A review of selective catalytic reduction of nitrogen oxides with hydrogen and carbon monoxide

A review of selective catalytic reduction of nitrogen oxides with hydrogen and carbon monoxide

Applied Catalysis A: General 421–422 (2012) 1–13 Contents lists available at SciVerse ScienceDirect Applied Catalysis A: General journal homepage: w...

1MB Sizes 10 Downloads 52 Views

Applied Catalysis A: General 421–422 (2012) 1–13

Contents lists available at SciVerse ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

Review

A review of selective catalytic reduction of nitrogen oxides with hydrogen and carbon monoxide Hideaki Hamada a,∗ , Masaaki Haneda b a Research Center for New Fuels and Vehicle Technology, National Institute of Advanced Industrial Science and Technology (AIST), AIST Tsukuba Central 5, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan b Ceramics Research Laboratory, Nagoya Institute of Technology, 10-6-29 Asahigaoka, Tajimi, Gifu 507-0071, Japan

a r t i c l e

i n f o

Article history: Received 7 December 2011 Received in revised form 28 January 2012 Accepted 3 February 2012 Available online 10 February 2012 Keywords: Nitrogen oxide Selective catalytic reduction Hydrogen Carbon monoxide Platinum Palladium Iridium Rhodium

a b s t r a c t The selective reduction of NO with hydrogen (H2 -SCR) and CO (CO-SCR) over platinum group metal catalysts in the presence of O2 is overviewed. In the case of H2 -SCR, Pt and Pd show high activity at low temperatures. The acidity of the support material greatly affects NO reduction activity and selectivity to N2 /N2 O. Although the activity of Ir and Rh for H2 -SCR is low, coexisting SO2 in the reaction gas considerably promotes NO reduction. The best support for Ir and Rh is SiO2 . Li and Zn additives improve the activity of Ir/SiO2 and Rh/SiO2 , respectively, by maintaining the active reduced metal state. For CO-SCR, on the other hand, Ir is almost the only active metal species. Coexisting SO2 is also essential for CO-SCR on Ir/SiO2 to occur. The role of SO2 for both H2 -SCR and CO-SCR on Ir/SiO2 is to keep Ir in the form of the catalytically active Ir metal state. The additions of WO3 and Nb2 O5 considerably promote the activity of Ir/SiO2 for CO-SCR, catalyzing CO-SCR even in the absence of SO2 . Ir metal interacting strongly with W oxide is the active species on WO3 -promoted Ir/SiO2 . Furthermore, the addition of Ba improves the performance of Ir/WO3 /SiO2 catalyst. © 2012 Elsevier B.V. All rights reserved.

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selective catalytic reduction with hydrogen (H2 -SCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Pt catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. Catalytic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. Effect of supports and additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Pd catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Ir and Rh catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selective catalytic reduction with carbon monoxide (CO-SCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Catalytic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Ir as an active species for CO-SCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Catalytically active state of Ir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Additive effect for Ir catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Influence of coexisting gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.1. Effect of SO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.2. Effect of H2 O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author. Tel.: +81 29 861 9329, fax: +81 29 861 4441. E-mail address: [email protected] (H. Hamada). 0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2012.02.005

2 2 2 2 2 4 5 7 7 8 9 9 11 11 11 12 13 13

2

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

1. Introduction Catalytic reduction of NO in vehicle exhaust plays an important role in reducing NOx which is a major atmospheric pollutant. Although the three-way catalyst for stoichiometric gasoline-fueled vehicles has been commercialized worldwide, this material cannot be applied to oxygen-rich exhaust emissions such as those from diesel and lean burn engines. The selective catalytic reduction of NO with NH3 (NH3 -SCR), which is also in practical use for NOx removal emitted from large-scale boilers, can be applied to oxygenrich exhaust. However, the use of NH3 as a reductant is difficult for vehicle applications. In this regard, a great number of studies have been conducted on the use of hydrocarbons as a reductant for the selective catalytic reduction of NO (HC-SCR) in oxygen-rich atmospheres [1–3], since the first reports by Iwamoto et al. [4] and Held et al. [5] following the patents of Volkswagen [6] and Toyota [7]. Many kinds of catalysts based on zeolite [8], metal oxide [9] and noble metals [10] have been reported to show activity for HC-SCR. Nevertheless, practical application of HC-SCR has been difficult due to its insufficient performance in real life applications. It has been long believed that H2 and CO cannot reduce NO selectively in oxygen-rich conditions because they are used as reductants in the stoichiometric three-way catalyst system as well as hydrocarbons. However, it has been proved recently that H2 and CO can also act as effective reductants for SCR. Since the selective catalytic reduction of NO with H2 (H2 -SCR) proceeds at relatively low temperatures, H2 -SCR is an attractive approach to the efficient removal of NOx in the exhaust of lean-burn and diesel engines, the temperature of which has become lower and lower due to the improvement of engine thermal efficiency. H2 can be formed by reforming of hydrocarbon fuels. H2 -SCR is also a promising measure for NOx treatment emitted from hydrogen fueled vehicles. The selective catalytic reduction of NO with CO (CO-SCR), on the other hand, is more attractive from a practical point of view, since CO is generally contained in vehicle exhausts and can be relatively more easily produced by engine operation compared with H2 . Moreover, the new type of internal combustion engines recently developed, such as homogeneous charge compression ignition (HCCI) engines, emit relatively high concentrations of CO [11], which can be used as a reductant for NO. This review article highlights the recent progress in H2 -SCR and CO-SCR. The first part overviews the studies on the H2 -SCR reaction using supported platinum group metal catalysts. Since Pt and Pd were initially reported to show high activity for H2 -SCR, the effects of supports and additives on the catalytic performance of Pt and Pd are extensively described. Although the activity of Ir and Rh for H2 -SCR is low, it has been reported by the authors’ group that the coexistence of SO2 promotes their H2 -SCR activity. Therefore, the special catalytic activity of Ir and Rh is described. The second part of the present article reviews the CO-SCR reaction, for which Ir is practically the only catalytically active metal. Since the activity of Ir is greatly affected by supports and additives, the catalytically active species of supported Ir is discussed in detail mainly based on our studies. The effect of coexisting gases, which is also an essential factor in determining the catalytic performance, is also described. 2. Selective catalytic reduction with hydrogen (H2 -SCR) H2 -SCR did not attract much attention before because of the superior efficiency of NH3 and hydrocarbon reductant for SCR of NO in oxygen-rich atmospheres. However, after it was confirmed that H2 can also act as an effective reductant for SCR of NO in lean conditions, many reports have been published on H2 -SCR, most of which use platinum group metals as the active

component. The performances of catalysts for H2 -SCR are summarized in Table 1. 2.1. Pt catalysts 2.1.1. Catalytic activity In 1971, Jones et al. [12] pointed out that H2 can react preferentially with NO over O2 in the reaction system of H2 –NO–O2 on commercially supported Pt catalysts, suggesting the possibility of H2 -SCR, although the reaction was not performed in excess oxygen conditions. Lamb and Tollefson [13] also reported the catalytic reduction of NO with H2 in the presence of O2 in low concentration and high velocity gas streams. Probably the first report to confirm H2 -SCR in net-oxidizing conditions is the one by Fu and Chuang [14]. They reported that stable NOx conversions of 60–80% were obtained at temperatures above 55 ◦ C and at space velocities from 3000 to 10,000 h−l for a feed containing 1000 ppm NOx, 1% H2 , and 3.2% O2 in N2 by using noble metal catalysts supported on styrene-di-vinylbenzene copolymer (SDB). The catalytic activity of the noble metals was found to be in the order of Pt > Pd–Ru > Pd > Ru  Au. In this catalytic reaction, low reaction temperature was essential to the selective NO–H2 reaction due to the competitive reaction between H2 and O2 , indicating the presence of a temperature window for H2 -SCR. It was also reported that introduction of H2 O did not affect the reaction because of the hydrophobic support material. In 1994, Wildermann [15] showed that a Pt–Mo catalyst supported on Al2 O3 is active for the selective NO–H2 reaction in the presence of 8% O2 . It was confirmed that Pt is the most active for H2 -SCR among noble metals, although undesirable N2 O formation is considerable in addition to N2 as a reduction product of NO. In 1997, Yokota et al. [16] investigated the catalytic performance of Pt-group metals for H2 -SCR and reported that Pt shows the highest NO reduction activity among platinum group metals when supported on Al2 O3 . For Pt, the effectiveness of catalyst support was in the order of ZSM-5 ∼ mordenite > SiO2 > Al2 O3 . It was noted that NO reduction over Pt/mordenite took place between 100 and 200 ◦ C and that NO conversion increased with increasing H2 concentration and NO concentration, although coexisting CO inhibited NO reduction. They also found the high catalytic performance of Pt–Mo–Na/SiO2 catalyst for H2 -SCR with a wide temperature window. The effect of Na and Mo addition was later studied by Burch and Coleman [17]. Frank et al. [18] investigated the kinetics and mechanism of H2 -SCR with Pt–Mo–Co/Al2 O3 catalyst, and established the kinetics of the NO/H2 /O2 reaction using a modified Langmuir-Hinshelwood model. An interesting feature of the reaction is the promoting effect of O2 on NO reduction under low partial pressures of O2 . Burch and Coleman [19] systematically examined Pt-group metals (Pt, Pd, Rh, and Ir) on Al2 O3 and SiO2 for catalytic activity for NO reduction in the presence of a large excess of O2 (500 ppm NO, 2000 ppm H2 , 6% O2 ). Although Pd, Rh and Ir catalysts were all found to be inactive towards NO reduction, NO was reduced effectively over Pt/Al2 O3 catalyst (50% conversion at 140 ◦ C) and over 1% Pt/SiO2 (75% conversion at 90 ◦ C), as shown in Fig. 1. It was noted that N2 O is also formed in large amounts in addition to N2 by H2 -SCR over Pt catalysts. The selectivity to N2 O was sensitive to reaction conditions; N2 O formation is increased in the presence of H2 O and decreased with increasing temperature. In general, increasing the NO concentration lowered the activity, whereas increasing the H2 concentration increased the activity for NO reduction. 2.1.2. Effect of supports and additives Following the above-mentioned early studies, a large number of reports have been published concerning H2 -SCR on Pt catalysts.

