Al2O3 nanocomposites

Nano Today (2009) 4, 220—226

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nanotoday

RAPID COMMUNICATION

Selective catalytic reduction of nitric oxide by propene over In2O3—Ga2O3/Al2O3 nanocomposites Pemakorn Pitukmanorom a, Jackie Y. Ying a,b,∗ a b

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, The Nanos, Singapore 138669, Singapore

Received 6 January 2009; accepted 6 March 2009 Available online 21 April 2009

KEYWORDS Selective catalytic reduction; Nitric oxide; Propene

Summary In2 O3 —Ga2 O3 /Al2 O3 nanocomposite catalysts combined the advantages of good propene selectivity offered by Ga2 O3 /Al2 O3 and high propene conversion provided by In2 O3 /Al2 O3 to yield catalysts with excellent de-NOx activity. A high NO conversion to N2 of 80% was achieved over the catalyst with 2 wt% In and 8 wt% Ga. This NO conversion was attained even in the presence of 15% O2 and 10% H2 O. The In2 O3 —Ga2 O3 /Al2 O3 nanocomposite remained active in the presence of SO2 . This combination of high activity, hydrothermal stability and sulfur resistance made our nanocomposite catalysts very interesting for applications in the selective catalytic reduction (SCR) of NOx . © 2009 Elsevier Ltd. All rights reserved.

Due to increasingly stringent government regulations on vehicle nitrogen oxides (NOx ) emission levels [1], the selective catalytic reduction (SCR) of NOx to N2 by hydrocarbons has gained a great deal of attention. In particular, the removal of NOx from lean-burn and diesel engines operating under an oxygen-rich atmosphere presents significant challenges since conventional three-way catalysts are ineffective in this environment [2—4]. Moreover, the presence of water vapor and sulfur oxides (SOx ) in the exhaust stream both inhibits catalyst activity and results in long-term catalyst instability [5]. Therefore, it is necessary to develop catalyst that is both active and stable under actual exhaust conditions.

∗

Corresponding author. Fax: +65 6478 9020. E-mail address: [email protected] (J.Y. Ying).

Since it was first reported that Cu-ZSM-5 catalyzed the SCR of NOx by hydrocarbons [6,7], numerous catalysts, including zeolites, supported noble metals and metal oxides, have been extensively examined for this reaction. Among the catalysts examined, metal oxides supported on alumina have shown the most promise for practical applications due to their superior hydrothermal stability compared to zeolites. The effectiveness of this type of catalysts for NOx reduction depends strongly on the dispersion and the local structure of metal additives, which in turn depend on the loading and catalyst preparation technique, as evident in the studies on Sn/Al2 O3 [8,9], Co/Al2 O3 [10], Ag/Al2 O3 [11,12], Ga/Al2 O3 [13,14], and In/Al2 O3 [15,16]. Although the mechanism of this reaction has not been fully elucidated, recent studies have emphasized the role of adsorbed nitrates and oxygenated hydrocarbons as intermediates in the SCR of NOx by propene [17—21]. It appears that for a catalyst to be effective, it must be able to activate the propene molecule,

1748-0132/$ — see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.nantod.2009.03.001

Selective catalytic reduction of nitric oxide by propene over In2O3—Ga2O3/Al2O3 nanocomposites

Scheme 1

Selective reduction of NO by C3 H6 .

as well as adsorb NOx on its surface at the temperature range of interest. Among the various Al2 O3 -supported materials that we investigated, which included Co, Cu, Zn, Ag and Sn, Ga2 O3 /Al2 O3 and In2 O3 /Al2 O3 nanocomposites were of the greatest interest when their performances were evaluated according to their N2 yield and propene selectivity. N2 yield represented the desirable NO conversion, and was given by the ratio of N2 detected to the theoretical amount of N2 that could be produced given the amount of NO in the feed stream. Propene selectivity indicated the efficiency of a catalyst in utilizing the available propene to react with NO instead of O2 , and was given by the ratio of propene that reacted with NO to the total propene converted by either NO (in NOx reduction) or O2 (in propene combustion). These competing reactions are represented in Scheme 1. Since the precise mechanism of NOx reduction by propene was not known, the propene selectivity reported in this work was calculated according to Scheme 1, which assumed that no O2 took part in the SCR of NO.

