mesopores
Macropore
25 nm
30 nm
Macropore
Nanoparticle diameter (nm)
Figure 2 | Visualization of spatially orthogonal Pd and Pt NPs. a, HAADF-STEM of NPs distributed across a section of the mesoporous SBA-15 framework
surrounded by macropores. b, Cross-sectional EDX compositions of the mesoporous SBA-15 framework intersecting either one (1) or two (2) macropores
depicted in a. c, High-resolution HAADF-STEM image highlighting stronger contrast of smaller, mesopore-confined Pt NPs. d, Particle size distributions of
monometallic and bimetallic hierarchically ordered macroporous–mesoporous SBA-15 demonstrating selective functionalization of macropores with large
Pd NPs and mesopores with small Pt NPs. e,f, Elemental mapping of mesopores (macropores) showing exclusive Pt (Pd) functionalization.
a stable product is reached19. Cascades thus offer great advantages cinnamaldehyde and cinnamic acid productivity for our spatially
in respect of atom economy, and economies of time, labour, orthogonal PdPt bimetallic catalyst with a variety of monometallic
resource management and waste generation, and permit the use of and bimetallic analogues. Monometallic catalysts, their combination
synthetically enabling intermediates that may not be practical to as a physical mixture, and conventionally synthesized bimetallic
isolate. Catalytic cascades, in which the product of a reaction catal- catalysts in which Pd and Pt were co-localized within the
ysed by species A undergoes a subsequent distinct transformation hierarchical pore network, proved ineffective, with either low
catalysed by a second species B, are hindered by the possibility of rates of alcohol oxidation, poor selectivity to the cinnamaldehyde
undesired interactions between the initial substrate and the second intermediate and/or very poor acid production, demonstrating
active site, or indeed between the two catalytic species20. Such highly the inability of Pd or Pt to either individually catalyse the
desirable ‘one-pot’ catalytic cascades therefore necessitate the spatial cascade reaction, or to communicate effectively when isolated
separation of each catalytic step. The catalytic advantage of spatially in discrete catalyst support particles. In contrast, their spatial
segregating Pt NPs within mesopores, accessible overwhelmingly compartmentalization within separate but interconnected pore
only through interconnected macropores containing Pd NPs, networks, mere nanometres apart, permits control over the reaction
was explored for the cascade oxidative dehydrogenation of sequence enabling oxidation of cinnamyl alcohol entering the
cinnamyl alcohol → cinnamaldehyde → cinnamic acid, the macropores to cinnamaldehyde over Pd, and subsequent aldehyde
last of these an important flavouring and essential oil21,22. Pd diffusion into the mesopores and oxidation to cinnamic acid
is highly selective for catalysing cinnamyl alcohol oxidation to over Pt, conferring an order of magnitude enhancement in
cinnamaldehyde23,24, but promotes decarbonylation of the resultant cinnamic acid yield. Only selectively functionalized materials in
aldehyde product; in contrast, Pt favours undesired hydrogenation which cinnamyl alcohol was able to react over Pd NPs within
of cinnamyl alcohol (driven by reactively formed surface hydrogen) macropores, before encountering Pt NPs within mesopores, permit
to 3-phenylpropionaldehyde25, but is highly selective towards the cascade oxidation.
cinnamaldehyde oxidation to the desirable cinnamic acid product26.
Localization of different functions within multimodal porous
An optimal catalyst design would therefore ensure that cinnamyl architectures can enhance selectivity in a cascade reaction only if
alcohol was oxidized over Pd before encountering Pt sites, while the rate of reactant diffusion is slow relative to the first step in
permitting the reactively formed cinnamaldehyde to subsequently the reaction sequence. Quantitative comparison of our spatially
access Pt sites for the selective production of cinnamic acid in orthogonal bimetallic catalyst with relevant controls evidences that
the second oxidation step. Such a goal is achievable only through the rate of cinnamyl alcohol selox over Pd indeed exceeds the
spatial control over the location of Pd and Pt within a hierarchical rate of cinnamyl alcohol diffusion from macropores → mesopores
catalyst, illustrated schematically in Supplementary Fig. 10, and (Supplementary Equation (1)). The benefits of spatially separating
demonstrated through our catalytic studies below.