Table 1 Activity of various catalysts for H2 -SCR. Catalyst

NO (ppm)

a

2,760 1,000 2,000 200–2000 1,000 1,000 500 500 800 800 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 2,500 2,500 2,500 1,000 1,000 500 500 2,500 1,000 1,000 1,000 1,000 1,000 1,000

H2 (%)

O2 (%)

0.90 1.0 0.6 0–0.6 0.4 0.4 0.2 0.2 0.3 0.3 0.5 0.5 0.5 1.0 0.4 0.5 0.5 0.5 0.4 1.0 1.0 1.0 0.3 0.8 0.3 0.4 1.0 1.0 0.5 0.3 0.3 0.6 0.6

0.91 3.20 6.00 6.00 6.00 6.00 6.00 6.00 10.00 10.00 6.70 6.70 6.70 2.00 10.00 6.70 6.70 6.70 10.00 5.00 5.00 5.00 5.00 2.00 10.00 10.00 5.00 5.00 5.00 0.65 0.65 5.00 5.00

Other gases

10% CO2 10% CO2

H2 O (%)

5 5

5 5 10 1000 ppm CO

20 ppm SO2 20 ppm SO2 20 ppm SO2 20 ppm SO2

10 10 6 6

W/F (g s cm−3 ) or GHSV (h−1 ) 1100 h−1 3000 h−1 n.a. n.a. 0.03 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.24 g s cm−3 0.24 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.12 g s cm−3 0.12 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.3 g s cm−3 80,000 h−1 80,000 h−1 80,000 h−1 0.18 g s cm−3 0.02 g s cm−3 0.03 g s cm−3 0.03 g s cm−3 0.09 g s cm−3 20,000 h−1 0.06 g s cm−3 0.0267 g s cm−3 0.0267 g s cm−3 0.0267 g s cm−3 0.0267 g s cm−3

166 45a 120 190 120 120 140 90 80 90 75 75 75 127 140 100 120 120 100 140 160 150 100 100 140 150 150 130 164 300 300 250 300

NO reduction efficiency

Ref.

NO conversion (%)

Selectivity

91 90 80 36 75 90 50 75 92 90 44 58 37 100 90 60 80 81 80 82 90 94 51 70 76 98 100 75 41 73 53 65 55

n.a. n.a. n.a. N2 /N2 O = 4 30%N2 , 70%N2 O 30%N2 , 70%N2 O 50%N2 , 50%N2 O 25%N2 , 75%N2 O 40%N2 , 60%N2 O 55%N2 , 45%N2 O 8%N2 , 92%N2 O 19%N2 , 81%N2 O 45%N2 , 55%N2 O 78%N2 , 22%N2 O 45%N2 , 18%NH3 , 37%N2 O 60%N2 , 40%N2 O 85%N2 , 15%N2 O 69%N2 , 31%N2 O 40%N2 , 60%N2 O 87%N2 , 13%N2 O 92%N2 , 8%N2 O 80%N2 , 20%N2 O 21%N2 , 79%N2 O 100%N2 90%N2 , 10%N2 O 92%N2 , 8%N2 O 78%N2 , 22%N2 O n.a. 76%N2 , 24%N2 O 61%N2 , 39%N2 O 91%N2 , 9%N2 O N2 /N2 O = 3/1 N2 /N2 O = 6/1

[12] [14] [16] [16] [17] [17] [19] [19] [20] [20] [22] [22] [22] [23] [25] [26] [26] [27] [28] [29] [30] [32] [36] [37] [38] [39] [40] [41] [42] [44] [44] [46] [47]

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Pt (PZ-1-168) 2% Pt/SDB 2% Pt/mordenite 2% Pt–Mo–Na/SiO2 1% Pt/Al2 O3 1% Pt/0.27Na/Al2 O3 1% Pt/Al2 O3 1% Pt/SiO2 1% Pt/Al2 O3 1% Pt/TiO2 –ZrO2 1% Pt/SiO2 1% Pt/SiO2 –Al2 O3 1% Pt/MFI 3% Pt/ZrO2 + H-ZSM-5 (1:3) 1% Pt/(Sn0.9 Fe0.1 )P2 O7 1% Pt/Si-MCM-41 1% Pt/Al-MCM-41 1% Pt/ZSM-5 1% Pt/CeO2 0.1% Pt/La0.5 Ce0.5 MnO3 0.1% Pt/La0.7 Sr0.2 Ce0.1 FeO3 0.1% Pt/Ce–Mg–O 1% Pd/TiO2 Pd/MFI 0.5% Pd/Ti-Pillared Clay 1% Pd–5% V2 O5 /20% TiO2 –Al2 O3 0.5% Pd/LaCoO3 La0.8 Sr0.2 Fe0.9 Pd0.1 O3 1% Pd/K2 O–6TiO2 0.5% Ir/SiO2 0.5% Rh/SiO2 5% Ir–Li/SiO2 5% Rh–Zn/SiO2

Tmax (◦ C)

Reaction conditions

The temperature may be higher due to reaction heat.

3

4

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Fig. 1. Catalytic activity of 1% Pt/Al2 O3 () and 1% Pt/SiO2 () for H2 -SCR [19]. NO = 500 ppm, H2 = 2000 ppm, O2 = 6%, W/F = 0.03 g s cm−3 .

These studies have shown that NO conversion activity and N2 /N2 O selectivity are strongly dependent on the support for Pt catalyst. Machida et al. [20] investigated the performance of Pt catalysts on various metal oxides for steady-state catalytic reactions of a gas mixture of 0.08% NO, 0.28% H2 , and 10% O2 . Their results indicate that the NO conversion to N2 /N2 O is strongly affected by the oxide supports. Al2 O3 and SiO2 exhibited high NO conversion below 100 ◦ C with the selectivity over 60% to N2 O, consistent with the results previously reported by Burch and Coleman [19]. The low activity of Pt on CeO2 -based oxides suggests that the basicity of support has a negative effect on the activity. By contrast, various binary oxides with solid acidity generally show good catalytic performance, among which TiO2 –ZrO2 exhibits higher activity with more formation of N2 than the other catalysts. They also studied the cause of the high NO–H2 reaction selectivity versus NO–O2 reaction, and concluded that nitrate formed by oxidative adsorption of NO covering the Pt surface inhibits H2 –O2 reaction. This study indicates the importance of acid-base properties of the supports. Satsuma et al. [21] studied the catalytic activity of supported Pt catalysts; they point out that the presence of strong acid sites on a metal oxide support results in a reaction of adsorbed NH4 + species in a flow of NO + O2 , leading not to N2 O but to N2 formation. They also [22] investigated the essential factors in determining the activity and selectivity of Pt catalysts supported on various zeolites and metal oxides; they found that high NO conversion was obtained over Pt/SiO2 –Al2 O3 and Pt/SiO2 , while N2 selectivity was high on Pt/zeolites, especially Pt/MFI. They concluded that less oxidized Pt shows higher turnover frequency for H2 -SCR, and that the N2 selectivity is related to the acid strength of the support, on which the concentration of surface NH4 + is higher, leading to N2 formation. In addition to NO reduction pathways to N2 /N2 O over Pt surfaces, a bifunctional mechanism was proposed, which includes NH3 formation via NO reduction by H2 on Pt surface, followed by storage of NH4 + on the Brønsted acid site of acidic supports, and the selective N2 formation by the well-established NH3 -SCR mechanism.

The importance of NH3 intermediates in H2 -SCR was also indicated by Nanba et al. [23], who reported that the formation of NH3 as a byproduct of H2 -SCR was substantial over Pt/ZrO2 . The NH3 thus formed was effectively converted to N2 by the selective reduction of NO with NH3 over H-ZSM-5 physically mixed with the Pt catalyst. Detailed investigation of the catalytic performance of Pt/ZrO2 revealed that the catalysts with high Pt dispersion showed high NH3 selectivity and those with agglomerated Pt particles exhibited higher N2 formation [24], and that reduced Pt sites promoted NH3 formation. In this reaction, decomposition of the ammonium nitrate that had accumulated on the catalyst surface seemed to be involved in the NO–H2 –O2 reaction. Recently, several reports have appeared concerning Pt catalysts on various acidic supports for H2 -SCR. The formation of ammonia and NOx reduction with high N2 selectivity was reported over a Pt catalyst supported on Fe-doped SnP2 O7 powder with strong solid acidity [25]. The good catalytic performances of Pt/Al-MCM41 [26] and Pt/ZSM-35 [27] were also reported. In the case of Pt/MCM-41, the introduction of Al species to MCM-41 at suitable Si/Al ratios showed great promotion effect on NO conversion as well as N2 selectivity. In situ FTIR spectra suggested that NH4 + is the key intermediate in H2 -SCR on these catalysts. In 2009, Itoh et al. investigated the catalytic performance of Pt supported on rare earth oxides for H2 -SCR [28]. Among Pt-supported rare earth oxide catalysts, only Pt/CeO2 showed good NO conversion, whereas the other catalysts showed very low NOx reduction activity due to the high solid basicity of the support materials. Perovskite is another interesting support material. In 2002, Costa et al. [29] reported the excellent activity of Pt/La0.5 Ce0.5 MnO3 catalyst for H2 -SCR. It is remarkable that the addition of H2 O into the feed stream widened the operating temperature window of the catalyst above 200 ◦ C with no negative effect on the catalyst stability. Thus, remarkable N2 selectivity values in the 80–90% range were observed in the 100–200 ◦ C range for H2 -SCR on the catalyst. A similar catalytic performance was observed for Pt/La0.7 Sr0.2 Ce0.1 FeO3 [30]. The Pt loading of the catalysts was only 0.1%. They also made transient isotopic kinetic studies to compare the reaction mechanism of H2 -SCR over Pt/La–Ce–Mn–O and Pt/SiO2 [31]. In the case of the Pt/SiO2 catalyst, NO reduction proceeds through the interaction of one reversibly chemisorbed and one irreversibly chemisorbed NOx species on the Pt surface, whereas this reduction proceeds through the interaction of two different NOx species irreversibly chemisorbed on the support in the case of Pt/La–Ce–Mn–O catalyst leading to N2 formation. It was found that the reaction path from NO to N2 O passes via oxidation of NO to NO2 with O2 . Costa and Efstathiou [32] also investigated the activity of 0.1% Pt supported on various metal oxides; they found that Pt/MgO–CeO2 catalyst prepared by sol–gel method showed excellent activity and selectivity to N2 formation. The authors claimed that this catalyst appears to be the most active, selective and stable one ever reported in the literature for H2 -SCR even in the presence of 5% H2 O or 20 ppm SO2 in the feed stream. The performance of Pt/MgO–CeO2 catalyst under process conditions was also evaluated in the low-temperature range of 150–200 ◦ C [33]. Steady state isotopic transient kinetic analysis coupled with temperature-programmed surface reaction experiment [34] suggests that the N-pathways of the reaction involve two different active chemisorbed NOx species, one present on the MgO and the other one on the CeO2 support surface. The influences of reaction temperature on the chemical structure and the surface concentration of active NOx in H2 -SCR over Pt/MgO–CeO2 were also studied in detail [35]. 2.2. Pd catalysts In 1998, Ueda et al. [36] investigated the catalytic activity of 1% Pt and Pd catalysts on various supports for H2 -SCR under the reaction

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Fig. 2. Catalytic activity of 1% Pd/Al2 O3 (), 1% Pd/TiO2 (), 1% Pt/Al2 O3 (䊉) and 1% Pt/TiO2 () for H2 -SCR [36]. NO = 1000 ppm, H2 = 3000 ppm, O2 = 5%, H2 O = 10%, W/F = 0.18 g s cm−3 .

