Solvent-free oxidation of primary carbon-hydrogen bonds in toluene using Au-Pd alloy nanoparticles

Article abstract of DOI:10.1126/science.1198458

Selective oxidation of primary carbon-hydrogen bonds with oxygen is of crucial importance for the sustainable exploitation of available feedstocks. To date, heterogeneous catalysts have either shown low activity and/or selectivity or have required activated oxygen donors. We report here that supported gold-palladium (Au-Pd) nanoparticles on carbon or TiO₂ are active for the oxidation of the primary carbon-hydrogen bonds in toluene and related molecules, giving high selectivities to benzyl benzoate under mild solvent-free conditions. Differences between the catalytic activity of the Au-Pd nanoparticles on carbon and TiO₂ supports are rationalized in terms of the particle/support wetting behavior and the availability of exposed corner/edge sites.

Full text of DOI:10.1126/science.1198458

Solvent-Free Oxidation of Primary Carbon-Hydrogen Bonds in
Toluene Using Au-Pd Alloy Nanoparticles
Lokesh Kesavan et al.
Science 331, 195 (2011);
DOI: 10.1126/science.1198458
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 July 20, 2012 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/331/6014/195.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2011/01/12/331.6014.195.DC1.html
This article cites 22 articles, 3 of which can be accessed free:
http://www.sciencemag.org/content/331/6014/195.full.html#ref-list-1
This article appears in the following subject collections:
Chemistry
http://www.sciencemag.org/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
2011 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

REPORTS
3. S. W. Hell, Nat. Biotechnol. 21, 1347 (2003).
4. G. Foo, D. M. Palacios, J. Swartzlander Jr., Opt. Lett. 30,
3308 (2005).
16. B. J. McMorran, A. D. Cronin, N. J. Phys. 11, 033021 (2009).
17. V. Y. Bazhenov, M. V. Vasnetsov, M. S. Soskin, JETP Lett.
52, 429 (1990).
24. A. Jesacher, S. Fürhapter, S. Bernet, M. Ritsch-Marte,
Phys. Rev. Lett. 94, 233902 (2005).
25. R. Danev, H. Okawara, N. Usuda, K. Kametani,
18. N. R. Heckenberg, R. McDuff, C. P. Smith,
H. Rubinsztein-Dunlop, M. J. Wegener, Opt. Quantum
Electron. 24, S951 (1992).
5. D. G. Grier, Nature 424, 810 (2003).
6. A. Mair, A. Vaziri, G. Weihs, A. Zeilinger, Nature 412,
313 (2001).
7. M. F. Andersen et al., Phys. Rev. Lett. 97, 170406 (2006).
8. S. Franke-Arnold, L. Allen, M. Padgett, Laser Photon. Rev.
2, 299 (2008).
K. Nagayama, J. Biol. Phys. 28, 627 (2002).
26. K. Schultheiß, F. Pérez-Willard, B. Barton, D. Gerthsen,
R. R. Schröder, Rev. Sci. Instrum. 77, 033701 (2006).
27. We thank S. Adam, G. Gallatin, M. Stiles, and J. H. Scott
for useful discussions. Supported by the NIST-CNST
NanoFab, NIST Division 637 FIB-SEM, and the
19. See supporting material on Science Online.
20. Although the beams in this experiment do traverse
magnetic fields in the lenses, the fields are confined to
very small intervals relative to the entire path of the electrons.
21. Assuming a 1-nA, 300-keV electron beam with 0.3-eV
energy spread, the longitudinal extent of a electron’s
wave packet along the optical axis is about 3 mm,
whereas the average separation distance between
consecutive electrons is 4 cm.
9. K. Bliokh, Y. Bliokh, S. Savel’ev, F. Nori, Phys. Rev. Lett.
99, 190404 (2007).
NIST-CNST/UMD-NanoCenter Cooperative Agreement.
The authors declare no competing financial interests.
10. B. J. McMorran, thesis, University of Arizona (2009).
11. M. Uchida, A. Tonomura, Nature 464, 737 (2010).
12. J. Leach, E. Yao, M. J. Padgett, N. J. Phys. 6, 71 (2004).
13. J. Verbeeck, H. Tian, P. Schattschneider, Nature 467, 301 (2010).
14. B. McMorran, J. D. Perreault, T. A. Savas, A. Cronin,
Ultramicroscopy 106, 356 (2006).
Supporting Online Material
www.sciencemag.org/cgi/content/full/331/6014/192/DC1
Materials and Methods
22. A. V. Volyar, V. G. Shvedov, T. A. Fadeeva, Tech. Phys.
Figs. S1 and S2
Lett. 25, 203 (1999).
15. A. Cronin, B. McMorran, Phys. Rev. A 74, 061602(R)
(2006).