the individual oxidation steps was further highlighted through a
Figure 3 summarizes the reaction kinetics (Supplementary spiking experiment in which cinnamic acid was directly introduced
Figs 11–14), comparing cinnamyl alcohol conversion and desired to the solution mixture at the start of the reaction, enabling
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved
3

NATURE MATERIALS DOI: 10.1038/NMAT4478
Received 4 August 2015; accepted 1 October 2015;
LETTERS
9,000
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
350
300
250
200
150
100
50
published online 16 November 2015
References
1. Parlett, C. M. A., Wilson, K. & Lee, A. F. Hierarchical porous materials:
Catalytic applications. Chem. Soc. Rev. 42, 3876–3893 (2013).
2. Climent, M. J., Corma, A., Iborra, S. & Sabater, M. J. Heterogeneous catalysis
for tandem reactions. ACS Catal. 4, 870–891 (2014).
3. Merino, E. et al. Synthesis of structured porous polymers with acid and basic
sites and their catalytic application in cascade-type reactions. Chem. Mater. 25,
981–988 (2013).
4. Balazs, A. C., Emrick, T. & Russell, T. P. Nanoparticle polymer composites:
Where two small worlds meet. Science 314, 1107–1110 (2006).
5. Cheong, S. W. & Mostovoy, M. Multiferroics: A magnetic twist for
ferroelectricity. Nature Mater. 6, 13–20 (2007).
0
6. Kim, J., Piao, Y. & Hyeon, T. Multifunctional nanostructured materials for
multimodal imaging, and simultaneous imaging and therapy. Chem. Soc. Rev.
38, 372–390 (2009).
7. Maeda, K. et al. Photocatalyst releasing hydrogen from water. Nature 440,
295 (2006).
8. Ragesh, P., Ganesh, V. A., Nair, S. V. & Nair, A. S. A review on ‘self-cleaning and
multifunctional materials’. J. Mater. Chem. A 2, 14773–14797 (2014).
9. Lu, W. & Lieber, C. M. Nanoelectronics from the bottom up. Nature Mater. 6,
841–850 (2007).
10. Zhou, W., Ding, L., Yang, S. & Liu, J. Orthogonal orientation control of carbon
nanotube growth. J. Am. Chem. Soc. 132, 336–341 (2010).
11. Paciello, A. & Santonicola, M. G. A supramolecular two-photon-active
hydrogel platform for direct bioconjugation under near-infrared radiation.
J. Mater. Chem. B 3, 1313–1320 (2015).
12. Chen, A. Y. et al. Synthesis and patterning of tunable multiscale materials with
engineered cells. Nature Mater. 13, 515–523 (2014).
13. Grunes, J., Zhu, J., Anderson, E. A. & Somorjai, G. A. Ethylene hydrogenation
over platinum nanoparticle array model catalysts fabricated by electron beam
lithography: Determination of active metal surface area. J. Phys. Chem. B 106,
11463–11468 (2002).
14. Yamada, Y. et al. Nanocrystal bilayer for tandem catalysis. Nature Chem. 3,
372–376 (2011).
Figure 3 | Reaction dynamics of cinnamyl alcohol oxidation.
Active site-normalized rates of cinnamyl alcohol conversion (blue), and
cinnamaldehyde (yellow) and cinnamic acid (green) production for the
one-pot cascade oxidation of cinnamyl alcohol → cinnamaldehyde →
cinnamic acid over a spatially orthogonal, hierarchical Pd macroporous–Pt
mesoporous SBA-15 catalyst (1 wt% in each metal), and a range of
bimetallic and monometallic SBA-15 analogues including a physical mixture
of 1 wt% Pd macroporous and 1 wt% Pd mesoporous SBA-15. Spatially
separating the two active sites, and first directing cinnamyl alcohol over Pd,
significantly enhances activity and selectivity towards desired selox
products. All reactions were performed at 150 ◦C and 5 bar O2.
its competitive adsorption with cinnamyl alcohol over Pd within 15. Dhainaut, J., Dacquin, J.-P., Lee, A. F. & Wilson, K. Hierarchical
macroporous-mesoporous SBA-15 sulfonic acid catalysts for biodiesel
the macropores (Supplementary Fig. 15). Cinnamic acid addition
synthesis. Green Chem. 12, 296–303 (2010).