conditions of 1000 ppm NO, 3000 ppm H2 , 5% O2 and 10% H2 O. As shown in Fig. 2, they found that Pd/TiO2 , in addition to Pt/Al2 O3 and Pt/TiO2 , showed good activity for H2 -SCR, with a conversion maximum at around 100 ◦ C, in contrast to the low activity of Pd/Al2 O3 . It is noteworthy that an additional conversion maximum was found at around 300 ◦ C over several Pd catalysts, especially on Pd/TiO2 . It was concluded that in situ generated NO2 reacts with H2 at 300 ◦ C. This paper confirms the activity of Pd for H2 -SCR as well as Pt. After the report by Ueda et al., Pd has also been studied extensively as the active catalyst species for H2 -SCR. In 2002, Wen [37] reported the activity of Pd/MFI zeolite catalyst prepared by sublimation method. A steady-state conversion of NO to N2 as high as 70% was achieved at 100 ◦ C over this catalyst. N2 was the only Ncontaining product with no formation of N2 O. Pd metal may be the active site for NO reduction and the oxidation of Pd metal with subsequent agglomeration of PdO would be responsible for the catalyst deactivation. The catalytic activity of Pd supported on pillared clays was reported by Qi et al. [38]. It was shown that the Ti-PILC-supported Pd catalyst gave two NO reduction peaks at around 140 and 240 ◦ C, with high NO conversion and N2 product selectivity, compared with Pd/TiO2 /Al2 O3 . The second NOx conversion peak was due to reaction between H2 and in situ generated NO2 , as already reported by Ueda et al. [36]. They also studied Pd/TiO2 –Al2 O3 [39] for H2 -SCR and found that the addition of V2 O5 to 1% Pd/TiO2 –Al2 O3 not only increased NO conversion but also widened the reaction temperature window to 250 ◦ C. Increasing the O2 concentration from 2% to 10% caused only a slight decrease in NO conversion. FT-IR studies showed that a significant amount of NH4 + was formed over the V2 O5 -containing catalyst while almost no NH4 + was formed over the V2 O5 -free sample, leading to the conclusion that the higher activity of 1% Pd–5% V2 O5 /TiO2 –Al2 O3 can partly be attributed to in situ formation of NH4 + . Perovskite oxides were also found to be effective as a support for Pd catalyst as well as Pt. Chiarello et al. [40] investigated the catalytic performance of 0.5% Pd/LaCoO3 prepared by flame-spray pyrolysis. This catalyst allowed full NO conversion at 150 ◦ C, with 78% selectivity to N2 . The high activity of the flame-spray pyrolysismade catalyst was attributed to the formation of segregated Co metal particles during the precalcination at 800 ◦ C, followed by reduction at 300 ◦ C. Two NO conversion maxima with respect to temperature were observed. The lower temperature peak was

5

ascribed to cooperation between the Pd and Co metal particles, with formation of active nitrates on cobalt reduced by hydrogen spillover from Pd, whereas the higher temperature peak involves N atoms formed by dissociative NO adsorption over Pd. Furfori et al. [41] examined a series of LaFeO3 -based catalysts for NO-H2 reactions. La0.8 Sr0.2 Fe0.9 Pd0.1 O3 showed the best catalytic performance, with 75% NO conversion at 130 ◦ C under the conditions of 1000 ppm NO, 1% H2 , and 5% O2 . The catalytic performance of the catalyst supported on a ceramic honeycomb monolith was also tested in a lab-scale test. A mechanistic analysis was also presented concerning the relationship between the activity and the reducibility of the B site determined by TPR experiments. Li et al. [42] reported the catalytic activity of Pd on K2 O–6TiO2 nanowires prepared by alkali treatment of TiO2 for H2 -SCR. Although NO conversion on 1% Pd/K2 O–6TiO2 was not so high (max 40%), the high N2 selectivity of 80% at maximum was noted with only 11% conversion of H2 . On the other hand, Pd/Al2 O3 and Pd/TiO2 exhibited maximum NO conversion values of about 50–60% at 100–130 ◦ C with N2 selectivity less than 40%. In situ DRIFT spectroscopy revealed surface-fixed nitrates and Pd-bound NO, but no NHx ad-species. It was concluded that the beneficial effect of the K2 O–6TiO2 support is due to stabilization of highly dispersed Pd with enhanced concentration of Pd0 –NO intermediates as well as the support basicity to enhance nitrate fixation. 2.3. Ir and Rh catalysts As described before, Pt and Pd generally show good catalytic activity for H2 -SCR, while the other Pt group metals have been reported to show little or no activity for the reaction. Since actual exhaust gases contain various coexisting gases, their influence is important from a practical viewpoint. In particular, the effect of SO2 on the catalytic activity is important for diesel applications, but this effect has not been investigated systematically. In 2001, Yoshinari et al. [43] reported a remarkable catalytic activity of Ir supported on SiO2 for H2 -SCR, the performance of which is depicted in Fig. 3. In the absence of O2 , Ir/SiO2 showed high activity for NO reduction with H2 above 300 ◦ C, but the increase of O2 concentration decreased NO conversion, which is a typical result in non-selective NO reduction. However, this situation was completely changed when SO2 was added to the reaction gas. In the

Fig. 3. Effect of O2 concentration on the activity of 0.5% Ir/SiO2 for NO reduction with H2 in the presence of SO2 [43]. NO = 1000 ppm, H2 = 3000 ppm, H2 O = 10%, SO2 = 20 ppm, W/F = 0.0267 g s cm−3 . NO conversion to N2 : (䊉) 0% O2 , () 0.65% O2 , () 1.3% O2 , () 3.9% O2 . H2 conversion: () 0% O2 , () 0.65% O2 , (♦) 1.3% O2 , () 3.9% O2 . Gray symbols indicate NO conversions to N2 O.

6

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Table 2 Activity of SiO2 -supported platinum group catalysts for H2 -SCR in the presence and absence of SO2 [44]. Catalyst

SO2 (ppm)

NO conversion to N2 (N2 O) (%)

0 20 0 20 0 20 0 20 0 20

0.5% Pt/SiO2 0.5% Pd/SiO2 0.5% Ru/SiO2 0.5% Ir/SiO2 0.5% Rh/SiO2

150 ◦ C

200 ◦ C

300 ◦ C

400 ◦ C

500 ◦ C

600 ◦ C

16 (34) 1 (0)

5 (10) 1 (0) 13 (5) 1 (0) 0 (0) 0 (2) 0 (0) 5 (4) 3 (0) 5 (3)

1 (2) 4 (4) 3 (1) 2 (1) 0 (0) 1 (0) 4 (0) 45 (28) 17 (2) 48 (5)

2 (1) 5 (3) 0 (1) 1 (1) 1 (0) 2 (0) 11 (2) 35 (8) 10 (1) 52 (1)

2 (1) 1 (1) 3 (0) 1 (1) 2 (0) 4 (0) 11 (1) 17 (1) 8 (0) 28 (1)

1 (1) 1 (1) 5 (1) 1 (1) 1 (0) 2 (0) 14 (0) 12 (1) 6 (1) 10 (1)

NO = 1000 ppm, H2 = 3000 ppm, O2 = 0.65%, H2 O = 10%, W/F = 0.0267 g s cm−3 .

presence of 20 ppm SO2 , Ir/SiO2 could not catalyze NO–H2 reaction (NO conversion less than 10% even at 600 ◦ C), indicating the poisoning effect of SO2 . On the other hand, it was found that NO reduction proceeds efficiently in the presence of 0.65% O2 with 73% NO conversion to N2 and N2 O at 300 ◦ C, although NO conversion tends to decrease at higher O2 concentrations. The experimental results clearly indicate that H2 -SCR takes place on Ir/SiO2 when SO2 coexists. Yoshinari et al. [44] screened SiO2 -supported Pt group metals for the H2 -SCR activity in the absence and presence of 20 ppm SO2 . The results are summarized in Table 2. Pt/SiO2 showed the highest activity for NO reduction to N2 and N2 O in the absence of SO2 . Pd/SiO2 was also active for H2 -SCR. However, the presence of SO2 considerably inhibited NO reduction on Pt/SiO2 and Pd/SiO2 . Ru/SiO2 was almost inactive irrespective of the presence of SO2 . Interestingly, NO reduction proceeded on Rh/SiO2 as well as Ir/SiO2 in the presence of SO2 , whereas NO was not reduced in the absence of SO2 . This result indicates that Rh/SiO2 as well as Ir/SiO2 show H2 SCR activity in the presence of SO2 . It should be noted from Table 3 that the SO2 promotion effect on SCR over Ir/SiO2 was observed not only for H2 but also for CO and various hydrocarbons. In the case of hydrocarbons, propene, n-decane and acetone served as an effective reductant, although propane did not serve as a reductant either in the absence or in the presence of SO2 . The CO-SCR reaction will be described in detail in the latter part of this review. Concerning the effect of support, the best support was found to be SiO2 for both Rh and Ir, although mild SO2 promoting effects were also observed for other supports such as Al2 O3 , TiO2 and zeolites. Table 4 shows the catalytic activity of Ir/SiO2 for the reaction of various gas mixtures with components chosen from NO, NO2 , O2 , H2 , and SO2 . Since NO was reduced more easily than NO2 , NO2 is probably not a reaction intermediate for the SCR of NO, although the SO2 promoting effect was also observed for NO2 reduction. As

Table 4 Several unit reactions over 0.5% Ir/SiO2 [44]. Reaction

NOx conversion to N2 (N2 O) (%) ◦

NO–H2 NO–H2 –SO2 NO–H2 –O2 NO–H2 –O2 –SO2 NO2 –H2 –O2 NO2 –H2 –O2 –SO2 H2 –O2 H2 –O2 –SO2





H2 conversion (%)

300 C

400 C

500 C

300 ◦ C

400 ◦ C

500 ◦ C

44 (0) 0 (0) 4 (0) 45 (28) 1 (0) 3 (0)

89 (0) 1 (1) 11 (2) 35 (8) 2 (0) 9 (0)

98 (0) 2 (3) 11 (1) 17 (1) 5 (0) 11 (1)

82 2 2 90 12 20 23 69

47 7 28 97 23 56 43 83

40 22 66 100 48 85 68 100

NO = 1000 ppm, NO2 = 1000 ppm, W/F = 0.0267 g s cm−3 .