Experimental To achieve high surface area materials with good dispersion of active components, alumina-based nanocomposite catalysts were prepared using a sequential chemical precipitation technique. In a typical synthesis, 15 g of aluminum nitrate nonahydrate in 100 ml of water was slowly added to a base solution, which consisted of 100 ml of water, 100 ml of ethanol, and 50 ml of tetraethylammonium hydroxide (35% w/v), to yield an aluminum hydroxide precipitate. The precipitation was conducted at room temperature and a pH of 11. After aging for 24 h, 50 ml of metal salts solution with appropriate concentrations of active components were then added. Metal nitrates were used as precursors, except in the case of tin, where tin chloride was employed. The precipitation of active components was done at a pH of 9 for gallium nitrate and indium nitrate. The suspension was stirred for an additional 24 h, and the resulting materials were centrifuged, washed and dried, before calcination in air at 700 ◦ C for 3 h. The powders were crushed and sieved to a mesh size of 80—120. The metal loadings of all the catalysts in this work were reported in wt% metal. To study the effects of synthesis methods, catalysts were also prepared by impregnation and sol—gel methods. Impregnated catalysts were prepared by adding an aqueous

221

solution of metal nitrates directly onto the calcined alumina support. Mixed oxide catalysts were synthesized by sol—gel method by adding a solution of metal nitrates in ethylene glycol to the boehmite sol prepared according to Maeda et al. [22]. All the catalysts were calcined in air at 700 ◦ C for 3 h. Catalytic activities under steady-state conditions were measured in a quartz tube fixed-bed reactor. A feed gas with molar composition of 0.1% NO, 0.1% C3 H6 , 15% O2 , and 0% or 10% H2 O with helium as a balance was fed to 0.2 g of catalyst at a total flow rate of 200 cm3 min−1 , which corresponded to a contact time of 0.06 g s cm−3 or a space velocity of ∼30,000 h−1 . Water was introduced into the gas stream through a syringe pump. The reactor effluent was analyzed with a HP 6890 Plus gas chromatograph equipped with a washed molesieve 13X column and a Porapak Q column. This allowed oxygen, nitrogen, nitrous oxide, carbon monoxide, carbon dioxide, water, methane, and propene to be separated and quantified. No nitrous oxide, methane, or any hydrocarbons other than propene were detected in catalytic studies in the presence of O2 . All carbons in the molar carbon balance could be accounted for with just CO, CO2 , and C 3 H6 . Prior to each NOx temperature-programmed desorption (TPD) experiment, 0.2 g of catalyst sample was purged in a flow of 15% O2 in helium at 200 cm3 min−1 at 600 ◦ C for 2 h. After cooling to 50 ◦ C, NOx adsorption was performed by exposing the sample to 0.1% NO and 15% O2 in helium at 200 cm3 min−1 for 2 h. Desorption of adsorbed species was performed at a heating rate of 2 ◦ C min−1 in a helium flow of 60 cm3 min−1 with 0.5% Ar as internal standard. The effluent was analyzed using a HP 6890 Plus gas chromatograph equipped with a HP 5972A Mass Selective Detector. Catalyst surface area and pore size distribution were calculated from the nitrogen adsorption isotherm (Micromeritics ASAP 2000) using the BET and BJH methods, respectively. Phase identification was achieved by X-ray diffraction (XRD) with a Siemens D5000 diffractometer. Scherrer’s analysis of X-ray peak broadening was employed to determine crystallite size. Particle morphology was studied using transmission electron microscopy (TEM) performed on a JEOL 2010 highresolution microscope. Elemental mapping was collected using a VG HB603 scanning transmission electron microscope (STEM) equipped with energy dispersive X-ray (EDX) analysis.

