`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 1 of 6
`
`
`
`1512 different Si/A1 ratios for the NO reduction under a simulated lean NOx condition has been systematically reported. In addition, changes in the physicochemical properties of the catalysts induced by hydrothermal aging are also investigated to understand the catalyst deactivation behavior. 2. EXPERIMENTAL A MOR type natural zeolite (NZ), mined from Youngil, Korea [6], and a synthetic MOR (Zeolon 900Na, PQ Corp.) were employed to obtain CuNZA and CuHM, respectively. Prior to Cu 2+ ion exchange, these two MOR materials were also dealuminated by acid treatment and steaming to increase Si/A1 ratio in the zeolite. Five CuZSM-5 catalysts with Si/A1 ratios ranging from 14 to 95 were prepared using the corresponding HZSM-5 purchased from Tosoh Corp. All Cu2+-exchanged zeolite catalysts were prepared by conventional liquid-phase and solid-state ion exchange methods that are described elsewhere [4,5]. Chemical analysis for Si, A1, Cu in catalysts was performed by AA. The degree of dealumination of the zeolites was confirmed by 29Si and 27A1 MAS NMR. The physicochemical properties of zeolite catalysts employed in this study are listed in Table 1. A simulated lean-bum condition containing 1,200 ppm NO, 1,600 ppm C3H6, 3.2% 02, 3,000 ppm CO, 1,000 ppm H 2, 10% CO2, 10% H20 and He (balance) was employed to examine the hydrothermal stability of the catalysts. The reaction products were analyzed by on-line gas chromatograph (Hewlett Packard 5890 Series II). The hydrothermally aged catalysts were characterized by XRD, ESR, XANES and BET measurements to investigate the cause of the catalyst deactivation. Table 1. Physicochemical properties of zeolite catalysts employed in the present study Catalyst Cu content Si/A1 Cu/A1 Surface area (wt.%) (m2/g) CuHM 1 2.02 6 0.16 368 CuHM2 4.20 5 0.30 CuHM3 2.55 12 0.34 434 CuHM4 1.03 12 0.14 CuHM5 1.73 22 0.45 450 CuNZA 1 4.37 4 0.25 179 CuNZA2 1.84 10 0.24 232 CuNZA3 1.75 14 0.28 CuNZA4 1.64 19 0.31 128 CuZSM-5-1 2.49 14 0.53 350 CuZSM-5-2 2.90 26 0.81 344 CuZSM-5-3 1.43 27 0.50 CuZSM-5-4 1.20 34 0.53 400 CuZSM-5-5 0.44 95 0.52 330
`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 2 of 6
`
`
`
`1513 3. RESULTS AND DISCUSSION 3.1 Hydrothermal stability of MOR and MFI type zeolite catalysts Figure 1 shows the hydrothermal stability of CuNZA2, CuHM1 and CuZSM-5-2 catalysts for NO reduction under the lean NOx condition. When the catalysts were aged at 800~ with 10% H20 for 6 h, the CuHM1 catalyst exhibited a deNOx efficiency of less than 5% at 450 ~ while 15 and 40% of NO conversion were still observed for CuNZA2 and CuZSM-5-2, respectively. Thus, the CuZSM-5-2 catalyst was found to exhibit the highest hydrothermal stability for NO reduction among these three types of zeolite catalysts. Another important observation is that the activity maintenance of the catalysts was proportional to the Si/A1 ratio of the catalyst. This clearly shows that the hydrothermal stability of zeolite catalysts for NO reduction under the lean NOx condition primarily depends on the Si/A1 ratio of the catalyst, despite the differences in the zeolite structure. To further elucidate the role of Si/A1 ratio of the catalysts for the hydrothermal stability, MOR type zeolite catalysts with different Si/A1 ratio were examined. Figure 2 shows that deNO~ activity of CuNZA catalyst was significantly improved by dealumination. Especially, the CuNZA4 catalyst with a Si/A1 ratio of 19 still showed more than 35% of NO conversion at 550~ even after the aging at 800~ with 10% H20 for 24 h.