23. R. F. Egerton, Electron Energy-Loss Spectroscopy in the
Electron Microscope (Springer, New York, 1996).
7 October 2010; accepted 23 November 2010
10.1126/science.1198804
dation of primary alcohols using oxygen (20).
This catalyst operates by establishing a reactive
hydroperoxy intermediate. Because these inter-
mediates are known to be involved in the en-
zymatic oxidation of primary carbon-hydrogen
bonds (21), we reasoned that it should be feasible
for Au-Pd nanoparticles to be active for the
oxidation of the primary carbon-hydrogen bonds
in toluene. Here we show that Au-Pd alloy nano-
particles prepared by a sol immobilization tech-
nique can give significantly improved activity for
the oxidation of toluene under mild solvent-free
conditions. These catalysts have TONs that are a
factor of ~30 greater than those of previous hetero-
geneous catalysts for this reaction and also dis-
play a remarkably high selectivity to benzyl benzoate.
We started by investigating the oxidation of
toluene in an autoclave reactor with O2 in the
absence of catalyst in order to determine the
blank baseline rate. O2 is a di-radical and can ini-
Solvent-Free Oxidation of Primary
Carbon-Hydrogen Bonds in Toluene
Using Au-Pd Alloy Nanoparticles
Lokesh Kesavan,1 Ramchandra Tiruvalam,2 Mohd Hasbi Ab Rahim,1 Mohd Izham bin Saiman,1
Dan I. Enache,1 Robert L. Jenkins,1 Nikolaos Dimitratos,1 Jose A. Lopez-Sanchez,1
Stuart H. Taylor,1 David W. Knight,1 Christopher J. Kiely,2 Graham J. Hutchings1*
Selective oxidation of primary carbon-hydrogen bonds with oxygen is of crucial importance for the
sustainable exploitation of available feedstocks. To date, heterogeneous catalysts have either
shown low activity and/or selectivity or have required activated oxygen donors. We report here that
supported gold-palladium (Au-Pd) nanoparticles on carbon or TiO2 are active for the oxidation
of the primary carbon-hydrogen bonds in toluene and related molecules, giving high selectivities
to benzyl benzoate under mild solvent-free conditions. Differences between the catalytic activity
of the Au-Pd nanoparticles on carbon and TiO2 supports are rationalized in terms of the
particle/support wetting behavior and the availability of exposed corner/edge sites.
elective oxidation of primary carbon- cesses environmentally unfriendly. Vapor-phase tiate homogeneous oxidation processes at ele-
hydrogen bonds is of crucial importance in oxidation has been considered, but the conver- vated temperatures and pressures. We found that
activating raw materials to form intermedi- sion must be limited to avoid overoxidation to such processes become substantial at 190°C un-
S
ates and final products for use in the chemical, CO2 and other byproducts (10). Attempts to over- der our conditions (fig. S1) (18). This important
pharmaceutical, and agricultural business sectors come these problems have prompted investiga- observation suggests that the earlier studies con-
(1). One class of raw materials is alkyl aromatics; tion of the use of supercritical CO2 and ionic ducted at 190°C (table S1) (18) may not in fact
toluene, for example, the simplest member of this liquids, but these unfortunately resulted in low have been heterogeneously catalyzed. In view of
class, can be oxidized to benzyl alcohol, ben- conversions (11, 12).
the potential role of O2 di-radicals, the maximum
zaldehyde, benzoic acid, and benzyl benzoate. Often, heterogeneous catalysts are preferred reaction temperature in our studies has been
These products are commercially significant as over homogeneous catalysts, because these ma- restricted to 160°C; at this temperature, the blank
versatile intermediates in the manufacture of phar- terials can be readily separated from the reaction reaction in the absence of catalyst but in the
maceuticals, dyes, solvents, perfumes, plasticiz- mixture. Heterogeneous catalysts can also be read- presence of support is negligible for short reaction
ers, dyestuffs, preservatives, and flame retardants. ily used in flow reactors, facilitating the efficient times (Table 1, entries 1, 3, and 4) and is very low
Commercially, benzaldehyde is produced by the production of materials using continuous pro- at longer reaction times (Table 1, entry 2). We
chlorination of toluene followed by saponifica- cesses. For the oxidation of toluene, there have initially investigated Au-Pd/TiO2 catalysts, pre-
tion (2), and benzoic acid is produced by the been many attempts to find a suitable oxidation pared by impregnation, because these had previ-
liquid-phase cobalt-catalyzed reaction of toluene catalyst, and to date these have used copper and ously been shown to be very active for H2O2
using oxygen at 165°C with acetic acid as sol- manganese (13–15), cobalt (16), or chromium (17) synthesis (19) and alcohol oxidation (20); how-
vent, but the conversion has to be limited to catalysts, but all of these perform very poorly ever, these catalysts were not found to be par-
<15% to retain high selectivities (3–9). The use with turnover numbers (TONs: mole of product ticularly active for toluene oxidation, although
of halogens and acidic solvents makes these pro- per mole of metal catalyst) of less than 100, even they did not produce any CO2 (table S2) (18).
at temperatures in excess of 190°C (table S1) (18).