Results and discussion To study the effect of synthesis procedure, Ga2 O3 /Al2 O3 catalysts with 10 wt% Ga were prepared using the impregnation method (I), the sol—gel method (S), and the sequential precipitation method (P). The NO reduction activities of these Ga2 O3 /Al2 O3 catalysts in the absence of water are shown in Fig. 1. Our Ga2 O3 /Al2 O3 (P) catalyst displayed the highest N2 yield of ∼80% at 450 ◦ C, which was better than the N2 yield of ∼70% from Ga2 O3 /Al2 O3 (S) catalyst. The N2 yields of these two catalysts were much higher than that of the Ga2 O3 /Al2 O3 (I) catalyst due to their higher propene conversions, which were attributed to higher Ga dispersions in Ga2 O3 /Al2 O3 (P) and Ga2 O3 /Al2 O3 (S) catalysts. Unlike Ga2 O3 /Al2 O3 (I) catalyst, Ga2 O3 /Al2 O3 (P) and Ga2 O3 /Al2 O3 (S) showed no distinct XRD peaks associated with the Ga2 O3 phase (Supplementary

222

P. Pitukmanorom, J.Y. Ying

Figure 1 N2 yield and propene conversion for NO reduction by propene over the Ga2 O3 /Al2 O3 catalysts with 10% Ga synthesized by () impregnation; () sol—gel; and () sequential precipitation. The reactions were run at a contact time of 0.06 g s cm−3 (∼30,000 h−1 ) with 0.1% NO, 0.1% C3 H6 , 15% O2 , and 0% H2 O. Table 1 loading.

BET surface area, pore volume and average pore diameter of Al2 O3 and Ga2 O3 /Al2 O3 catalysts with 10 wt% of Ga

Catalyst

Method

Surface area (m2 g−1 )

Pore volumea (cm3 g−1 )

Mean pore diametera (nm)

Al2 O3 Ga2 O3 /Al2 O3 (I) Ga2 O3 /Al2 O3 (S) Ga2 O3 /Al2 O3 (P)

— Impregnation Sol—gel Sequential Precipitation

200 153 199 210

0.49 0.39 0.65 0.61

8.2 10.0 11.4 10.0

a

BJH method from the adsorption isotherm.

Data Fig. S1), which indicated that the Ga species were well dispersed over the latter two catalysts. Table 1 also shows that the Ga2 O3 /Al2 O3 (P) catalyst possessed the highest surface area, which indicated that our sequential precipitation approach could be used to prepare catalysts with high Ga2 O3 dispersion and excellent surface area. Of all the metal oxide-coated catalysts that we examined, Ga2 O3 /Al2 O3 with 10 wt% Ga exhibited the best N2 yield of ∼80% in the absence of water at 450 ◦ C. However, the N2 yield of this catalyst was reduced to ∼60% when 10% water was introduced into the feed stream due to a decrease in propene conversion from 98% to 57%. Such inhibition effect could be attributed to the competitive adsorption between water and one or more of the reactants [4] as the presence of water caused a significant decrease in propene conversion along with an increase in selectivity (Supplementary Data Fig. S2), suggesting that the amount of propene on the surface was reduced. In the presence of 10% water, pure Al2 O3 and Ga2 O3 /Al2 O3 have higher selectivities for NO reduction by propene than In2 O3 /Al2 O3 (Table 2). The introduction of Ga2 O3 improved the N2 yield of Al2 O3 by enhancing the propene conversion, without compromising the propene selectivity. In contrast, In2 O3 /Al2 O3 displayed good N2 yield due to high propene conversion despite its relatively low selectivity. Recent studies have proposed that the role of SnO2 /Al2 O3 and In2 O3 /Al2 O3 in the SCR of NO was to selectively oxidize propene to acrolein, which further reacted on Al2 O3 to form nitrogen [21,23,24]. Taking advantage of the good propene selectivity offered by Ga2 O3 /Al2 O3 and the high propene conversions provided by SnO2 /Al2 O3 and In2 O3 /Al2 O3 , alumina-based nanocompos-