promoted undesired cinnamyl alcohol decarbonylation over Pd at
16. Wainwright, S. G. et al. True liquid crystal templating of SBA-15 with reduced
the expense of its selective oxidation to cinnamaldehyde. Spatial
microporosity. Micropor. Mesopor. Mater. 172, 112–117 (2013).
separation of the selective oxidation step over Pd that produces
cinnamaldehyde, from that over Pt that produces cinnamic acid, and
17. Zhao, D. Y. et al. Triblock copolymer syntheses of mesoporous silica with
periodic 50 to 300 angstrom pores. Science 279, 548–552 (1998).
controlling the sequence of these two reactions such that cinnamic 18. Mazumder, V. & Sun, S. Oleylamine-mediated synthesis of Pd
nanoparticles for catalytic formic acid oxidation. J. Am. Chem. Soc. 131,
acid is formed only after the alcohol has already undergone
4588–4589 (2009).
oxidation to the aldehyde is thus critical to achieving good selox
19. Nicolaou, K. C., Edmonds, D. J. & Bulger, P. G. Cascade reactions in total
performance. Indeed, our spatially orthogonal bimetallic material
synthesis. Angew. Chem. Int. Ed. 45, 7134–7186 (2006).
outperforms all literature catalysts (Supplementary Table 2) in terms
of both turnover frequencies (TOFs) for cinnamyl alcohol oxidation,
20. Motokura, K., Tada, M. & Iwasawa, Y. Heterogeneous organic base-catalyzed
reactions enhanced by acid supports. J. Am. Chem. Soc. 129, 9540–9541 (2007).
and selectivity to cinnamaldehyde in high-activity systems, and 21. Burt, S. Essential oils: Their antibacterial properties and potential applications
in foods—A review. Int. J. Food Microbiol. 94, 223–253 (2004).
crucially, is the only heterogeneous catalyst to our knowledge able to
22. Soto-Vaca, A., Gutierrez, A., Losso, J. N., Xu, Z. & Finley, J. W. Evolution of
produce quantifiable yields of cinnamic acid in a one-pot oxidation
phenolic compounds from color and flavor problems to health benefits.
of cinnamyl alcohol.
J. Agric. Food Chem. 60, 6658–6677 (2012).
The ability to control the spatial patterning of functionalities
within the pore networks of multimodal architectures confers
unique and flexible properties on the resulting materials, which
are critical to the development of, for example, cascade reactions
in which the reaction sequence is controlled by the location of
different catalytically active sites. Our methodology is extendable
to diverse catalytic cascades featuring chemically incompatible
(for example, acid–base3,27,28) functions, and, for example, selective
functionalization of mesopores and macropores with different
fluorescent markers to afford new sensing devices able to
discriminate between analytes of different molecular dimensions;
macropore and mesopore dimensions are independently tunable,
and may be incorporated into different host matrices29.
23. Hackett, S. F. J. et al. High-activity, single-site mesoporous Pd/Al2O3 catalysts
for selective aerobic oxidation of allylic alcohols. Angew. Chem. Int. Ed. 119,
8747–8750 (2007).
24. Lee, A. F. et al. Reaction-driven surface restructuring and selectivity control in
allylic alcohol catalytic aerobic oxidation over Pd. J. Am. Chem. Soc. 133,
5724–5727 (2011).
25. Durndell, L. J. et al. Selectivity control in Pt-catalyzed cinnamaldehyde
hydrogenation. Sci. Rep. 5, 9425 (2015).
26. Durndell, L. J., Parlett, C. M. A., Hondow, N. S., Wilson, K. & Lee, A. F. Tunable
Pt nanocatalysts for the aerobic selox of cinnamyl alcohol. Nanoscale 5,
5412–5419 (2013).
27. Zhang, F., Jiang, H., Li, X., Wu, X. & Li, H. Amine-functionalized GO as an
active and reusable acid–base bifunctional catalyst for one-pot cascade
reactions. ACS Catal. 4, 394–401 (2014).
28. Zeidan, R. K., Hwang, S.-J. & Davis, M. E. Multifunctional heterogeneous
catalysts: SBA-15-containing primary amines and sulfonic acids. Angew. Chem.
Int. Ed. 45, 6332–6335 (2006).