There is clearly a need to develop heterogeneous impregnation method can be relatively large
catalysts for toluene oxidation that have greatly (typically >6 nm) and have substantial composi-
The Au-Pd nanoparticles synthesized by the
1Cardiff Catalysis Institute, School of Chemistry, Cardiff Uni-
versity, Main Building, Park Place, Cardiff CF10 3AT, UK.
2Department of Materials Science and Engineering, Lehigh
University, 5 East Packer Avenue, Bethlehem, PA 18015–3195,
USA.
improved activity while retaining selectivity.
tional variations, which limits their reactivity.
Recently, we have shown that Au-Pd alloy Therefore, in order to design more effective cat-
nanoparticles are very effective for the direct alysts, we decided to investigate Au-Pd nano-
synthesis of hydrogen peroxide (19) and the oxi- particles with smaller median particle sizes (2 to
*To whom correspondence should be addressed. E-mail:
hutch@cardiff.ac.uk
www.sciencemag.org SCIENCE VOL 331 14 JANUARY 2011
195

REPORTS
5 nm) and a more controlled composition. These tigated the effect of the Au-Pd molar ratio (Table zoic acid, and benzyl benzoate. By itself, Au was
were prepared by sol immobilization of Au-Pd 1, entries 5 to 12) using these sol-immobilized not active for this reaction, but the addition of Pd
colloids, with carbon and TiO2 as supports (18), materials supported on carbon. For all catalysts, significantly enhanced the conversion, demon-
a method that afforded tightly controlled particle no CO2 formation was observed; the only pro- strating a clear synergistic effect for the Au-Pd
size distributions and composition. We inves- ducts were benzyl alcohol, benzaldehyde, ben- catalysts as compared with the monometallic species.
Table 1. Comparison of catalytic activity for the oxidation of toluene in the absence of solvent with O2. Catalyst mass varied between 0.2 and 0.6 g to give a
substrate/metal molar ratio of 6500. Toluene, 10 to 20 ml; stirring rate, 1500 rpm; TON calculated on the basis of the total metal. All metal catalysts were prepared
using the sol immobilization method (18). T, temperature.
Selectivity (%)
Entry
Catalyst
Metal
(%)
Au:Pd
molar ratio
T (°C)
PO
(bar)
Time
(hours)
Conversion
(%)
Benzyl
alcohol
Benzaldehyde
Benzoic
acid
Benzyl
benzoate
TON
2
1
2
3
4
5
6
7
8
None*
None*
Carbon*
TiO2*
Au/C*
-
-
-
-
-
-
-
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
120
120
80
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
6
7
48
7
7
7
7
7
7
7
7
7
7
7
7
7
48
48
48
48
48
48
0.5
7
0.2
2.9
0.1
0.2
0.2
0.3
1.5
4.8
5.3
5.2
4.3
1.6
1.6
2.1
2.2
50.8
24.1
10.6
4.0
18.3
5.8
11.4
0.1
9.0
28.4
1.8
0.9
1.2
1.9
9.6
3.9
0.8
2.9
2.2
0.1
0.5
0.2
1.1
8.6
0.1
2.1
0.3
0.1
0.2
69.4
43.6
72.6
6.0
81.9
57.6
63.4
12.7
8.3
1.9
38.2
0.1
85.1
0
6.2
3.1
10.3
11.1
10.3
7.3
3.3
11.4
1.0
2.3
4.5
2.8
13.1
4.8
0.1
2.9
10.4
12.4
16.0
8.9
8.1
7.8
-
-
-
-
11
17
95
310
350
340
280
105
105
135
141
3300
1570
690
260
60
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1:0
7:1
3:1
1:1.85
1:2
1:3
1:7
0:1
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
1:1.85
Au-Pd/C*
Au-Pd/C*
Au-Pd/C*
Au-Pd/C*
Au-Pd/C*
Au-Pd/C*
Pd/C*
Au/C+Pd/C†
Au-Pd/TiO2*
Au-Pd/TiO2*
Au-Pd/C‡
Au-Pd/TiO2‡
Au-Pd/C‡
Au-Pd/TiO2‡
Au-Pd/C‡
Au-Pd/C‡
Au-Pd/C§
Au-Pd/C§
Au-Pd/C§
Au-Pd/C§
31.4
76.1
79.3
79.3
69.5
36.4
61.6
89.5
89.0
94.3
95.5
79.7
88.1
57.2
91.4
85.5
94.8
91.8
85.5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
8.5
13.6
56.4
26.2
6.6
6.5
1.1
1.2
7.1
6.0
34.2
5.7
0.9
4.9
1.8
21.5
82.9
94.4
160
160
160
160
160
320
29
350
1347
1534
10
10
10
10
11.6
1.8
0.8
0.8
3.1
7.3
13.3
24
27
1.0
*Reaction conditions: toluene, 20 ml; substrate/metal molar ratio, 6500; mass of catalyst, variable.