H2 = 3000 ppm,

O2 = 0.65%,

H2 O = 10%,

already described, SO2 inhibited the NO reduction with H2 in the absence of O2 . The effect of SO2 on H2 –O2 reaction is interesting because this is a side reaction inhibiting the reaction of NO and H2 . The table shows that SO2 did not hinder but enhanced the H2 –O2 reaction, indicating that SO2 also has a promoting effect. Consequently, we conclude that SO2 does promote the essential catalytic activity of supported Ir species. The detailed SO2 promoting effect on H2 -SCR over Ir/SiO2 [44] and Rh/SiO2 [45] was further investigated by experiments using intermittent feeds of SO2 . As shown in Fig. 4, NO conversion

Table 3 Selective reduction of NO over 0.5% Ir/SiO2 with various reductants [44]. Reductant (ppm)

H2 (3000) CO (3000) C3 H6 (1000) C3 H8 (1000) n-C10 H24 (300) (CH3 )2 CO (1000)

SO2 (ppm)

0 20 0 20 0 20 0 20 0 20 0 20

NO conversion to N2 (N2 O) (%) 300 ◦ C

400 ◦ C

500 ◦ C

600 ◦ C

4 (0) 45 (28) 1 (0) 23 (7) 0 (0) 1 (1) 1 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (1)

11 (2) 35 (8) 23 (4) 49 (9) 1 (0) 24 (7) 1 (0) 0 (1) 0 (0) 3 (2) 1 (0) 6 (2)

11 (1) 17 (1) 9 (2) 31 (3) 61 (2) 74 (5) 1 (0) 1 (1) 1 (0) 70 (3) 4 (0) 16 (1)

14 (0) 12 (1) 3 (0) 4 (1) 91 (1) 81 (2) 1 (0) 1 (1) 19 (1) 37 (3) 10 (1) 28 (1)

NO = 1000 ppm, O2 = 0.65%, H2 O = 10%, W/F = 0.0267 g s cm−3 .

Fig. 4. Response of NO and H2 conversion to the intermittent feed of SO2 over 0.5% Ir/SiO2 catalyst [44]. NO = 1000 ppm, H2 = 3000 ppm, O2 = 0.65%, H2 O = 10%, SO2 = 0 or 20 ppm, T = 400 ◦ C, W/F = 0.0267 g s cm−3 . () NO conversion to N2 , () NO conversion to N2 O, () H2 conversion.

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

7

Fig. 6. Effect of 150 ppm SO2 addition on the catalytic activities of 0.02% Ir/SiO2 (), 0.02% Ir/Al2 O3 () and 0.02 t% Ir/silicalite (䊉) at 400 ◦ C [49]. NO = 1000 ppm, CO = 8000 ppm, O2 = 1%, total flow rate = 100 cm3 min−1 , catalyst weight = 0.1 g. Fig. 5. Effect of alkali metal additives on the catalytic activity of 5% Ir/SiO2 for H2 -SCR [46]. NO = 1000 ppm, H2 = 6000 ppm, O2 = 5%, H2 O = 6%, SO2 = 20 ppm, W/F = 0.0267 g s cm−3 . () no additive, () 0.06% Li, (䊉) 0.2% Na, () 0.34% K, () 0.73% Rb, () 1.1% Cs.

increased quickly after the introduction of SO2 . However, the subsequent removal of SO2 did not result in a rapid decrease of NO conversion. This means that the effect of coexisting SO2 was not completely lost after the removal of SO2 in the gas phase, suggesting that surface adsorbed SO2 plays an important role in the activity promotion. The detailed mechanism of SO2 promoting effect will be described later in the section on CO-SCR reactions. In order to improve the catalytic activity of Ir/SiO2 , Hamada et al. [46] investigated metal additive effects and found that additions of alkali and alkaline earth metal oxides are effective in promoting the NO reduction activity. The extent of the promoting effect was in the order of Li > Na> K > Rb > Cs for alkali metals (Fig. 5) and Ba > Sr > Ca > Mg for alkaline earth metals, among which Li was the most effective. Interestingly, the promoting effect was observed especially at high O2 concentrations. From XRD measurements, it was concluded that Ir metal is the active species and that the additives prevent the catalyst deactivation by O2 as an oxidationretardant for Ir metal. In the case of Rh/SiO2 , Hasegawa et al. [47] found that Zn is the best additive for promoting the NO reduction activity. The structural characterizations of ZnO/Rh/SiO2 catalyst and in situ FT-IR measurements revealed that Zn additive also prevents catalyst deactivation as an oxidation-retardant for the active Rh metal species and promotes a reaction step that generates NHx species as the key reaction intermediate. 3. Selective catalytic reduction with carbon monoxide (CO-SCR) 3.1. Catalytic activity CO is one of the most practical reductants for the removal of NO because it is present in exhaust emissions from vehicles or is relatively easily produced by engine control. However, CO was not regarded as an effective reductant for SCR until recently because CO is easily consumed by the reaction with O2 when supported noble metal catalysts are employed. CO-SCR was first reported by Tauster and Murrell [48]. They measured the catalytic activity of 0.1% Ir/Al2 O3 (0.1 g) for CO-SCR using a reaction gas mixture composed of 0.2% NO, 1.0% CO and 0.75% O2 diluted in He at a flow rate of 100 L h−1 . They reported that the NO conversion was as high as 90% at 400 ◦ C. This indicates that NO reacts preferentially with CO rather than with O2 . The interesting catalytic performance of Ir/Al2 O3 was

explained by a higher probability of adsorbing NO molecules onto a surface site compared with O2 molecules. They also emphasized that 0.001% Ir/Al2 O3 is also highly active for the CO-SCR, although the effective temperature window shifts to a region about 100 ◦ C higher. After the first report by Tauster and Murrell, various materials have been investigated in order to develop effective catalysts for CO-SCR. The CO-SCR performance of various catalysts reported so far is summarized in Table 5. In 2000, Ogura et al. [49] reported that NO can be successfully reduced to N2 with CO over 0.02% Ir/silicalite catalyst under the reaction conditions of 1000 ppm NO, 7500 ppm CO and 1% O2 . In addition to Ir/silicalite, Ir/SiO2 and Ir/Al2 O3 can also effectively catalyze the CO-SCR reaction, although the latter is not so active. The above-mentioned reports suggest that Ir is a promising catalytic component for CO-SCR. Ogura et al. examined the effect of SO2 on the activity of Ir/silicalite, Ir/SiO2 and Ir/Al2 O3 for CO-SCR. As given in Fig. 6, Ir/SiO2 showed higher NO conversion than Ir/silicalite in the initial run in the absence of SO2 at 400 ◦ C. However, NO conversion on Ir/SiO2 dropped by addition of 150 ppm SO2 , although the activity of Ir/SiO2 was recovered by elimination of SO2 from the reaction gas. On the other hand, it is of interest that the catalytic activity of Ir/silicalite was not influenced by coexisting 150 ppm SO2 . Yoshinari et al. [43,44] and Haneda et al. [50,51] examined the effects of metal oxide support on the activity of Ir catalysts for COSCR, which was carried out in the absence and presence of SO2 . They found that CO-SCR takes place quite efficiently (NO conversion 49% (to N2 ) and 9% (to N2 O) at 400 ◦ C) on Ir/SiO2 when SO2 coexists in the reaction gas, although the maximum NO conversion on Ir/Al2 O3 , Ir/TiO2 and Ir/ZSM-5 catalysts was as low as 15%. The most outstanding feature was that the coexistence of SO2 and O2 is essential for NO reduction on Ir/SiO2 to occur. This is quite a favorable characteristic for the treatment of diesel exhaust containing SO2 . The positive effect of SO2 , which is also observed for H2 -SCR as described before, is discussed in detail later in this review. Accoroding to the reports by Ogura et al. [49], Yoshinari et al. [43,44] and Haneda et al. [50,51], silica-based materilas such as silicalite and SiO2 seem to be effective supports for Ir. On the other hand, Wang et al. [52] reported that Ir/ZSM-5 (Si/Al = 50) was active for NO reduction with CO under lean conditions. Shimokawabe et al. [53,54] measured the catalytic activity of Ir catalysts supported on various metal oxides. They indicated that the use of metal oxides with high oxidation number such as WO3 , Nb2 O5 and Ta2 O5 is effective for the CO-SCR reaction. Thus, metal oxide support would play an important role in the catalytic activity of Ir. In addition to Ir, various supported noble metal catalysts, such as Pt/Al2 O3 [55,56], Pt/SiO2 [57], Pt/TiO2 [58], Pt/WO3 /CeZrO [59],

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

n.a. n.a. n.a. 9.0 6.0 n.a. 3.8 2.4 0 30 n.a. 4 0 4.7 0.3 n.a.

c

The activity was evaluated in the presence of 0.1% C3 H8 , 0.5% H2 and 12% CO2 . The activity was evaluated in the presence of 10% CO2 . The activity was evaluated in the presence of 0.5% H2 and 12% CO2 . a

b

2,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 500 5,000 2,000 230 500 10,000 400

1.0 0.75 0.75 0.3 0.6 1.5 0.3 0.3 0.3 1.5 0.5 0.6 0.069 1.5 1.0 0.04

0.75 1.0 1.0 0.65 5.0 2.0 5.0 5.0 5.0 1.5 2.0 5.0 10.5 9.0 0.5 2.0

0 0 150 20 20 0 2 2 2 0 0 0 0 0 0 0

0 0 0 10 6 0 1 1 1 0 0 6 0 6 0 0

0.036 g s cm−3 0.06 g s cm−3 0.06 g s cm−3 0.027 g s cm−3 0.027 g s cm−3 50000 h−1 0.06 g s cm−3 0.06 g s cm−3 0.06 g s cm−3 0.1 g s cm−3 0.115 g s cm−3 n.a. 0.075 g s cm−3 50000 h−1 0.12 g s cm−3 50000 h−1

3.2. Ir as an active species for CO-SCR

0.1% Ir/Al2 O3 0.02% Ir/silicalite 0.02% Ir/silicalite 0.5% Ir/SiO2 5% Ir/SiO2 0.1% Ir/ZSM-5a 5% Ir/WO3 5% Ir/Nb2 O5 5% Ir/Ta2 O5 1% Pt/TiO2 1% Pt/WO3 /CeZrO 2% Pt/12.3% MoO3 –0.11% Na2 O/SiO2 b 3% Pd/Ce0.6 Zr0.4 O2 2% Rh/H-Betac 0.5% Cu/Al2 O3 10% MnO2 /TiO2

NO (ppm)

CO (%)

O2 (%)

SO2 (ppm)

H2 O (%)

W/F (g s cm−3 ) or GHSV (h−1 )

400 370 400 400 350 310 300 320 325 400 350 450 400 350 500 200

90 95 55 49 62 92 25 14 25 60 10 16 59 56 7.7 90

Ref. To N2 O (%) To N2 (%)

NO conversion Tmax (◦ C) Reaction conditions Catalysts

Table 5 Activity of various catalysts for CO-SCR.