29. Dacquin, J.-P. et al. An efficient route to highly organized, tunable
macroporous-mesoporous alumina. J. Am. Chem. Soc. 131,
12896–12897 (2009).
Methods
Methods and any associated references are available in the online
version of the paper.
4
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT4478
LETTERS
Acknowledgements
characterized the Pd colloids. C.M.A.P., M.A.I. and N.S.H. undertook materials
This work was supported by the EPSRC (EP/G007594/4). A.F.L. was supported by an
EPSRC Leadership Fellowship, K.W. by a Royal Society Industry Fellowship, and S.K.B.
by a Durham University Addison Wheeler Fellowship and The Leverhulme Trust ECF
schemes. L.M.B. acknowledges the EPSRC for a studentship. Electron microscopy access
was provided through the Leeds EPSRC Nanoscience and Nanotechnology Research
Equipment Facility (LENNF) (EP/K023853/1), the University of Birmingham Nanoscale
Physics Laboratory, and DU GJ Russell Microscopy Facility.
characterization. A.F.L. wrote the manuscript.
Additional information
Supplementary information is available in the online version of the paper. Reprints and
permissions information is available online at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to A.F.L.
Author contributions
Competing financial interests
A.F.L., C.M.A.P. and K.W. planned the experiments. C.M.A.P. synthesized all porous
materials and performed catalytic testing. L.M.B. and S.K.B. synthesized and
The authors declare no competing financial interests.
NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials
5
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT4478
LETTERS
Methods
hierarchically ordered macroporous–mesoporous SBA-15 with an aqueous solution
of dihydrogen hexachloroplatinate. The parent support (0.6 g) was stirred in the
aqueous salt solution (3 cm3, 0.01575 g Pt salt, nominal 1 wt% loading) for 18 h in
the dark. A dry powder was obtained by gentle heating of the slurry at 50 ◦C for
10 h, followed by 100 ◦C reduction under H2 (10 cm3 min−1) for 1 h, yielding
monometallic Pt NPs. The bimetallic Pd macropore/Pt mesopore material was
produced from the preceding Pt functionalized material (0.3 g) through
impregnation with 6.5 cm3 of a solution of the preformed colloidal Pd NPs in
hexane (0.46 mg of Pd NP cm−3, nominal 1 wt% loading). The resulting solid was
stirred in solution for 1 h before solvent evaporation at room temperature to leave a
dry powder. The monometallic Pd material was produced identically except with
omission of the initial aqueous platinum salt impregnation.
Polystyrene colloidal nanospheres. Monodispersed non-crosslinked polystyrene
spheres were produced by adapting literature methods30. Styrene (105 cm3) was
washed five times with sodium hydroxide solution (0.1 M, 1:1 vol/vol) followed by
five washes with distilled water (1:1 vol/vol) to remove polymerization inhibitors.
The washed organic phase was added to nitrogen-degassed water (850 cm3) at
80 ◦C followed by dropwise addition of aqueous potassium persulphate solution
(0.24 M, 50 cm3) with 300 r.p.m. agitation. The reaction proceeded for 22 h, after
which the solution had turned white owing to the formation of polystyrene
nanospheres. Solid product was recovered and colloidal crystal arrangement was
induced by centrifugation (Hereus Multifuge X1 with Thermo Fiberlite F15-8x50cy
Fixed-Angle Rotor operated at 8,000 r.p.m./7,441g, for 1 h). The resulting highly
ordered polystyrene colloidal nanosphere crystalline matrix was finally ground to a
fine powder for use as the hard macropore-directing template.
Pt and Pd in mesopores. Platinum and palladium NPs were deposited selectively
within the mesopore domains by incipient wetness impregnation of the
hierarchically ordered macroporous–mesoporous SBA-15, in which the
macropores had been previously hydrophobized by triethoxy(octyl)silane, with an
aqueous solution of dihydrogen hexachloroplatinate and tetraamine palladium
nitrate. The parent support (0.25 g) was stirred in the aqueous salt solution (1 cm3,
0.0066 g Pt salt and 0.0070 g Pd salt, nominal 1 wt% loading of each) for 18 h in the
dark. A dry powder was obtained by gentle heating of the slurry at 50 ◦C for 10 h,
followed by calcination under air at 500 ◦C for 2 h and subsequent 200 ◦C reduction
under H2 (10 cm3 min−1) for 1 h, yielding Pt and Pd NPs.