†Reaction of a physical mixture composed of 1 wt % Au/C and 1 wt % Pd/C. Toluene,
20 ml; substrate/metal molar ratio, 6500.
‡Reaction conditions: toluene, 10 ml; substrate/metal molar ratio, 6500; mass of catalyst, 0.2 g.
§Reaction conditions: toluene, 20 ml;
substrate/metal molar ratio, 1625; mass of catalyst, 1.6 g.
Table 2. Comparison of the catalytic activity for oxidation of substituted toluenes in the absence of solvent with O2. Catalyst mass, 0.4 g; substrate, 20 ml; stirring
rate, 1500 rpm; 160°C; 10 bar O2; TOF and TON calculated on the basis of the total metal. Au-Pd/C (1:1.85 Au:Pd molar ratio), 1 wt % total metal was prepared
using the sol immobilization method (18).
Selectivity (%)
Substrate
Time
(hours)
Conversion
(%)
Benzyl
alcohol
Benzaldehyde
Benzoic
acid
Benzyl benzoate
TON
Toluene
1
4
7
1.1
2.2
4.4
2.7
2.2
1.0
27.5
18.7
12.4
6.9
8.6
10.4
62.9
70.5
76.2
284
n-Methoxy
benzyl
alcohol
0.8
n-Methoxy
benzaldehyde
n-Methoxy
benzoic
acid
30.3
25.8
26.9
46.5
30.6
43.3
n-Methoxybenzyl
n-methoxybenzoate
Toluene
Esters
C-C
products
4- Methoxytoluene
3-Methoxytoluene
2-Methoxytoluene
1
4
7
1
4
7
1
4
7
2.3
6.1
10.6
1.7
2.0
3.7
3.6
6.4
11.1
18.7
11.8
7.4
16.4
33.3
12.6
8.8
24.1
43.5
49.0
8.1
2.2
17.0
0.7
0.9
0.2
0.1
7.4
19.5
6.8
22.7
15.7
11.8
7.1
5.0
4.5
1.2
0.8
1.2
0.3
0.4
0.8
18.1
12.9
11.6
19.0
12.5
19.0
2.3
0.8
0.5
1.4
1.1
0.9
1.4
1.0
0.8
581
204
619
63.8
66.8
61.1
10.0
13.0
3.6
7.3
2.5
5.2
196
14 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org

REPORTS
A physical mixture of the separate Au/C and sion (table S6) (18), confirming the wider ap- investigated using a lower substrate/metal molar
Pd/C catalysts showed no enhancement (Table 1, plicability of our catalysts. ratio and showed that the conversion continued to
entry 13), highlighting the molecular-scale nature Given these initial promising results, we in- increase steadily, fully depleting the toluene after
of the synergy. The observed synergy is due to vestigated the oxidation of toluene at longer re- 110 hours (Fig. 1), while the selectivity to benzyl
electronic and morphological features because action times (Table 1, entries 16 and 17). In this benzoate also increased. Monometallic Au and
transmission electron microscopy (TEM) shows regime, much higher toluene conversions were Pd catalysts showed very low conversion (8%)
that the mean particle size of the nanoparticles attained and no CO2 was observed. Once again over 110 hours (table S8) (18). The toluene con-
decreases slightly on the addition of Au to Pd only four products were observed: benzyl al- version was also found to increase linearly with
(table S3). When Au-rich catalysts were used, cohol, benzaldehyde, and benzoic acid in trace catalyst mass (fig. S3). Hence, with appropriate
some benzyl alcohol was formed, but this was amounts and benzyl benzoate as the dominant tuning of catalyst loading and reaction condi-
readily oxidized to the main product benzalde- product (~95%) for both catalysts. The carbon- tions, we are confident that complete and selec-
hyde, as this sequential oxidation is rapid at this supported catalyst was typically twice as active tive conversion of toluene to desirable products
temperature (20). We consider the initial oxida- over this longer time scale, exhibiting a TON of can be achieved (Table 1, entries 22 to 25 show
tion of toluene to involve a surface hydroperoxy 3300 with ~3150 mol of benzyl benzoate per results of an initial optimization). Furthermore, by
intermediate formed from the interaction of the mole of metal produced over the reaction period. increasing the reaction time at 140°C, we can ob-
metal with oxygen and toluene. However, as the We investigated the use of lower reaction tem- tain a 95% yield of benzyl benzoate (table S9) (18).