ites containing both Ga and either Sn or In oxides were synthesized in various compositions. The total loadings of Ga and either Sn or In were kept constant at 10 wt%. The catalytic activities of the alumina-based nanocomposites containing 10 wt% Ga, 4 wt% Sn and 6 wt% Ga, and 4 wt% In and 6 wt% Ga illustrated that the introduction of Sn or In in addition to Ga led to improvements in propene conversion, giving rise to greater N2 yield (Fig. 2). In particular, a combination of 4 wt% In and 6 wt% Ga led to an excellent yield of 80% at 450 ◦ C. The metal oxide/Al2 O3 nanocomposites containing 10 wt% Ga, 4 wt% Sn and 6 wt% Ga, and 4 wt% In and 6 wt% Ga gave rise to selectivities of 11.8%, 10.7%, and 9.5% at 450 ◦ C, respectively. Thus, the nanocomposites approach provided for excellent synergism between the active components, so that substantial increases in propene conversions were realized without significant reduction in selectivities for the latter two catalysts.

Table 2 Catalytic activities of pure ␥-Al2 O3 and metal oxide-coated Al2 O3 catalysts with 10 wt% of metal loading.a . Catalyst

C 3 H6 conversion (%)

Selectivity (%)

N2 yield (%)

␥-Al2 O3 Ga2 O3 /Al2 O3 In2 O3 /Al2 O3

51 57 83

10.8 11.8 7.8

49 60 59

a The reactions were run at a contact time of 0.06 g s cm−3 (∼30,000 h−1 ) at 450 ◦ C with 0.1% of NO, 0.1% of C3 H6 , 15% of O2 , and 10% of H2 O.

Selective catalytic reduction of nitric oxide by propene over In2O3—Ga2O3/Al2O3 nanocomposites

223

Figure 2 N2 yield and propene conversion for NO reduction by propene over metal oxide-coated Al2 O3 nanocomposites containing () 10 wt% Ga; () 4 wt% Sn and 6 wt% Ga; and () 4 wt% In and 6 wt% Ga. The reactions were run at a contact time of 0.06 g s cm−3 (∼30,000 h−1 ) with 0.1% NO, 0.1% C3 H6 , 15% O2 , and 10% H2 O.

Figure 3 Effect of In loading on () N2 yield; () propene conversion; and () propene selectivity for NO reduction by propene at 450 ◦ C over In2 O3 —Ga2 O3 /Al2 O3 nanocomposites with a total Ga and In loading of 10 wt%. The reactions were run at a contact time of 0.06 g s cm−3 (∼30,000 h−1 ) with 0.1% NO, 0.1% C3 H6 , 15% O2 , and 10% H2 O.

To optimize the catalytic activities of In2 O3 —Ga2 O3 /Al2 O3 nanocomposites, various combinations of Ga and In were examined, while keeping the total loading of Ga and In constant at 10 wt%. In all formulations, no Ga2 O3 XRD peaks were observed, while In2 O3 peaks were only visible at an In loading of 8 wt% or higher, indicating that these oxides were highly dispersed on the support (Supplementary Data Fig. S3). In2 O3 did not react with Al2 O3 , and In loading led to increase in catalyst surface area due to suppression of grain growth and sintering. The nanocomposites containing 10 wt% Ga and 10 wt% In have BET surface areas of 210 m2 g−1 and 298 m2 g−1 , respectively (Table 3). In2 O3 —Ga2 O3 /Al2 O3 nanocomposites allowed for significantly greater propene conversion than Ga2 O3 /Al2 O3 , with Table 3

some compromise in propene selectivity (Fig. 3). The introduction of only 2 wt% In led to a dramatic increase in the propene conversion and complete propene conversion was achieved at 500 ◦ C for the nanocomposite containing 2 wt% In and 8 wt% Ga. Selectivity was found to gradually decrease with increasing In loading (Fig. 3). Thus, the best N2 yields were obtained with the In2 O3 —Ga2 O3 /Al2 O3 catalysts with 2—4 wt% In and 8—6 wt% Ga. At 450 ◦ C, these catalysts were able to convert 80% of NO to N2 even with high partial pressures of oxygen and water. We believe that this remarkable lean-NOx activity arose in part from the high dispersion of Ga2 O3 and In2 O3 . The synergism between these oxide components could have allowed the activated propene species created on In2 O3 to be utilized in the reduction of NO to

BET surface area, pore volume and average pore diameter of In2 O3 —Ga2 O3 /Al2 O3 nanocomposites.