`Reaction Temperature (oC) Figure 1. Hydrothermal stability of CuNZA2 (O,O), CuHM1 (1,[:]) and CuZSM-5-2 (A,A) catalysts. Closed and open symbols indicate fresh catalysts and hydrothermally aged ones at 800~ for 6 h with 10% H20, respectively. 100
`Reaction Temperature (~ Figure 2. Hydrothermal stability of the dealuminated CuNZA catalsyts (cid:12)9 CuNZA2 (O), CuNZA3 (I) and CuNZA4 (A). These catalysts were hydrothermally aged at 800~ for 24 h with 10% HzO. 3.2 Effect of Si/Ai ratio of MOR and MFI type zeolite catalysts To understand the origin of the significant enhancement of the deNOx activity over the dealuminated CuHM catalysts, the activities of CuHM catalysts with different Si/A1 and Cu/A1 ratios have been examined and are given in Figure 3. Note that the Si/A1 ratio of
`
`t ' c i t , O, = ...... 1
`
`0~"
`
`8O o z
`
`q..
`o
`t-
`o
`"~
`o >
`c
`o
`O
`
`40
`
`20
`
`300
`
`400
`
`500
`
`600
`
`700
`
`300
`
`400
`
`500
`
`600
`
`700
`
`100
`
`sO
`
`Z
`
`604
`
`l
`
`~=
`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 3 of 6
`
`o 1
`
`
`100
`
`. ao 8 =o 0 30G 404) 500 84)0 7'00 Reaction Temperature (~ Figure 3. Effect of Si/A1 and Cu/A1 ratms of the dealuminated CuHM catal.ysts on their NO removal actw]ty : CuHM1 (O), CuHM2 (1), CuHM3 ((cid:12)9 and CuHM4 ((cid:12)9 loo ] 8.0 i "~ 4o i i 0 .... ] 300 400 500 600 700 Reaction Temperature (~ Figure 4. Effect of Si/A1 ratio of the CuZSM-5 catalysts on their NO removal activity : CuZSM-5-1 (O), CuZSM-5-3 (ll), CuZSM-5-4 ((cid:12)9 and CuZSM-5-5 ((cid:12)9 CuHM1 is quite similar to that of CuHM2, while these two catalysts contain Cu/A1 ratios of 0.16 and 0.30, respectively. However, the conversion of NO was found to be not distinctive for the catalysts, even though the reaction temperature exhibiting the maximum NO removal activity of the catalysts is slightly shifted. For CuHM1 and CuHM4 catalysts with similar Cu/A1 ratios, on the contrary, a significant improvement in the deNOx activity was observed after dealumination. A similar result was also observed from the distinctive deNOx activities of CuHM2 and CuHM3. Therefore, the improvement in the NO removal activity of CuHM catalysts upon dealumination cannot be attributed to the difference in Cu/AI ratio of the catalyst but the increase in Si/A1 ratio. This speculation can be further supported by the catalytic results obtained from CuNZA and CuZSM-5 with different Si/A1 ratios. Figure 4 shows NO removal activity of CuZSM-5 with similar Cu/A1 ratios. There may be an optimal S1/A1 ratio for CuZSM-5 catalyst to remove NO by hydrocarbons. Therefore, it is most likely that the Si/A1 ratio of the zeolite catalyst is still one of the most critical properties determining NO removal activity of the catalyst for this reaction system. 3.3 Characterization of the zeolite catalysts When CuZSM-5-2 catalyst was hydrothermally aged at 900~ structural collapse of the catalyst was clearly observed by a comparison of the intensity of the XRD peak of the catalyst with respect to the aging duration shown in Figure 5. As the aging time and temperature increase, the structure of the catalyst was progressively destroyed. Thus, the intensity of the characteristic diffraction peaks of the catalyst aged at 900~ for 24 h was significantly reduced as depicted in Figure 5(0. This might well reflect not only the loss of the deNOx activity of the catalyst but also the reduction of the catalyst surface area by hydrothermal aging. A remarkable decrease of XRD peak intensity by hydrothermal aging
`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 4 of 6
`
`1514
`
`
`1515 was also observed for CuHM1 and CuNZA2 catalysts. For the dealuminated catalysts, the degree of structural destruction of the catalysts examined by XRD well agreed with the rate of the activity maintenance of the catalyst. This again supports the speculation that the Si/AI ratio of the catalyst plays an important role for the destruction of zeolite structure by hydrothermal aging as expected.