fraction of Pd in the alloy was increased, the se- peratures (Table 1, entries 18 though 20) and
We consider that the high selectivity to benzyl
lectivity to benzyl benzoate became dominant. The observed some activity, albeit slight, even at 80°C. benzoate 6 could result from four possible mech-
optimum catalyst composition found was a 1:2 In a further set of experiments, we attempted to anisms: (i) coupling of the aldehyde 3 and the
Au:Pd molar ratio (approximately 1:1 by weight). enhance the activity of the catalyst at 80°C by alcohol 2 to give the hemiacetal 5, followed by
For this catalyst, the turnover frequency (TOF: using t-butyl hydroperoxide as a co-oxidant (table oxidation to the ester (Scheme 1); (ii) direct ther-
mole of product formed per mole of metal per S7) and found that the conversion increased mal (catalyzed) dehydrative esterification between
hour) of toluene oxidation after 7 hours of re- appreciably when both the carbon- and TiO2- acid 4 and alcohol 2 (Scheme 1); (iii) Cannizzaro
action was ~50. We also investigated this Au-Pd supported catalysts were used; TON values of reaction between alcohol 2 and aldehyde 3 via
composition using a TiO2-supported material 850 to 1200 can be achieved even under these alkoxide 7 (Scheme 2); and (iv) Tishchenko cou-
(Table 1, entry 14) and found that it was also mild conditions. However, using this hydro- pling of two aldehydes 3 via “dimer” 8 directly to
active, but less so than the carbon-supported ma- peroxide as a co-oxidant generated a different the ester 6 (Scheme 3).
terial, displaying a TOF of 20 hour−1 after 7 hours product profile. At long reaction times and high
of reaction. conversions, benzoic acid became the main pro- shown that methyl formate is formed on an Au
Recently, Bäumer and co-workers (22) have
To demonstrate the general applicability of duct, together with appreciable amounts of benz- catalyst when methanol is oxidized with O2
the AuPd/C catalyst, we oxidized 2-, 3-, and 4- aldehyde and benzyl alcohol; only negligible under very dilute conditions; we consider that
methoxytoluene (Table 2), and 2-, 3-, and 4- amounts of benzyl benzoate were formed. As this precedent favors mechanism (i) in the present
nitrotoluene (table S4) (18). No CO2 was with O2 as the sole oxidant, when using the case. We performed several control experiments
observed, and the reactivity trend (4-methoxy- hydroperoxide co-oxidant, the carbon-supported to rule out the other three possible pathways.
~ 2-methoxy- >3-methoxy ~ toluene > 2-nitro- > catalyst was about twice as active as the TiO2- Heating the oxidized intermediates with O2 in the
3-nitro ~ 4-nitro) is indicative of the involvement supported Au-Pd nanoparticles.
absence of catalyst gave little ester (table S10),
of electron-deficient intermediate(s). Additional The reaction profile over an initial 7-hour establishing the involvement of the catalyst in
products formed were identified as a family of reaction period was investigated using a higher steps beyond the initial toluene-to-alcohol oxida-
esters and C-C coupling products (table S5) (18). substrate-to-metal ratio at 160°C (fig. S2). This tion. Heating mixtures of alcohol 2 and acid 4 in
We have also investigated the reaction of xy- experiment showed that the toluene conversion the absence of O2, with and without catalyst, did
lenes; the catalysts are equally effective for the increases linearly over this time scale and that the not result in ester formation (tables S11 and S12),
oxidation of this substrate and formed the al- selectivity to benzyl benzoate increases as the ruling out direct esterification (4 → 6; Scheme 1).
dehyde, acid, and esters as products, with the selectivity to benzaldehyde decreases. The re- Similarly, treating alcohol 2 with aldehyde 3 also
relative amounts being dependent on the conver- action profile over a longer time scale was also failed to produce ester 6 (tables S11 and S12),
ruling out the Cannizzaro mechanism (Scheme
Fig. 1. Toluene conver-
sion and selectivity to ben-
zyl alcohol, benzaldehyde,
benzoic acid, and benzyl
benzoate: reaction tem-
perature, 160°C (433 K);
0.1 MPa partial pressure
2). Finally, we investigated reactions of the alde-
hyde 3 alone in the absence of O2, with and
without catalyst, and again observed no ester 6
formation (tables S11 and S12), precluding the
Tishchenko mechanism (Scheme 3). Hence we
conclude that the high selectivities observed are
consistent with the involvement of the hemiacetal
5 (Scheme 1).
of oxygen (PO ); 20 ml of
2
toluene; 0.8 g of catalyst
[1 weight % (wt %)
Au-Pd/C prepared by sol im-
mobilization with a 1:1.85
Au/Pd ratio]; toluene/metal
molar ratio of 3250; and
reaction time, 110 hours.