Ga loading (wt%)

In loading (wt%)

Surface area (m2 g−1 )

Pore volumea (cm3 g−1 )

Mean pore diametera (nm)

10 8 6 4 2 0

0 2 4 6 8 10

210 223 243 285 312 298

0.61 0.82 0.96 1.02 1.19 1.13

10.0 12.6 13.6 13.1 14.4 15.6

a

BJH method from the adsorption isotherm.

224

P. Pitukmanorom, J.Y. Ying Table 4 Amount of NO desorbed at 200—600 ◦ C from In2 O3 —Ga2 O3 /Al2 O3 nanocomposites. Ga loading

In loading

NO desorbed

(wt%)

(wt%)

(␮mol NO g−1 )

(␮mol NO m−2 )

0 0 2 4 6 8 10

130 190 230 250 290 285 220

0.67 0.92 1.02 1.03 1.08 1.00 0.86

0 10 8 6 4 2 0

Figure 4 TPD profiles of NO and O2 for In2 O3 —Ga2 O3 /Al2 O3 nanocomposite with 2 wt% In and 8 wt% Ga. 0.2 g of the sample was pretreated in a flow of 15% O2 in helium at 600 ◦ C for 2 h, followed by NO adsorption in 0.1% NO and 15% O2 at 50 ◦ C for 2 h. Desorption was performed at a heating rate of 2 ◦ C min−1 in a helium flow of 60 cm3 min−1 .

N2 by the more selective Ga2 O3 /Al2 O3 , rather than being consumed in the combustion process. Although the activity of the In2 O3 —Ga2 O3 /Al2 O3 nanocomposite was limited by inadequate hydrocarbon conversion at low temperatures, the SCR activity at higher temperatures where good NOx reduction generally occurred was hindered by low selectivity. This decrease in selectivity coincided with the temperature at which surface NOx species decomposed during TPD experiment (Fig. 4), suggesting that these surface nitrates played an important role in the SCR process. The temperature at which the catalyst is active is also expected to depend on the nature and concentration of hydrocarbons. The presence of a wide variety of hydrocarbons in the exhaust could result in a wider operating-temperature window for the In2 O3 —Ga2 O3 /Al2 O3 catalyst. Studies on the formation and reaction of surface nitrates by Haneda et al. [25], Meunier et al. [18], and Shimizu et al. [17] showed that adsorbed NOx species over aluminabased catalysts were highly reactive towards hydrocarbons.

Upon exposure to hydrocarbons, these species would react to form N2 at rates similar to those found in steady-state NO reduction. To quantify the amount of NOx adsorbed on the surface of our nanocomposites, TPD experiments were performed to obtain the NO and O2 desorption profiles (Fig. 4). Two desorption peaks were observed in all nanocomposites examined—–NO desorption at 120 ◦ C, and a simultaneous desorption of NO and O2 at ∼400 ◦ C due to nitrate decomposition. The presence of either Ga or In increased the amount of NO desorbed over the high-temperature region from the alumina-based catalyst, but the most significant increases in NO desorbed was associated with the nanocomposites that contained both Ga and In (Table 4). The amount of NOx adsorbed increased from 130 ␮mol g−1 (0.67 ␮mol m−2 ) for Al2 O3 to 190 ␮mol g−1 (0.92 ␮mol m−2 ) for Ga2 O3 /Al2 O3 to 230 ␮mol g−1 (1.02 ␮mol m−2 ) for In2 O3 —Ga2 O3 /Al2 O3 with 2 wt% In and 8 wt% Ga. These findings indicated that the number of NOx adsorption sites on the catalysts was increased with the addition of Ga or In. The addition of In species might have improved the dispersion of Ga species, besides increasing the BET surface area of the catalyst (Table 3). The resistance of the In2 O3 —Ga2 O3 /Al2 O3 nanocomposite against SO2 poisoning was examined at 450 ◦ C in the presence of 100 ppm SO2 . Exposure to SO2 led to a reduction in the N2 yield from ∼80% to 60% over the first 10 h, at which point, the N2 yield appeared to stabilize at 60% (Fig. 5). The reduction in N2 yield could be attributed to decreases in