Hierarchically ordered SBA-15. The hierarchical, bimodal silica support was
synthesized using a modified true liquid crystal templating technique16 to
incorporate the polystyrene nanospheres as macropore-directing hard templates.
Pluronic P123 (2 g) was sonicated with hydrochloric acid-acidified water (pH 2,
2 g) at 40 ◦C to a homogeneous gel. Tetramethoxysilane (4.08 cm3) was added and
stirred rapidly for 5 min at 800 r.p.m. to form a homogeneous liquid. Immediately
following this change in physical state the polystyrene colloidal crystals (6 g ground
to a fine powder) were added with agitation at 100 r.p.m. for 1 min to homogenize
the mix. The resulting viscous mixture was heated under vacuum (100 mbar) at
40 ◦C to remove the evolved methanol. After 2 h the solid was exposed to the
atmosphere at room temperature for 24 h to complete precursor condensation.
Pt and Pd in macropores. Platinum and palladium NPs were deposited selectively
within the macropore domains by incipient wetness impregnation of macropore
template-extracted macroporous–mesoporous SBA-15 (with the mesopores
blocked by the Pluronic P123 template) with an aqueous solution of dihydrogen
hexachloroplatinate and tetraamine palladium nitrate. The parent support (0.3 g)
was stirred in the aqueous salt solution (1 cm3, 0.0066 g Pt salt and 0.0070 g Pd salt,
nominal 1 wt% loading of each) for 18 h in the dark. A dry powder was obtained by
gentle heating of the slurry at 50 ◦C for 10 h, followed by calcination under air at
500 ◦C for 2 h and subsequent 200 ◦C reduction under H2 (10 cm3 min−1) for 1 h,
yielding Pt and Pd NPs.
Stepwise template extraction and macropore hydrophobization. The preceding
parent silica support (10 g) was stirred in toluene (100 cm3) at −8 ◦C for 1 min. The
solid was recovered by vacuum filtration and briefly washed with cold toluene. The
extraction protocol was subsequently repeated four times to fully extract the
polystyrene template, affording an empty macropore network without removal of
the P123 mesopore template. The resulting solid (2 g) was stirred in
triethoxy(octyl)silane (6 cm3) for 3 min and recovered by vacuum filtration before
drying overnight at room temperature. This step introduced hydrophobic character
selectivity into the macropores. The macroporous solid (∼2 g) was subsequently
refluxed in methanol (400 cm3) for 18 h to fully extract the Pluronic P123
mesopore-directing agent, and recovered by filtration and washing three times with
methanol, to yield a macroporous–mesoporous support with differing
Pt and Pd in mesopores and macropores. Platinum and palladium NPs were
deposited throughout both mesopores and macropores by incipient wetness
impregnation of the fully detemplated hierarchically ordered
macroporous–mesoporous SBA-15 (in which macropore hydrophobization
was omitted) with an aqueous solution of dihydrogen hexachloroplatinate and
tetraamine palladium nitrate. The parent support (0.25 g) was stirred in the aqueous
salt solution (1 cm3, 0.0066 g Pt salt and 0.0070 g Pd salt, nominal 1 wt% loading of
each) for 18 h in the dark. A dry powder was obtained by gentle heating of the slurry
at 50 ◦C for 10 h, followed by calcination under air at 500 ◦C for 2 h and subsequent
200 ◦C reduction under H2 (10 cm3 min−1) for 2 h, yielding Pt and Pd NPs.
hydrophobicity between the interconnected pore networks.