Pt–Mo–Na/SiO2 [16], Pd/Ce0.6 Zr0.4 O2 [60] and Rh/Na-Beta zeolite [61], have been reported to show activity for CO-SCR. Although comparison of the catalytic activity under the same reaction conditions is difficult, their catalytic activity seems to be not so high, compared with that of Ir catalysts. Among the reports, Nakatsuji et al. [61] emphasized that the activity of Rh/Na-Beta zeolite is dramatically enhanced by the presence of SO2 . However, Haneda et al. [51] measured the activity of silica-supported noble metal catalysts and found that Pt/SiO2 , Rh/SiO2 and Pd/SiO2 showed little catalytic activity for CO-SCR, irrespective of coexisting SO2 (Table 6). The CO-SCR reaction was also reported to take place over supported metal oxide catalysts such as Cu/Al2 O3 [62,63] and over TiO2 -supported transition metal oxide catalysts (MOx/TiO2 ; M = Cr, Mn, Fe, Ni, Cu) [64]. Comparison of the activity of various supported catalysts under the same conditions, however, revealed that the activity of Cu/Al2 O3 is not very high, and that supported Ir catalyst is the most active [53]. Taking into account the reports published so far, Ir is probably the most effective catalytically active metal for CO-SCR.

[48] [49] [49] [43,44] [51] [52] [54] [54] [54] [58] [59] [16] [60] [61] [62] [64]

8

In order to gain insight into understanding the specific properties of Ir, researchers have studied the reaction behavior of NO or CO on the well-defined Ir crystal surfaces by means of surface science techniques [65–72]. The exposure of NO onto Ir (2 1 1) surface gives NO species adsorbed on the atop site of (1 1 1) terrace and the bridge site of the (1 0 0) step [67]. On faceted Ir (2 1 0) that contains (1 1 0) and (3 1 1) faces, NO adsorbs on atop and bridge sites at low NO coverage but only on atop sites at high NO coverage [68]. Both atop NO and bridge NO is decomposed to N2 through the recombination of atomic nitrogen produced by dissociation of adsorbed NO species. The NO adsorbed on Ir (1 1 1) surface is also decomposed to N2 [65]. On the other hand, the adsorption of CO onto Ir single crystal surface causes the formation of CO species adsorbed only on atop sites, irrespective of the type of crystal surface [67,69–72]. The coadsorption of and reaction between NO and CO on Ir single crystal surfaces have also been investigated. Fujitani et al. [73–77] intensively investigated the adsorption behavior and reaction properties of NO and CO on Ir (1 1 1) and Rh (1 1 1). Two NO adsorption states, indicative of fcc-hollow (bridge) sites and atop sites, were revealed to be present on the Ir (1 1 1) surface at saturated coverage by means of X-ray photoelectron spectroscopy (XPS), high-resolution electron energy loss spectroscopy (HR-EELS) and infrared reflection absorption spectroscopy (IRAS). NO adsorbed on hollow sites dissociated to Na and Oa at temperatures above 10 ◦ C. The dissociated Na desorbed to form N2 by recombination of Na at 301 ◦ C and by a disproportionation reaction between atop-NO and Na at 198 ◦ C. CO species preadsorbed on Ir (1 1 1) inhibited the adsorption of NO on atop sites, whereas adsorption on hollow sites was not affected by the coexistence of CO. The adsorbed CO reacted with dissociated Oa and desorbed as CO2 at 301 ◦ C. A similar reaction mechanism has also been proposed by Chen et al. [78], who investigated adsorption sites and reactions of coadsorbed NO and CO on planar Ir (2 1 0) and faceted Ir (2 1 0) by means of temperature programmed desorption (TPD) and density functional theory (DFT). Fujitani et al. [75–77] also reported that adsorption behavior and reaction properties of NO and CO-preadsorbed on Rh (1 1 1) are quite different from those on Ir (1 1 1). NO adsorption takes place on fcc-hollow sites at low NO coverage, while it occurs on atop and hcp-hollow sites at high NO coverage. CO adsorbs initially on the atop sites and then on the hollow (fcc + hcp) sites. The adsorption of NO on the fcc-hollow and atop sites is inhibited by CO species preadsorbed on each type of site (atop and/or hollow) of Rh (1 1 1). Namely, NO and CO seem to competitively adsorb on Rh (1 1 1). This

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

9

Table 6 Activity of SiO2 -supported platinum group metal catalysts for CO-SCR in the presence and absence of SO2 [51]. Catalyst

5% Pt/SiO2 5% Pd/SiO2 5% Ir/SiO2 5% Rh/SiO2

SO2 (ppm)

0 20 0 20 0 20 0 20

NO conversion to N2 (N2 O) (%) 150 ◦ C

200 ◦ C

300 ◦ C

350 ◦ C

400 ◦ C

450 ◦ C

500 ◦ C

0 (3.2) 0 (0.5) 0 (2.3)

0 (1.2) 0 (1.0) 0 (0.2) 0 (0.7) 7.4 (2.2) 3.3 (0.9) 0 (0.7) 0 (0.6)

0 (1.3) 0 (0.7) 0 (0.2) 0 (1.2) 54 (3.6) 11 (2.4) 0 (0.2) 0 (1.1)

0 (0.3) 0 (0.4) 0 (0) 0 (0.7) 61 (2.0) 62 (6.0) 0 (0.2) 0 (1.7)

0 (0.2) 0 (0.6) 0 (0) 0 (0.5) 43 (1.5) 28 (3.2) 0 (0) 0 (1.3)

0 (0) 0 (0.3) 0 (0) 0 (0.2) 44 (0) 14 (1.5) 0 (0) 0 (0.6)

0 (0) 0 (0.2) 0 (0) 0 (0) 79 (0) 4.1 (0.8) 0 (0) 0 (0.2)

0 (0.3)

NO = 1000 ppm, CO = 6000 ppm, O2 = 5%, SO2 = 0 or 20 ppm, H2 O = 6%, W/F = 0.0267 g s cm−3 .

finding clearly suggests that the NO reduction by CO on Rh catalyst is inhibited in the presence of excess CO. This finding is in accordance with the reports by Haneda et al. [50,51] in which Rh/SiO2 showed little catalytic activity for CO-SCR under the reaction conditions using a reaction gas mixture composed of 1000 ppm NO, 5% O2 , 6000 ppm CO and 6% H2 O. In conclusion, two NO adsorption sites, atop and hollow sites, are present on the Ir surface, and NO adsorption on the latter sites does not compete with CO adsorption, although NO and CO competitively adsorb on Rh surface. The specific property of the Ir surface would be responsible for the high catalytic activity for CO-SCR reactions. 3.3. Catalytically active state of Ir The CO-SCR reaction over Ir/SiO2 pretreated with H2 was found to be a structure-sensitive reaction from the relationship between the turnover frequency (TOF) and Ir dispersion [79]. Fig. 7 shows the change in the TOF at 200 and 220 ◦ C on 1% Ir/SiO2 as a function of Ir dispersion. The TOF gradually increased with decreasing Ir dispersion at both temperatures. In particular, the TOF at 220 ◦ C increased rapidly in the dispersion region below 10%, suggesting that very large crystallites have high specific activity. However, it should be noted that the increased activity per surface atom at low dispersion is offset by the decrease in the number of surface atoms. It was revealed that the activity of Ir/SiO2 for CO-SCR strongly depends on pretreatment conditions. As shown in Fig. 8, Ir/SiO2 pretreated with O2 showed little activity for NO reduction, whereas high activity was achieved for Ir/SiO2 reduced with H2 , suggesting

Fig. 7. TOF for NO reduction with CO at 200 (䊉) and 220 ◦ C () on 1% Ir/SiO2 as a function of Ir disperison [79]. NO = 500 ppm, CO = 3000 ppm, O2 = 10%, SO2 = 1 ppm, H2 O = 1%, total flow rate = 90 cm3 min−1 , catalyst weight = 0.04 g.

that Ir metal rather than Ir oxide is the catalytically active species [51]. Therefore, it can be concluded that the low catalytic activity of highly dispersed Ir on SiO2 is due to catalyst deactivation by oxidation of active Ir metal to IrO2 . Actually, characterization of Ir/SiO2 using TPO, XRD and FT-IR spectroscopy following CO adsorption showed that larger Ir crystallites are more difficult to oxidize and much easier to stabilize in metallic state under reaction conditions [79]. 3.4. Additive effect for Ir catalysts As described in Section 3.1., the catalytic performance of Ir catalyst is dependent on the metal oxide support employed. In addition to SiO2 [43,44,50,51], metal oxides with a high oxidation number such as WO3 , Nb2 O5 and Ta2 O5 are also effective [53,54]. In particular, Ir/WO3 is an interesting catalyst, which shows high catalytic activity for CO-SCR both in the absence and presence of SO2 [80,81]. However, the specific surface area of the metal oxides is not so high (less than 20 m2 g−1 ). Therefore, it is expected that an increment in the surface area of the metal oxides would enhance the catalytic activity of Ir catalysts. Nanba et al. [82,83] and Takahashi et al. [84] independently investigated the effect of WO3 addition on high surface area SiO2 as a support for Ir catalyst. Nanba et al. [82,83] optimized the conditions for catalyst preparation and found that the optimum composition is 0.5% Ir and 1–30% WO3 (Fig. 9). The most effective pretreatment condition is sequential treatment involving

Fig. 8. Effect of pretreatment conditions on the activity of 5% Ir/SiO2 for NO reduction with CO in the presence of O2 and SO2 [51]. NO = 500 ppm, CO = 6000 ppm, O2 = 5%, SO2 = 20 ppm, H2 O = 6%, W/F = 0.0267 g s cm−3 . () treated in flowing 5% O2 /He at 600 ◦ C, () treated in flowing 10% H2 –6% H2 O/He at 600 ◦ C.

10

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Fig. 9. Dependence of CO-SCR activity on WO3 loading in the SiO2 support at 260 ◦ C [82]. NO = 500 ppm, CO = 5000 ppm, O2 = 10%, H2 O = 1%, total flow rate = 225 cm3 min−1 , catalyst weight = 0.1 g. (䊉) NOx conversion, () N2 selectivity.

calcination in the presence of O2 and H2 O, followed by reduction and then re-oxidation under mild conditions. TEM observations and H2 -TPR measurements suggested that metallic Ir and slightly reduced tungsten oxide (WO3−x : 0 < x < 1) species are produced by the second reduction step to create Ir–WO3−x interaction, and that the third re-oxidation under mild conditions decreases the reducibility of tungsten oxide, which is important for maintaining the active metallic Ir surface. Takahashi et al. [84] also reported that the catalytic activity of Ir/WO3 /SiO2 catalyst strongly depends on the catalyst pretreatment conditions. The activity of Ir/WO3 /SiO2 is significantly enhanced by treatment in 6% H2 O/He at 600 ◦ C after reduction in H2 at 600 ◦ C, whereas the activity of Ir/WO3 /SiO2 oxidized at 600 ◦ C following reduction is quite low (Table 7). On the basis of structural characterizations of the catalysts treated under different conditions, it was concluded that Ir metal species interacting strongly with W oxide (denoted as Ir–WOx) are created by H2 O treatment as the active species for NO reduction with CO. This suggests that the dispersion of Ir and W oxide and their interactions are key factors in determining the catalytic activity of Ir/WO3 /SiO2 . It is noted that the Ir–WOx species is preferentially created by hightemperature air calcination [85]. As for additives to SiO2 support other than WO3 , Tamai et al. [86] reported the promotive effect of Nb2 O5 on the activity of Ir/SiO2 for the CO-SCR. It is of interest that adding Nb2 O5 to Ir/SiO2 (Nb2 O5 /Ir/SiO2 ) is more effective than using Nb2 O5 /SiO2 support for Ir (Ir/Nb2 O5 /SiO2 ). The addition of Nb2 O5 makes Ir metal, which Table 7 Effect of pretreatment on the catalytic activity of 5% Ir/10% WO3 /SiO2 for CO-SCR [84]. Pretreatment condition

NO conversion (%) (N2 selectivity (%))

CO conversion (%)

Without pretreatment 5% O2 in He 6% H2 O in He 6% H2 O + 3000 ppm CO in He 6% H2 O + 5% O2 in He 6% H2 O + 5% O2 + 3000 ppm CO in He

28 (90) 0 (0) 53 (91) 42 (89) 0 (0) 50 (89)

92 3 98 99 2 35

Pretreatment temperature: O2 = 5%, CO = 3000 ppm, W/F = 0.0267 g s cm−3 .