Open circles indicate con-
version, squares indicate
selectivity to benzyl alco-
hol, diamonds indicate
selectivity to benzaldehyde,
To determine the origin of the differences in
activity between the two catalysts, we have char-
acterized the common starting sol (18) and the
two sol-immobilized materials using EM. Figure
S4 (18) shows a representative scanning TEM
(STEM) high-angle annular dark field (HAADF)
image of the starting colloidal particles deposited
onto a carbon film. The nanoparticles have a
mean particle size of 2.9 nm (fig. S5A) and were
found to be homogeneous Au-Pd alloys. More
than 80% of the particles were icosahedral or
decahedral (multiply twinned), with the remain-
Time (h)
triangles indicate selectivity to benzoic acid, and solid circles indicate selectivity to benzyl benzoate.
www.sciencemag.org SCIENCE VOL 331 14 JANUARY 2011
197

REPORTS
der being cub-octahedral or just single/double- character. Such particles preferentially exposed G to J) and comparatively few single/double-
twinned in character. They were also distinctly distinct {111}- and {200}-type facets. The larger twinned (Fig. 2K) and cub-octahedral (Fig. 2L)
rounded in shape because they were still coated particles (Fig. 2F) still tended to be multiply particles present. It is plausible that the strong
with protective polyvinyl alcohol ligands. After twinned and exclusively exposed {111}-type fac- interaction of the Au-Pd metal with the TiO2
deposition of the colloids onto either activated et planes. In addition, the Au-Pd particles tended support in the former case templates many of the
amorphous carbon or TiO2, the sol-immobilized to form an extended flat interface structure with smaller multiply twinned particles to restructure
material was dried at 120°C for 3 hours. The the crystalline TiO2 substrate (fig. S6), which as cub-octahedral particles or simple single/
low-temperature drying process in each case caused could serve to improve particle adhesion and in- double-twinned crystals during the drying step.
a very modest size increase in the Au-Pd par- hibit sintering at higher temperatures.
ticles, leading to HAADF-measured mean sizes In contrast, the corresponding HAADF im- of the Au-Pd catalysts immobilized on carbon
for the TiO2- and carbon-supported catalysts of ages (Fig. 2, G to I) from the carbon-supported and TiO2 showed them to have a Pd/Au ratio of
3.9 and 3.7 nm, respectively (fig. S5 B and C). samples show that the Au-Pd particles are more 2.1 and 2.2, respectively, and confirmed that in
X-ray photoelectron spectroscopy (XPS) analysis
The most important difference noted between rounded and have a much lower ability to wet the both cases, the Pd was predominantly in the
the two samples relates to the morphology of the amorphous carbon support. The distribution of metallic state (23).
Au-Pd particles. In contrast to the starting colloid, Au-Pd particle morphologies on the carbon sup-
The AuPd/C sample had approximately
many of the smaller particles were now found to port was much closer to that of the starting double the catalytic activity of the AuPd/TiO2
be highly faceted, and primarily cub-octahedral colloid, with icosahedral and decahedral (five- sample, despite having a very similar size dis-
(Fig. 2, A to D) or singly twinned (Fig. 2E) in fold-twinned) particles predominating (Fig. 2, tribution of Au-Pd particles. This suggests that
simple metal surface area considerations are not
dominating the catalytic activity in this instance,
because the total numbers of exposed surface
Scheme 1
atoms are almost identical (24), and the TONs
per surface-exposed atoms for the most-active
catalysts are 1.03 × 104 for Au-Pd/C (Table 1,
entry 16) and 0.56 × 104 for Au-Pd/TiO2 (Table
1, entry 17) (tables S14 and S15) (18). In ad-
dition, the similarity in surface composition, as
evidenced by XPS, does not provide a clue as to
the source of the activity difference. The Au-Pd/
TiO2 catalyst does clearly have more support/
particle periphery sites relative to the Au-Pd/C
catalyst, by virtue of its flatter, better wetting in-
terface. However, such periphery sites are prob-
ably not implicated in the catalytic process in
this instance, because the Au-Pd/TiO2 catalyst
displays the lower activity. It is possible that the
disparity in catalytic activity may be related to
subtle differences in the number of low coordi-
nation facet-edge and corner sites in the two
cases. The flatter, more faceted, Au-Pd particles
have fewer of these low coordination number
sites, whereas the more irregular, rougher Au-Pd
Scheme 2
Scheme 3
A
B
C
D
E
F
L
G
H
I
J
K
Fig. 2. Representative (A to F) STEM HAADF micrographs of Au-Pd nanoparticles in the Au-Pd/TiO2 sol-immobilized sample. Representative (G to L) STEM
HAADF micrographs of Au-Pd nanoparticles in the Au-Pd/C sol-immobilized sample.