Figure 5 Effect of SO2 on the () N2 yield; () propene conversion; and () selectivity for NO reduction by propene at 450 ◦ C over the In2 O3 —Ga2 O3 /Al2 O3 nanocomposite with 2 wt% In and 8 wt% Ga. The reactions were run at a contact time of 0.06 g s cm−3 (∼30,000 h−1 ) with 0.1% NO, 0.1% C3 H6 , 15% O2 , and 10% H2 O. SO2 (100 ppm) was introduced in the feed stream for the first 24 h.

Selective catalytic reduction of nitric oxide by propene over In2O3—Ga2O3/Al2O3 nanocomposites both propene conversion and selectivity as a result of SO2 adsorption and sulfate formation on the catalyst surface. The presence of sulfate species has been shown to reduce nitrate formation over Ga2 O3 —Al2 O3 catalyst due to competitive adsorption between the two species on the basic oxygen sites [26]. Thus, the quantity of adsorbed NOx species over our nanocomposite catalyst was expected to decrease when SO2 was added to the feed stream. This poisoning of NOx adsorption sites [26,27] could cause the selectivity to decrease. However, upon the removal of SO2 from the feed stream after 24 h of exposure, 85% of the original N2 yield was recovered within 5 h, and ∼90% of the original N2 yield was obtained after 24 h. This was due to the high degree of recovery in propene conversion, despite some loss in catalyst selectivity. The resistance of the In2 O3 —Ga2 O3 /Al2 O3 nanocomposite against poisoning by SO2 would be of major importance since most fuels would have some sulfur content.

Conclusions Nanocrystalline alumina-supported catalysts active for the SCR of NO by propene were synthesized by a controlled chemical precipitation technique. The Ga2 O3 /Al2 O3 catalyst prepared by this method exhibited excellent NO reduction activity due to its high surface area and good dispersion of Ga species. However, the activity of this catalyst was lowered in the presence of 10% H2 O due to reduced propene conversion. By introducing a combination of metal oxide coatings of Ga2 O3 , SnO2 and In2 O3 on alumina, excellent catalytic activity could be achieved in the presence of water vapor despite some decrease in propene selectivity. N2 yields as high as ∼80% were achieved over In2 O3 —Ga2 O3 /Al2 O3 nanocomposites at 450 ◦ C. The significantly improved activity was attributed to synergism between better propene activation by In and improved NOx adsorption. The In2 O3 —Ga2 O3 /Al2 O3 nanocomposites remained active even in the presence SO2 , where only a temporary reduction in N2 yield was observed.

Conflict of interest To avoid a conflict of interest, Younan Xia acted as Editor for this paper.

Acknowledgments The authors thank Michael Frongillo and Anthony GarrattReed of the MIT CMSE for their assistance with electron microscopy. This work was sponsored by Singapore-MIT Alliance.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.nantod. 2009.03.001.

References [1] B.I. Bertelsen, Top. Catal. 16/17 (2001) 15.