600 ◦ C. Reaction conditions: H2 O = 6%, SO2 = 1 ppm,

NO = 500 ppm, Temp. = 280 ◦ C,

is the catalytically active species, more resistant against oxidation. This is probably the main reason for the improvement of activity. Nb2 O5 /Ir/SiO2 having 10% Nb2 O5 showed a maximum NO conversion of 80%. The excellent activity of Nb2 O5 /Ir/SiO2 can be attributed to the strong interaction of Ir with Nb2 O5 , which is similar to that of Ir and WO3 . The role of the additives to SiO2 support is to stabilize the catalytically active Ir species in metallic state by the strong interaction with Ir. The strong interaction of Ir–WO3 and Ir–Nb2 O5 causes the outstanding feature that Ir/WO3 /SiO2 and Nb2 O5 /Ir/SiO2 can effectively catalyze the CO-SCR reaction even in the absence of SO2 [82,83,86]. The same phenomenon was also observed for Ir/WO3 catalyst [80,81]. On the contrary, Ir/SiO2 shows high activity only in the presence of SO2 [43,44,50,51]. The high activity of Ir/WO3 /SiO2 and Nb2 O5 /Ir/SiO2 in the absence of SO2 is related to the role of SO2 in stabilizing the catalytically active reduced Ir site, which will be described later. Namely, the catalytically active reduced Ir site is stabilized by the interactions with WO3 and Nb2 O5 during the reaction even in the presence of excess O2 and absence of SO2 . The catalytic activity of supported Ir catalysts for the CO-SCR reaction is also effectively improved by other additives. Haneda et al. [87] screened the additive effect of 19 metal elements and found that the catalytic activity for NO reduction at temperatures below 360 ◦ C increased when Li, Na, Mg, Sr, Ba, W, Co, Ce, Zn, Pt, Rh, Au or Ru were doped to Ir/SiO2 (Table 8). In particular, a significant increase of activity was attained by Sr and Ba doping. The addition of K, Cs and Al caused an increase in the activity around 400 ◦ C. However, Cu and Pd decreased the activity of Ir/SiO2 over the entire temperature range. In the case of Ba, a molar ratio of Ba/Ir = 1/10 is the most effective to enhance the activity of Ir/SiO2 at low temperature. Temperature-programmed oxidation (TPO) of pre-reduced Ba/Ir/SiO2 indicated that the re-oxidation temperature of the reduced Ir species in Ba/Ir/SiO2 increases with increasing Ba loading up to a molar ratio of Ba/Ir = 1/10. FT-IR spectroscopy analysis following CO adsorption in the presence of O2 and SO2 revealed that the surface of Ir species is stabilized as metallic state by Ba additive under the reaction conditions. Consequently, it was concluded that the role of Ba is to prevent catalyst deterioration by retarding the oxidation of Ir metal, which is the catalytically active species. Ba additive can also promote the catalytic performance of Ir/WO3 /SiO2 by suppressing the oxidation of partially reduced WOx as well as Ir metal, leading to the stable Ir–WOx species in catalytically active state during the reaction [88,89]. The addition of a small amount of H2 into the reaction gas is also effective to stabilize the catalytically active Ir–WOx sites during the reaction, enhancing the catalytic activity of Ba/Ir/WO3 /SiO2 catalyst. [90]. Nanba et al. [89] reported a unique catalytic performance of the physical mixture of Ba/Ir/WO3 /SiO2 and inert SiO2 that NOx conversion increases with increasing amounts of SiO2 in the physical mixture. It was noted that a certain amount of HNCO species was formed over the physical mixture. HNCO-TPD revealed that Ba/Ir/WO3 /SiO2 itself possesses strong adsorption sites for NH3 formed by HNCO hydrolysis, whereas the physical mixture has a smaller number of strong adsorption sites for NH3 due to the poisoning effect of SiO2 additive. Although NH3 species adsorbed on Ba/Ir/WO3 /SiO2 are easily oxidized to NOx in the presence of NO and O2 as a side reaction, the unfavorable NH3 oxidation is suppressed on the physical mixture, improving N2 formation via the reaction of HNCO with NO. The durability of Ba/Ir/WO3 /SiO2 , which is one of the best catalysts for CO-SCR, was checked by using simulated exhaust gases in laboratory test [91–93]. The durability tests were performed by repeating heating (200 → 600 ◦ C) and cooling (600 → 200 ◦ C) cycles up to 100 times. Although Ir/SiO2 and Ir/WO3 /SiO2 catalysts were deactivated as the number of heating-cooling cycles increased,

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

11

Table 8 Catalytic activity of metal-doped 5% Ir/SiO2 (M/Ir = 1/10) for CO-SCR [87]. NO conversion to N2 (N2 O) (%)

Ir/SiO2 Li/Ir/SiO2 Na/Ir/SiO2 K/Ir/SiO2 Cs/Ir/SiO2 Mg/Ir/SiO2 Sr/Ir/SiO2 Ba/Ir/SiO2 W/Ir/SiO2 Mo/Ir/SiO2 Cu/Ir/SiO2 Co/Ir/SiO2 Ce/Ir/SiO2 Zn/Ir/SiO2 Al/Ir/SiO2 Pt/Ir/SiO2 Pd/Ir/SiO2 Rh/Ir/SiO2 Au/Ir/SiO2 Ru/Ir/SiO2

260 ◦ C

280 ◦ C

300 ◦ C

320 ◦ C

340 ◦ C

360 ◦ C

400 ◦ C

450 ◦ C

500 ◦ C

0.6 (0) 4.3 (1.0) 4.5 (1.2)

1.5 (0) 11 (2.2) 12 (2.5) 4.3 (0.4)

2.0 (0) 18 (4.2) 18 (3.8) 0.9 (0) 0.9 (0)

5.4 (1.1) 46 (7.8) 56 (7.2) 3.7 (1.0) 4.2 (0) 0.6 (0) 3.5 (0.5) 2.0 (0) 28 (4.6)

3.7 (0.7) 25 (3.5) 24 (3.7) 7.8 (1.2) 1.6 (0) 6.2 (2.0) 46 (9.4) 56 (8.4) 12 (2.2) 7.9 (1.2) 0.8 (0) 8.1 (1.5) 4.7 (0.9) 28 (5.6) 0.5 (0) 3.8 (0.7) 0.6 (0) 6.7 (3.7) 9.3 (1.5) 6.2 (1.1)

7.9 (1.2) 36 (4.1) 32 (4.0) 13 (1.6) 4.1 (0) 21 (3.0) 32 (7.4) 37 (7.0) 28 (3.9) 11 (2.0) 1.0 (0) 14 (2.1) 8.9 (0.7) 32 (4.7) 1.1 (0) 8.2 (1.1) 0.6 (0) 17 (5.3) 19 (2.4) 12 (1.9)

13 (1.6) 33 (3.9) 29 (3.5) 16 (1.9) 6.0 (1.0) 21 (3.2) 22 (4.5) 22 (4.8) 31 (5.6) 8.7 (2.2) 1.0 (0) 18 (2.1) 13 (1.7) 23 (3.7) 2.0 (0) 14 (1.5) 0.6 (0) 20 (4.8) 27 (2.9) 19 (2.6)

14 (1.6) 25 (2.5) 23 (2.8) 18 (2.0) 10 (1.3) 21 (2.6) 17 (2.6) 12 (2.8) 22 (5.5) 7.8 (2.2) 1.8 (0) 17 (1.9) 14 (1.6) 16 (2.2) 5.1 (0.9) 17 (1.5) 0.7 (0) 17 (3.5) 27 (2.9) 23 (3.0)

8.8 (1.2) 12 (1.3) 14 (1.7) 23 (2.2) 17 (1.8) 21 (2.6) 13 (1.5) 6.1 (1.7) 14 (3.2) 4.2 (1.5) 2.2 (0) 7.8 (0.6) 9.0 (1.3) 6.5 (0) 26 (3.6) 12 (1.0) 0.8 (0) 15 (2.2) 20 (1.9) 24 (2.9)

2.0 (0) 2.4 (0) 6.5 (0.9) 21 (2.0) 13 (0.9) 9.8 (1.5) 16 (2.1) 6.7 (1.3) 7.9 (1.7) 2.5 (0.6) 1.6 (0) 1.4 (0) 3.6 (0.8) 2.0 (0) 21 (3.5) 6.3 (1.3) 0.9 (0) 7.6 (1.6) 13 (1.6) 17 (2.8)

0.9 (0) 1.1 (0) 3.5 (0) 11 (1.2) 5.1 (0) 2.9 (0) 10 (1.5) 4.8 (1.1) 4.7 (0) 2.2 (0) 1.6 (0) 0.7 (0) 1.5 (0) 1.6 (0) 6.1 (1.6) 2.3 (0) 2.5 (0) 1.9 (0.7) 6.8 (1.2) 8.7 (1.7)

1.3 (0) 11 (2.3) 0.9 (0) 0.8 (0) 1.1 (0) 1.4 (0)

1.7 (0) 0.5 (0) 2.0 (1.5) 3.4 (0) 2.9 (0)

NO = 500 ppm, CO = 3000 ppm, O2 = 5%, SO2 = 1 ppm, H2 O = 6%, W/F = 0.0267 g s cm−3 .