198
14 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org

REPORTS
7. H. V. Borgaonkar, S. R. Raverkar, S. B. Chandalia,
particles have substantially more of them. If these and fresh toluene was added. No metal was ob-
low coordination number corner and/or edge po- served to have leached into the liquid phase
sitions are implicated as active sites for toluene during reaction, and the decanted liquid showed
oxidation, then their relatively higher occur- no further reaction in the absence of catalyst
rence in the Au-Pd/C sample could account for (fig. S7 and table S16). The reaction was then
the superior performance displayed by this cat- allowed to proceed for a further 7 hours, and the
alyst. Another possible explanation could lie in whole process was repeated a further two times.
the difference in the distribution of Au-Pd par- The reaction profile obtained with the decanta-
ticle morphologies found in the two catalyst tion experiments was identical to that obtained
samples. The Au-Pd/C catalyst predominantly with the fresh catalyst (fig. S8). Detailed STEM
has multiply twinned (icosahedral and decahe- characterization shows that there is minimal
dral) particles, which tend to have {111} facet particle growth or morphology change for the
terminations. In comparison, the Au-Pd/TiO2 Au-Pd/C catalysts when studied over an ex-
materials show an increased fraction of cub- tended reaction period (Fig. 1), during which
octahedral and singly/doubly twinned particles, catalysts were recovered after 31 and 65 hours of
which exhibit mixed {100}/{111} facet ter- reaction, followed by two reuse cycles of 7 hours
minations. Hence, the increasing proportion of (figs. S9 and S10). Therefore, it is clear that any
{100}-type facets in the Au-Pd/TiO2 sample cor- sintering or structural modification of these highly
relates with a lowering of the catalytic activity, active catalysts is minimal, and we consider them
Ind. Eng. Chem. Prod. Res. Dev. 23, 455 (1984).
8. M. L. Kantam et al., Catal. Lett. 81, 223 (2002).
9. Snia-Viscosa, Hydrocarbon Proc. 134, 210 (1977).
10. F. Konietzni, U. Kolb, U. Dingerdissen, W. F. Maier,
J. Catal. 176, 527 (1998).
11. J. Zhu, A. Robertson, S. C. Tsang, Chem. Commun.
(Camb.) 18, 2044 (2002).
12. K. R. Seddon, A. Stark, Green Chem. 4, 119 (2002).
13. X. Li et al., Catal. Lett. 110, 149 (2006).
14. F. Wang et al., Adv. Synth. Catal. 347, 1987 (2005).
15. J. Gao, X. Tong, X. Li, H. Miao, J. Xu, J. Chem. Technol.
Biotechnol. 2002, 620 (2007).
16. R. L. Brutchey, I. J. Drake, A. T. Bell, T. Tilley,
Chem. Commun. (Camb.) 2005, 3736 (2005).
17. A. P. Singh, T. Selvam, J. Mol. Catal. Chem. 113,
489 (1996).
18. See supporting on Science Online for detailed methods.
19. J. K. Edwards et al., Science 323, 1037 (2009).
20. D. I. Enache et al., Science 311, 362 (2006).
21. J. Colby, D. I. Stirling, H. Dalton, Biochem. J. 165,
395 (1977).
22. A. Wittstock, V. Zielasek, J. Biener, C. M. Friend,
M. Bäumer, Science 327, 319 (2010).
23. N. Dimitratos et al., Phys. Chem. Chem. Phys. 11,
5142 (2009).
24. A. Borodziński, M. Bonarowska, Langmuir 13, 5613 (1997).
25. We acknowledge the support of the Dow Chemical
Company through the Dow Methane Challenge.
and preparation strategies need to avoid them.
In a final set of experiments, we investigated
the stability of the catalysts, because it is crucial
to confirm that high-activity catalysts can be re-
used. With the Au-Pd/TiO2 catalyst, the reaction
was stopped after 7 hours, and the catalyst was
recovered by decantation. Identical conversion
was obtained on reuse of the Au-Pd/TiO2 catalyst
(Table 1, entries 14 and 15). For the Au-Pd/C
catalyst, the reaction was stopped after 7 hours
and the catalyst was allowed to settle. The liquid
phase was then carefully removed by decantation
to be stable and reusable.
References and Notes
1. R. A. Sheldon, J. K. Kochi, Metal-Catalyzed Oxidations of
Organic Compounds (Academic Press, New York, 1981).