Palladium NP synthesis. Near-monodisperse palladium NPs of 5.6 ± 0.8 nm
diameter were prepared by adapting the protocol of ref. 18 to employ a readily
available borane complex, and extending the duration of particle ageing at 90 ◦C to
obtain larger NPs. Synthesis was carried out using standard Schlenk techniques
under an argon atmosphere. After evacuation of Pd(acac)2 (73 mg, Alfa Aesar) in a
3-neck round-bottom flask and backfilling with Ar (repeated three times),
oleylamine (15 cm3, Acros Organics, 80–90%) was added and the flask heated to
60 ◦C while stirring. Addition of borane triethylamine (0.52 cm3, Aldrich, 97%)
turned the solution from pale yellow to pale brown and was immediately followed
by heating to 90 ◦C within 15 min, during which time the solution turned black
indicating colloidal NP formation. Heating was continued at 90 ◦C for 90 min
before cooling to room temperature. Ethanol (Fisher Scientific, HPLC grade, about
30 cm3) was added to this suspension, precipitating the NPs, which were then
extracted by centrifugation (8,000 r.p.m., 20 min, 50 cm3 plastic centrifuge tube,
prewashed with ethanol). The resulting solid was redispersed in hexane (about
4 cm3, Fisher Scientific, reagent grade), and the volume of hexane was evaporated
to around 2 cm3 under flowing argon before precipitation by the addition of the
minimum quantity of ethanol and separation by centrifugation (6,000 r.p.m.,
10 min). Washing in about 2 cm3 hexane and precipitation with ethanol, followed
by centrifugation was repeated a further two times to remove any excess oleylamine
and other residual synthetic agents. The solid was finally redispersed and stored in
hexane (30 cm3) until further use. The Pd content of this NP solution was
determined by inductively coupled plasma optical emission spectrometry to be 11.0
± 0.12 mg (in 30 cm3), indicating that around 43% of the initial Pd is present in the
NPs after purification. The as-prepared NPs were characterized by TEM by casting
one droplet of NP solution onto a holey carbon-coated copper grid (Agar
Scientific) and evaporation to dryness. TEM imaging was performed using a JEOL
2100F FEG TEM with a Schottky field-emission source, equipped with an Oxford
INCAx-sight Si(Li) detector for energy-dispersive spectroscopy. The accelerating
voltage was 200 kV. The particle size distribution was obtained from imaging 6
different areas of the grid and measuring the diameter over 800 individual NPs. No
variation in particle size was apparent in different regions of the grid.
Pt in mesopores and macropores with Pd in macropores. Platinum NPs were
deposited within the mesopore and macropore domains by incipient wetness
impregnation of the fully detemplated hierarchically ordered
macroporous–mesoporous SBA-15 (in which macropore hydrophobization was
omitted) with an aqueous solution of dihydrogen hexachloroplatinate. The parent
support (0.25 g) was stirred in the aqueous salt solution (1 cm3, 0.0066 g Pt salt,
nominal 1 wt% loading) for 18 h in the dark. A dry powder was obtained by gentle
heating of the slurry at 50 ◦C for 10 h, followed by 100 ◦C reduction under H2
(10 cm3 min−1) for 2 h, yielding Pt NPs. Pd NPs were selectively deposited within
the macropores by impregnation with 5.4 cm3 of a solution of the preformed
colloidal Pd NPs in hexane (0.46 mg of Pd NP cm−3, nominal 1 wt% loading). The
solid was stirred in solution for 1 h before solvent evaporation at room temperature
to leave a dry powder.
Pd in mesopores and macropores and Pt in mesopores. Platinum NPs were
deposited selectively within the mesopore domains by incipient wetness
impregnation of the hydrophobic, hierarchically ordered
macroporous–mesoporous SBA-15 with an aqueous solution of dihydrogen
hexachloroplatinate. The parent support (0.25 g) was stirred in the aqueous salt
solution (1 cm3, 0.0066 g Pt salt, nominal 1 wt% loading) for 18 h in the dark. A dry
powder was obtained by gentle heating of the slurry at 50 ◦C for 10 h, followed by
100 ◦C reduction under H2 (10 cm3 min−1) for 1 h, yielding Pt NPs. The solid was
then calcined under air at 500 ◦C for 2 h to remove the organic octyl groups
(hydrophobicity) from the macropores. Palladium NPs were deposited within the
mesopore and macropore domains by incipient wetness impregnation with an
aqueous solution of tetraamine palladium nitrate. The solid (0.25 g) was stirred in
the aqueous salt solution (1 cm3, 0.0070 g Pd salt, nominal 1 wt% loading) for 18 h.
A dry powder was obtained by gentle heating of the slurry at 50 ◦C for 10 h,
followed by calcination under air at 500 ◦C for 2 h and subsequent 200 ◦C reduction
under H2 (10 cm3 min−1) for 2 h, yielding Pt and Pd NPs.