Ba/Ir/WO3 /SiO2 showed high and stable catalytic performance even after 100 cycles: NO conversion on the former two catalysts was 45–50% at 400 ◦ C in the 100th run, while it was 63% at 391 ◦ C on the latter one. The practical performance of Ba/Ir/WO3 /SiO2 in monolithic form was also evaluated under real diesel exhaust gas conditions. It was confirmed that about 45% NO conversion can be attained at SV < 6000 h−1 . 3.5. Influence of coexisting gases 3.5.1. Effect of SO2 Exhaust gases usually contain SO2 that is poisonous to catalysts, although the sulfur contents in automotive fuels have recently been severely restricted in many countries. Therefore, reaction inhibition and catalyst poisoning by coexisting SO2 have been extensively studied for the SCR reactions. In many cases, coexisting SO2 causes either irreversible or reversible deactivation of HC-SCR catalysts. On the other hand, it is of interest that the H2 -SCR and CO-SCR reactions on Ir catalyst are not severely inhibited by coexisting SO2 . Ogura et al. [49] reported that the catalytic activity of 0.02% Ir/silicalite catalyst for CO-SCR is not influenced by coexisting 150 ppm SO2 (Fig. 6). Shimokawabe et al. [80,81] showed that the presence of 100 ppm SO2 causes a significant deactivation of Ir/WO3 for the CO-SCR, but its negative effect is completely removed by coexisting O2 , indicating that the coexistence of O2 and SO2 does not inhibit the CO-SCR reaction. Furthermore, it should be noted that the presence of both O2 and SO2 is essential for the H2 -SCR and CO-SCR reactions on Ir catalysts to occur as already described in this review. The H2 -SCR and CO-SCR reactions over Ir/SiO2 [43,44,50,51] and promoted Ir/SiO2 such as Ba/Ir/SiO2 [87] and Li/Ir/SiO2 [46] do not take place in the absence of SO2 , whereas their catalytic activity was significantly enhanced by the addition of 1–20 ppm SO2 into the reaction gas. It is also noteworthy that the NO reduction activity is not influenced by changing the SO2 concentration range from 1 to 150 ppm. Interested in the role of SO2 in the H2 -SCR and CO-SCR reactions over supported-Ir catalysts, Fujitani et al. [94] investigated the adsorption and reactivity of SO2 on the Ir (1 1 1) surfaces by surface science techniques. X-ray photoelectron spectroscopy (XPS) measurements showed that SO2 is molecularly adsorbed on the Ir (1 1 1) surface at −73 ◦ C. Then, the adsorbed SO2 on the Ir (1 1 1) surface undergoes disproportionation to atomic sulfur and SO3 at 27 ◦ C,

followed by SO3 desorption: 3SO2 (a) → 2SO3 (a) + S(a). The remaining atomic sulfur reacts with surface oxygen on Ir to be removed from the surface: S(a) + 2O(a) → 2SO2 (g). The Ir surface, thus, reverts to its initial metallic state. Haneda et al. [51] investigated the valence state of the Ir surface by FT-IR using CO as a probe molecule. Fig. 10 shows FT-IR spectra of CO species adsorbed onto Ir/SiO2 recorded in different compositions of flowing gas at 250 ◦ C. When 0.6% CO/He was exposed to Ir/SiO2 , a strong IR band at 2079 cm−1 assignable to linearly bonded CO onto Ir0 sites was observed (Fig. 10A(a)). The exposure of CO in the presence of O2 to Ir/SiO2 caused a shift of IR band to 2107 cm−1 indicating the adsorption of CO onto Irı+ sites (Fig. 10A(b)). The surface of Ir seems to be easily oxidized in the presence of O2 . On the other hand, when CO was exposed to Ir/SiO2 in the presence of O2 and SO2 , two distinct IR bands assignable to Ir0 –CO and Irı+ –CO were observed at 2074 and 2107 cm−1 , respectively (Fig. 10A(c)). This clearly indicates that coexisting SO2 participates in the stabilization of Ir0 sites on which NO reduction with CO proceeds. It is of interest that the introduction of SO2 into the CO–O2 /He flowing gas caused the appearance of an IR band at 2074 cm−1 due to Ir0 –CO, and its band intensity increased over time on stream (Fig. 10B). In conclusion, the role of coexisting SO2 is not only to stabilize but also to create Ir0 sites, which is the catalytically active species for the SCR as described in Section 3.3., in oxidizing atmospheres. This finding is in good agreement with the findings obtained on single-crystal model catalysts by Fujitani et al. [94]. 3.5.2. Effect of H2 O In addition to SO2 , car exhausts invariably contain high concentrations of H2 O. Therefore, the influence of H2 O on the SCR reactions has been extensively studied. For HC-SCR, the presence of H2 O was generally detrimental to NO reduction. On the other hand, coexisting H2 O does not affect the catalytic activity of Rh/SiO2 and Ir/SiO2 for the H2 -SCR and CO-SCR reactions [44,46]. No significant change in the NO conversion was observed as H2 O concentration was changed in the range of 0–10% H2 O. Furthermore, an interesting promoting effect of coexisting H2 O on the activity of Ir/WO3 /SiO2 catalyst for CO-SCR was observed [95,96]. As shown in Fig. 11, little NO reduction occurred over the entire temperature range when H2 O was not present in the reaction gas. On the other hand, the presence of 0.75% H2 O caused a significant increase in NO conversion as well as CO conversion,

12

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

Fig. 10. (A) FT-IR spectra of CO adsorbed on 5% Ir/SiO2 after the exposure of (a) 0.6% CO/He, (b) 0.6% CO–5% O2 /He and (c) 0.6% CO–5% O2 –20 ppm SO2 /He at 250 ◦ C for 30 min. (B) FT-IR spectra of CO adsorbed on 5% Ir/SiO2 after introducing 20 ppm SO2 into the 0.6% CO–5% O2 /He flowing for (a) 0, (b) 15 and (c) 30 min [51].

and the catalytic activity of Ir/WO3 /SiO2 increased with increasing H2 O concentration up to 6%. When the reaction was carried out from 400 ◦ C in the absence of H2 O, however, the NO conversions on Ir/WO3 /SiO2 were very similar to those in the presence of

H2 O. In addition, it was confirmed that the change of NO conversion to the intermittent feed of H2 O is not reversible, suggesting that coexisting H2 O does not directly participate in the reaction as a reactant. Structural characterizations by XRD and Raman spectroscopy revealed that the reduced Ir sites are stabilized and/or regenerated when exposed to high temperature in the presence of H2 O, the role of which is to produce H2 in situ via WGS reaction. H2 -TPR measurements indicated that the strong Ir–W oxide interaction is created during the reaction in the presence of H2 O. The creation of the strong Ir–W oxide interaction was also visually confirmed by the measurement of STEM/HAADF images as well as that of element distribution by EDX. Consequently, it was concluded that the promoting effect of coexisting H2 O can be attributed to WGS reactions to produce H2 , which causes Ir to be in reduced state, leading to the formation of the catalytically active Ir–WOx species.

4. Conclusions

Fig. 11. Effect of H2 O concentration on the activity of 0.5% Ir/10% WO3 /SiO2 for NO reduction with CO in the presence of O2 and SO2 [96]. NO = 500 ppm, CO = 3000 ppm, O2 = 5%, SO2 = 1 ppm, H2 O = 0–10%, total flow rate = 90 cm3 min−1 , catalyst weight = 0.04 g. () 0% H2 O, () 0.75% H2 O, () 1.5% H2 O, (䊉) 6% H2 O, () 10% H2 O.

Supported platinum-group metals are active for the selective reduction of NO with hydrogen (H2 -SCR) or CO (CO-SCR). In the case of H2 -SCR, Pt and Pd show high activity for NO reduction to N2 and N2 O. The oxidation state of Pt and the acidity/basicity of the support are important factors affecting the catalytic activity and selectivity of Pt. High N2 selectivity is realized by using acidic supports, which add an additional reaction route in which NH3 is formed via NO reduction by H2 on Pt surface, followed by storage of NH4 + on the acid site and selective N2 formation by the NH3 SCR mechanism. The activity of Pd is also much affected by support materials. While coexisting SO2 inhibits H2 -SCR on Pt and Pd, it promotes NO reduction on Ir and Rh catalyst especially when supported on SiO2 . Li and Zn additives improve the activity of Ir/SiO2 and Rh/SiO2 , respectively, by maintaining the active metal species on the catalysts. For CO-SCR, only Ir shows high activity. Coexisting SO2 is also essential for the SCR of NO on Ir/SiO2 . The role of SO2 for both H2 -SCR and CO-SCR is to reduce Ir species even in O2 -rich atmospheres and to keep Ir in the form of catalytically active Ir metal state. The addition of WO3 and Nb2 O5 promotes the catalytic activity of Ir/SiO2 . Ir-WOx composite species is the active site on the

H. Hamada, M. Haneda / Applied Catalysis A: General 421–422 (2012) 1–13

WO3 -promoted Ir/SiO2 , which enables NO reduction even in the absence of SO2 . The addition of Ba improves further the activity of Ir/WO3 /SiO2 catalyst. The Ba-added Ir/WO3 /SiO2 can remove NOx effectively in actual diesel exhaust by CO-SCR. Acknowledgements A part of this work has been supported by the New Energy and Industrial Technology Development Organization (NEDO) under the sponsorship of the Ministry of Economy, Trade and Industry (METI) of Japan. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

[31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45]