2. W. Partenheimer, Catal. Today 23, 69 (1995).
3. Y. Ishii, S. Sakaguchi, T. Iwahama, Adv. Synth. Catal.
343, 393 (2001).
Supporting Online Material
www.sciencemag.org/cgi/content/full/331/6014/195/DC1
Materials and Methods
Figs. S1 to S10
Tables S1 to S16
4. K. Nomiya, K. Hashino, Y. Nemoto, M. Watanabe,
J. Mol. Catal. Chem. 176, 79 (2001).
5. T. G. Carrell, S. Cohen, G. C. Dismukes, J. Mol. Catal.
Chem. 187, 3 (2002).
6. S. Saravanamurugan, M. Palanichany, V. Murugesan,
References
29 September 2010; accepted 14 December 2010
10.1126/science.1198458
Appl. Catal. A Gen. 273, 143 (2004).
between these fields. Therefore, combining am-
phiphilicity with colloidal rigidity, we study “Janus
spheres” that are hydrophobic on one hemi-
sphere and negatively charged on the other.
An earlier publication from this laboratory de-
scribed some structures that these hybrid ma-
terials form (18). Here, we address the kinetics of
self-assembly at the single-particle level, showing
that small, kinetically favored isomers join to
form highly ordered but nonequilibrium large-
scale structures.
A critical design rule is that the range of
interparticle interactions (hydrophobic attraction
and electrostatic repulsion) must be short relative
to particle size and that the interactions must be
reversible. Clustering then favors densely packed
Supracolloidal Reaction Kinetics of
Janus Spheres
Qian Chen,1 Jonathan K. Whitmer,1,2 Shan Jiang,1 Sung Chul Bae,1 Erik Luijten,3,4* Steve Granick1,2,5
*
Clusters in the form of aggregates of a small number of elemental units display structural,
thermodynamic, and dynamic properties different from those of bulk materials. We studied the kinetic
pathways of self-assembly of “Janus spheres” with hemispherical hydrophobic attraction and found key
differences from those characteristic of molecular amphiphiles. Experimental visualization combined
with theory and molecular dynamics simulation shows that small, kinetically favored isomers fuse,
before they equilibrate, into fibrillar triple helices with at most six nearest neighbors per particle. The time
scales of colloidal rearrangement combined with the directional interactions resulting from Janus geometry
make this a prototypical system to elucidate, on a mechanistic level and with single-particle kinetic
resolution, how chemical anisotropy and reaction kinetics coordinate to generate highly ordered structures.
lusters, an intermediate level of matter structures that result from isotropic interactions be- structures with at most six nearest neighbors per
between building block (atom, molecule, tween building blocks are well understood (5–8), particle, in contrast to the more open and less
or particle) and bulk phase, are found ubiq- it is more challenging to understand clusters formed ordered structures formed by particles whose in-
C
uitously in nature and technology—for example, from the common case of directional noncovalent teraction range is larger (13). At very low salt
in the nucleation of bulk phases (1), nanoparticles interactions (9–17). On this question, we note that concentrations, particles repel one another electro-
(2), and protein aggregates in biology (3,4). Whereas for molecular amphiphiles, such as surfactants, statically, whereas at high salt concentrations, van
phospholipids, and many block copolymers, the der Waals forces cause the particles to aggregate
segregation of their polar and nonpolar portions irreversibly (19). Therefore, we consider interme-
is a major mechanism steering the spontaneous diate concentrations of monovalent salt at which
formation of microstructured mesophases with amphiphilic clusters self-assemble (20).
1Department of Materials Science and Engineering, University
of Illinois, Urbana, IL 61801, USA. 2Department of Physics,
University of Illinois, Urbana, IL 61801, USA. 3Department of
Materials Science and Engineering, Northwestern University,
Evanston, IL 60208, USA. 4Department of Engineering Sciences
and Applied Mathematics, Northwestern University, Evanston,
IL 60208, USA. 5Department of Chemistry, University of Illinois,
Urbana, IL 61801, USA.
fascinating and useful structures (9–11). Sim-
If the hydrophobic patch is too small, as-
ilarly, colloidal particles, larger than molecules sembly admits clusters composed of at most four
but small enough to sustain Brownian motion, particles. However, increase in patch size allows
also assemble into clusters owing to directional such clusters to grow into larger assemblies, with
noncovalent interactions (12–17). A largely un- two constraints: First, particles must approach
solved problem is the question of commonality closely enough to experience hydrophobic attrac-
*To whom correspondence should be addressed. E-mail:
luijten@northwestern.edu (E.L.); sgranick@illinois.edu (S.G.)
www.sciencemag.org SCIENCE VOL 331 14 JANUARY 2011
199