Platinum NP impregnation. Platinum NPs were deposited selectively within the
mesopore domains by incipient wetness impregnation of the hydrophobic,
NATURE MATERIALS | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT4478
LETTERS
Materials characterization. Nitrogen porosimetry was undertaken on a
Quantachrome Autosorb IQTPX porosimeter with analysis using ASiQwin v3.01
software. Samples were degassed at 150 ◦C for 12 h before recording N2
aberration probe corrector (CEOS GmbH) and a Bruker XFlash 5030 EDX. A
Hitachi SU8230 cold field-emission SEM operating at 20 kV was used for
simultaneous imaging of secondary, backscattered and dark-field scanning
transmission electron signals. Samples were prepared for microscopy by dispersion
in methanol and drop-casting onto a copper grid coated with a holey carbon
support film (Agar Scientific). Images were analysed using ImageJ 1.41 software.
adsorption/desorption isotherms. BET surface areas were calculated over the
relative pressure range 0.02–0.2. Mesopore properties were calculated applying the
BJH (Barrett–Joyner–Halenda) method to the desorption isotherm for relative
pressures >0.35, and fitting of isotherms to the relevant DFT (density functional
theory) kernel within the software package. Powder X-ray diffraction patterns were
Catalytic cascade oxidation. Catalytic aerobic selective oxidations were performed
in a 100 cm3 Buchi miniclave stirred batch reactor on a 75 cm3 scale at 150 ◦C.
Catalyst (12.5 mg) was added to reaction mixtures containing 4.2 mmol cinnamyl
alcohol (0.562 g), an internal standard (mesitylene, 0.1 cm3), and toluene solvent
(75 cm3) at 150 ◦C under 5 bar oxygen and stirring. Reactions were periodically
sampled for off-line gas chromatography analysis using a Varian 3800GC with an
8400 autosampler fitted with a CP-Sil5 CB column (15 m × 0.25 mm × 0.25 µm).
Conversion, selectivity and yields were calculated through calibration to reference
compounds and quoted ±2%. Turnover frequencies for cinnamyl alcohol
conversion and cinnamaldehyde production are quoted relative to the surface
density of PdO sites (determined by XPS), and for cinnamic acid production
relative to the surface density of PtO2 sites (determined by XPS), being the
respective active sites for selective oxidation over Pd NPs and Pt NPs (refs 23,24,26).
´
recorded using a Bruker D8 diffractometer employing a Cu Kα (1.54 Å) source
fitted with a Lynx eye high-speed strip detector. Low-angle patterns were recorded
for 2θ =0.3◦–8◦ with a step size of 0.01◦. Wide-angle patterns were recorded for
2θ =10◦–80◦ with a step size of 0.02◦. Contact angle measurements were carried
out on a Kruss DSA100 drop shape analyser, fitted with a digital camera for
continuous data collection. Water drop shapes were analysed 10 s after deposition
using DSA3 software. Thermogravimetric analysis was conducted using a Stanton
Redcroft STA 780 thermal analyser at 10 ◦C min−1 under flowing N2/O2 (80:20 v/v
20 cm3 min−1). Scanning electron microscopy (SEM) images were recorded on a
Carl ZEISS SUPRA 55-VP operating at 25 kV. Samples were supported on
aluminium stubs each backed with carbon tape. Transmission electron microscopy
(TEM) imaging of the silica support and preformed Pd NPs was performed using a
JEOL 2100F FEG TEM with a Schottky field-emission source, equipped with an
Oxford INCAx-sight Si(Li) detector for energy-dispersive spectroscopy (EDX).
High-resolution (scanning) transmission electron microscopy (S)TEM images were
recorded on either a FEI Tecnai F20 FEG TEM operating at 200 kV equipped with
an Oxford Instruments X-Max SDD EDX detector (10 nm diameter spot size), or a
JEOL 2100F FEG STEM operating at 200 keV and equipped with a spherical
References
30. Sen, T., Tiddy, G. J. T., Casci, J. L. & Anderson, M. W. Synthesis and
characterization of hierarchically ordered porous silica materials. Chem. Mater.
16, 2044–2054 (2004).
NATURE MATERIALS | www.nature.com/naturematerials
© 2015 Macmillan Publishers Limited. All rights reserved