R. Burch, J.P. Breen, F.C. Meunier, Appl. Catal. B 39 (2002) 283–303. R. Burch, Catal. Rev. 46 (2004) 271–334. Z. Liu, S.I. Woo, Catal. Rev. 48 (2006) 43–89. M. Iwamoto, H. Yahiro, Y. Yu-u, S. Shundo, N. Mizuno, Appl. Catal. 69 (1991) L15–L19. W. Held, A. König, T., Richter, L. Puppe, SAE Paper 900496, 1990. W. Held, A. König, Ger. Offen. DE 3642018 (Volkswagen), 1987. Y. Fujitani, H., Muraki, S., Kondoh, M. Fukui, Ger. Offen. DE 3735151 (Toyota Cetral R&D Labs.), 1988. M. Iwamoto, H. Yahiro, Catal. Today 22 (1994) 5–18. H. Hamada, Catal. Today 22 (1994) 21–40. R. Burch, Catal. Today 26 (1995) 185–206. D. Jun, K. Ishi, N. Iida, JSME Int. J. Ser. B 46 (2003) 60–67. J.H. Jones, J.T. Kummer, K. Otto, M. Shelef, E.E. Weaver, Environ. Sci. Technol. 5 (1971) 790–798. A. Lamb, E.L. Tollefson, Can. J. Chem. Eng. 53 (1975) 68–73. L. Fu, K.T. Chuang, Energy Fuels 3 (1989) 740–743. A. Wildermann, Ph.D. Thesis, University of Erlangen, Nurnberg, 1994. K. Yokota, M. Fukui, T. Tanaka, Appl. Surf. Sci. 121-122 (1997) 273–277. R. Burch, M.D. Coleman, J. Catal. 208 (2002) 435–447. B. Frank, G. Emig, A. Renken, Appl. Catal. B 19 (1998) 45–57. R. Burch, M.D. Coleman, Appl. Catal. B 23 (1999) 115–121. M. Machida, S. Ikeda, D. Kurogi, T. Kijima, Appl. Catal. B 35 (2001) 107–116. A. Satsuma, M. Hashimoto, J. Shibata, H. Yoshida, T. Hattori, Chem. Commun. 143 (2003) 1698–1699. J. Shibata, M. Hashimoto, K. Shimizu, H. Yoshida, T. Hattori, A. Satsuma, J. Phys. Chem. B 108 (2004) 18327–18335. T. Nanba, C. Kohno, S. Masukawa, J. Uchisawa, N. Nakayama, A. Obuchi, App. Catal. B 46 (2003) 353–364. T. Nanba, K. Sugawara, S. Masukawa, J. Uchisawa, A. Obuchi, Top. Catal. 42–43 (2007) 129–133. M. Saito, M. Itoh, J. Iwamoto, K. Machida, Chem. Lett. 37 (2008) 1226–1227. P. Wu, L. Li, Q. Yu, G. Wu, N. Guan, Catal. Today 158 (2010) 228–234. Q. Yu, M. Richterb, F. Konga, L. Li, G. Wu, N. Guan, Catal. Today 158 (2010) 452–458. M. Itoh, K. Motoki, M. Saito, J. Iwamoto, K. Machida, Bull. Chem. Soc. Jpn. 82 (2009) 1197–1202. C.N. Costa, V.N. Stathopoulos, V.C. Belessi, A.M. Efstathiou, J. Catal. 197 (2001) 350–364. C.N. Costa, P.G. Savva, C. Andronikou, P.S. Lambrou, K. Polychronopoulou, V.C. Belessi, V.N. Stathopoulos, P.J. Pomonis, A.M. Efstathiou, J. Catal. 209 (2002) 456–471. C.N. Costa, A.M. Efstathiou, J. Phys. Chem. 108 (2004) 2620–2630. C.N. Costa, A.M. Efstathiou, Environ. Chem. Lett. 2 (2004) 55–58. C.N. Costa, P.G. Savva, J.G. Fierro, A.M. Efstathiou, Appl. Catal. B 75 (2007) 147–156. C.N. Costa, A.M. Efstathiou, J. Phys. Chem. C 111 (2007) 3010. P.G. Savva, A.M. Efstathiou, J. Catal. 257 (2008) 324–333. A. Ueda, T. Nakao, M. Azuma, T. Kobayashi, Catal. Today 45 (1998) 135–138. B. Wen, Fuel 81 (2002) 1841–1846. G. Qi, R.T. Yang, L.T. Thompson, Appl. Catal. A 259 (2004) 261–267. G. Qi, R.T. Yang, F.C. Rinaldi, J. Catal. 237 (2006) 381–392. G.L. Chiarello, D. Ferri, J. Grunwaldt, L. Forni, A. Baiker, J. Catal. 252 (2007) 137–147. S. Furfori, N. Russo, D. Fino, G. Saracco, V. Specchia, Chem. Eng. Sci. 65 (2010) 120–127. L. Li, F. Zhang, N. Guan, E. Schreier, M. Richter, Catal. Commun. 9 (2008) 1827–1832. T. Yoshinari, K. Sato, M. Haneda, Y. Kintaichi, H. Hamada, Catal. Commun. 2 (2001) 155–158. T. Yoshinari, K. Sato, M. Haneda, Y. Kintaichi, H. Hamada, Appl. Catal. B. 41 (2002) 157–169. T. Yoshinari, K. Sato, M. Haneda, Y. Kintaichi, H. Hamada, J. Jpn. Petrol. Inst. 46 (2003) 264–271.

13

[46] H. Hamada, Y. Kintaichi, M. Haneda, H. Kudo, Y. Nagao, T. Yoshinari, K. Sato, Trans. Mat. Res. Soc. Jpn. 29 (2004) 2171–2174. [47] Y. Hasegawa, M. Haneda, Y. Kintaichi, H. Hamada, Appl. Catal. B 60 (2005) 41–47. [48] S.J. Tauster, L.L. Murrell, J. Catal. 41 (1976) 192–195. [49] M. Ogura, A. Kawamura, M. Matsukata, E. Kikuchi, Chem. Lett. 29 (2000) 146–147. [50] M. Haneda, T. Yoshinari, K. Sato, Y. Kintaichi, H. Hamada, Chem. Commun. (2003) 2814–2815. [51] M. Haneda, Pusparatu, Y. Kintaichi, I. Nakamura, M. Sasaki, T. Fujitani, H. Hamada, J. Catal. 229 (2005) 197–205. [52] A. Wang, L. Ma, Y. Cong, T. Zhang, D. Liang, Appl. Catal. B 40 (2003) 319–329. [53] M. Shimokawabe, N. Umeda, Chem. Lett. 33 (2004) 534. [54] H. Inomata, M. Shimokawabe, A. Kuwana, M. Arai, Appl. Catal. B 84 (2008) 783–789. [55] A. Abu-Jrai, A. Tsolakis, Int. J. Hydrogen Energy 32 (2007) 2073–2080. [56] A. Abu-Jrai, A. Tsolakis, A. Megaritis, Int. J. Hydrogen Energy 32 (2007) 3565–3571. [57] M.-S. Li, K. Seshan, J. Lefferts, Chin. J. Chem. 25 (2007) 435–438. [58] J.A. Wang, A. Cuan, J. Salmones, N. Nava, S. Castillo, M. Morán-Pineda, F. Rojas, Appl. Surf. Sci. 230 (2004) 94–105. [59] H. Zhu, J.-R. Kim, S.-K. Ihm, React. Kinet. Catal. Lett. 97 (2009) 207–215. ˜ A. Toledo, S. Castillo, M. Morán[60] L.F. Chen, G. González, J.A. Wang, L.E. Norena, Pineda, Appl. Surf. Sci. 243 (2005) 319–328. [61] T. Nakatsuji, T. Yamaguchi, N. Sato, H. Ohno, Appl. Catal. B 85 (2008) 61–70. [62] T. Yamamoto, T. Tanaka, R. Kuma, S. Suzuki, F. Amano, Y. Shimooka, Y. Kohno, T. Funabiki, S. Yoshida, Phys. Chem. Chem. Phys. 4 (2002) 2449–2458. [63] T. Yamamoto, T. Tanaka, S. Suzuki, R. Kuma, K. Teramura, Y. Kou, T. Funabiki, S. Yoshida, Top. Catal. 18 (2002) 113–118. [64] P.M. Sreekanth, P.G. Smirniotis, Catal. Lett. 122 (2008) 37–42. [65] J.C.L. Cornish, N.R. Avery, Surf. Sci. 235 (1990) 209–216. [66] S. Khatua, G. Held, D.A. King, Surf. Sci. 586 (2005) 1–14. [67] I. Nakamura, K. Suzuki, A. Takahashi, M. Haneda, H. Hamada, T. Fujitani, J. Vac. Sci. Technol. A 25 (2007) 1143–1146. [68] W. Chen, T.E. Madey, A.L. Stottlemyer, J.G. Chen, P. Kaghazchi, T. Jacob, J. Phys. Chem. C 112 (2008) 19113–19120. [69] K.J. Lyons, J. Xie, W.J. Mitchell, W.H. Weinberg, Surf. Sci. 325 (1995) 85–92. [70] U. Burghaus, J. Ding, W.H. Weinberg, Surf. Sci. 384 (1997) L869–L874. [71] C. Tang, S. Zou, M.W. Severson, M.J. Weaver, J. Phys. Chem. B 102 (1998) 8796–8806. [72] S. Zou, M.J. Weaver, Surf. Sci. 446 (2000) L95–L100. [73] T. Fujitani, I. Nakamura, Y. Kobayashi, A. Takahashi, M. Haneda, H. Hamada, J. Phys. Chem. B 109 (2005) 17603–17607. [74] T. Fujitani, I. Nakamura, A. Takahashi, M. Haneda, H. Hamada, J. Catal. 253 (2008) 139–147. [75] I. Nakamura, Y. Kobayashi, H. Hamada, T. Fujitani, Surf. Sci. 600 (2006) 3235–3242. [76] T. Fujitani, I. Nakamura, J. Vac. Soc. Jpn. 52 (2009) 61–66. [77] T. Fujitani, I. Nakamura, Catal. Surv. Asia 13 (2009) 22–29. [78] W. Chen, Q. Shen, R.A. Bartunski, P. Kaghazchi, T. Jacob, ChemPhysChem 11 (2011) 2515–2520. [79] M. Haneda, T. Fujitani, H. Hamada, J. Mol. Catal. A 256 (2006) 143–148. [80] M. Shimokawabe, M. Niitsu, H. Inomata, N. Iwasa, M. Arai, Chem. Lett. 34 (2005) 1426–1427. [81] H. Inomata, M. Shimokawabe, M. Arai, Appl. Catal. A 332 (2007) 146–152. [82] T. Nanba, S. Shinohara, J. Uchisawa, S. Masukawa, A. Ohi, A. Obuchi, Chem. Lett. 35 (2006) 450–451. [83] T. Nanba, S. Shinohara, S. Masukawa, J. Uchisawa, A. Ohi, A. Obuchi, Appl. Catal. B 84 (2008) 420–425. [84] A. Takahashi, I. Nakamura, M. Haneda, T. Fujitani, H. Hamada, Catal. Lett. 112 (2006) 133–138. [85] M. Haneda, N. Aoki, K. Arimitsu, H. Hamada, Bull. Chem. Soc. Jpn. 82 (2009) 1023–1029. [86] T. Tamai, M. Haneda, T. Fujitani, H. Hamada, Catal. Commun. 8 (2007) 885–888. [87] M. Haneda, H. Kudo, Y. Nagao, T. Fujitani, H. Hamada, Catal. Commun. 7 (2006) 423–426. [88] A. Takahashi, T. Fujitani, I. Nakamura, Y. Katsuta, M. Haneda, H. Hamada, Chem. Lett. 35 (2006) 420–421. [89] T. Nanba, K. Wada, S. Masukawa, J. Uchisawa, A. Obuchi, Appl. Catal. A 380 (2010) 66–71. [90] M. Haneda, K. Chiba, A. Takahashi, M. Sasaki, T. Fujitani, H. Hamada, Catal. Lett. 118 (2007) 159–164. [91] M. Sasaki, A. Takahashi, M. Haneda, T. Fujitani, H. Hamada, J. Jpn. Petrol. Inst. 50 (2007) 94–101. [92] M. Sasaki, T. Mita, T. Tsujimura, T. Nanba, A. Obuchi, M. Haneda, H. Hamada, J. Jpn. Petrol. Inst. 51 (2008) 359–360. [93] M. Sasaki, A. Sultana, M. Haneda, H. Hamada, Top. Catal. 52 (2009) 1803–1807. [94] T. Fujitani, I. Nakamura, Y. Kobayashi, A. Takahashi, M. Haneda, H. Hamada, Surf. Sci. 601 (2007) 1615–1622. [95] M. Haneda, H. Hamada, Chem. Lett. 37 (2008) 830–831. [96] M. Haneda, H. Hamada, J. Catal. 273 (2010) 39–49.