225

[2] M. Amiridis, T. Zhang, R.J. Farrauto, Appl. Catal. B 10 (1996) 203. [3] A. Fritz, V. Pitchon, Appl. Catal. B 13 (1997) 1. [4] R. Burch, J.P. Breen, F.C. Meunier, Appl. Catal. B 39 (2002) 283. [5] H.H. Kung, M.C. Kung, Catal. Today 30 (1996) 5. [6] M. Iwamoto, H. Yahiro, Y. Yu-u, S. Shundo, N. Mizuno, Shokubai 32 (1990) 430. [7] W. Held, A. Koenig, T. Richter, L. Puppe, SAE Paper (1990) 900496. [8] P.W. Park, H.H. Kung, D.-W. Kim, M.C. Kung, J. Catal. 184 (1999) 440. [9] J. Li, J. Hao, L. Fu, Z. Liu, X. Cui, Catal. Today 90 (2004) 215. [10] J.Y. Yan, M.C. Kung, W.M.H. Sachtler, H.H. Kung, J. Catal. 172 (1997) 178. [11] N. Bogdanchikova, F.C. Meunier, M. Avalos-Borja, J.P. Breen, A. Pestryakov, Appl. Catal. B 36 (2002) 287. [12] K. Shimizu, J. Shibata, H. Yoshida, A. Satsuma, T. Hattori, Appl. Catal. B 30 (2001) 151. [13] K. Shimizu, M. Takamatsu, K. Nishi, H. Yoshida, A. Satsuma, T. Tanaka, S. Yoshida, T. Hattori, J. Phys. Chem. B 103 (1999) 1542. [14] M. Haneda, Y. Kintaichi, H. Shimada, H. Hamada, J. Catal. 192 (2000) 137. [15] J. Li, J. Hao, X. Cui, L. Fu, Catal. Lett. 103 (2005) 75. [16] J.A. Perdigón-Melon, A. Gervasini, A. Auroux, J. Catal. 234 (2005) 421. [17] K. Shimizu, H. Kawabata, A. Satsuma, T. Hattori, J. Phys. Chem. B 103 (1999) 5240. [18] F.C. Meunier, J.P. Breen, V. Zuzaniuk, M. Olsson, J.R.H. Ross, J. Catal. 187 (1999) 493. [19] Y. Chi, S.S.C. Chuang, J. Catal. 190 (2000) 75. [20] T. Maunula, J. Ahola, H. Hamada, Appl. Catal. B 26 (2000) 173. [21] P.W. Park, S.C. Ragle, C.L. Boyer, M.L. Balmer, M. Engelhard, D. McCready, J. Catal. 210 (2002) 97. [22] K. Maeda, F. Mizukami, S. Niwa, M. Toba, M. Watanabe, K. Masuda, J. Chem. Soc. Faraday Trans. 88 (1992) 97. [23] A. Yezerets, Y. Zhang, P.W. Park, M.C. Kung, H.H. Kung, Stud. Surf. Sci. Catal. 130 (2000) 629. [24] J.-H. Lee, A. Yezerets, M.C. Kung, H.H. Kung, Chem. Commun. (2001) 1404. [25] M. Haneda, Y. Kintaichi, T. Mizushima, N. Kakuta, H. Hamada, Appl. Catal. B 31 (2001) 81. [26] M. Haneda, Y. Kintaichi, H. Hamada, Appl. Catal. B 31 (2001) 251. [27] T.N. Angelidis, S. Christoforou, A. Bongiovanni, N. Kruse, Appl. Catal. B 39 (2002) 197. Dr. Pemakorn Pitukmanorom received his B.A. (1999) and Ph.D. (2004) in Chemical Engineering from University of Cambridge and Massachusetts Institute of Technology, respectively. He joined the Institute of Bioengineering and Nanotechnology, Singapore in 2004 as Research Scientist where he worked on polymeric drug delivery systems for applications in tissue engineering. Prof. Jackie Y. Ying received her B.E. and Ph.D. from The Cooper Union and Princeton University, respectively. She joined the faculty at MIT in 1992, where she was Professor of Chemical Engineering until 2005. She has been the Executive Director of the Institute of Bioengineering and Nanotechnology in Singapore since 2003. For her research in nanostructured materials, Prof. Ying has been recognized with the American Ceramic Society Ross C. Purdy Award, David and Lucile

226 Packard Fellowship, Office of Naval Research and National Science Foundation Young Investigator Awards, Camille Dreyfus TeacherScholar Award, American Chemical Society Faculty Fellowship Award in Solid-State Chemistry, Technology Review TR100 Young

P. Pitukmanorom, J.Y. Ying Innovator Award, and American Institute of Chemical Engineers Allan P. Colburn Award. She was elected a World Economic Forum Young Global Leader, and a member of the German National Academy of Sciences, Leopoldina.