`
`~
`
`
`
`~
`
`
`
`~
`L
`
`'
`
`L
`
`.
`
`,
`
`30
`
`20
`
`r .-.,.---
`
`(e) --
`
`7'0
`
`60
`
`50
`
`~_t"L,.
`
`10
`
`;0
`20
`Figure 5. X-ray powder diffraction pattern of
`
`CuZSM-5-2 catalysts : (a) fresh, (b) aged at 700~ without H20 for 2 h, (c) aged at 800~ with 10% H,O for 4 h, (d) aged at 800~ with 10% H20 for 6 h,-(e) aged at 800~ with 10% H20 for 24 h and (f) aged at 900~ with 10% H20 for 24 h. The destruction of the catalyst structure upon the hydrothermal aging may alter the ionic state of Cu 2+ ions on the catalyst surface. (a) Figure 6. Cu 2§ ESR spectra of CuZSM- 5-2 catalysts : (a) fresh, (b) aged at 800~ with 10% H20 for 4 h, (c) aged at 800~ with 10% H20 for 24 h and (d) aged at 900~ with H20 for 24 h. An alteration of the chemical environment of Cu 2+ ions on the hydrothermally aged catalyst surface was examined by ESR as shown in Figure 6. The low field hyperfine structure of CuZSM-5-2 catalyst was notably reduced when aged at 900~ with 10% H20 for 24 h. This suggests that Cu 2§ ions on the catalyst surface were probably transferred to copper oxide species by the sintering of the metal ions through the structural collapse of the catalyst. Such an alteration of Cu 2§ ions was also observed for CuHM1 and CuNZA2 catalysts after hydrothermal aging. Therefore, it is clear that the loss of Cu 2+ ions on the catalyst surface occurred upon hydrothermal aging and thereby caused a significant loss of NO removal activity of the catalysts. Additionally, the alteration of Cu 2+ ions to the copper oxides has been confirmed by X-ray absorption near edge structure (XANES). As shown in Figures 7 and 8, isolated Cu 2§ ions were transformed to Cu20 after hydrothermal aging, as already examined by XRD and ESR studies. Therefore, hydrothermal aging caused the structural destruction of zeolite and the decrease of surface area, and then Cu 2+ ions transformed to copper oxides that may not be active reaction sites for the present reaction system. Based upon the peak intensity of Cu20, the zeolite with a higher Si/A1 ratio results in less formation of Cu20 on the catalyst surface.
`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 5 of 6
`
`(b)
`80
`
`
`o
`
`8960
`
`8980
`
`9040
`
`9020
`9000
`(eV)
`
`Figure 7. XANES spectra of CuHM catalysts : (a) CuHMl(fresh), (b) CuHM3(fresh), (c) CuHM5(fresh), (d) CuHM1 (aged), (e) CuHM3 (aged), (f) CuHM5(aged) and (g) Cu20. These catalysts were hydrothermally aged at 800~ for 24 h with 10% H,O. A 5 \ ni < i | | 8960 8980 9000 9020 9040 Energy (eV) Figure 8. XANES spectra of CuNZA catalysts : (a) CuNZA2(fresh), (b) CuNZA3(fresh), (c) CuNZA5(fresh), (d) CuNZA2(aged), (e) CuNZA3(aged), (f) CuNZA5(aged) ) and (g) Cu20. These catalysts were hydrothermally aged at 800~ for 24 h with 10% H20. Therefore, it can be concluded that the Si/A1 ratio of the zeolite catalyst is one of the most important characteristics for the hydrothermal stability of zeoilte catalyst. 4. CONCLUSIONS Hydrothermal stability of MOR and MFI type zeolite catalysts for NO reduction by hydrocarbons has been observed to become stronger as the Si/A1 ratio of the catalyst increases. It is evident that the Si/A1 ratio of the zeolite catalyst is the most critical characteristic determining the hydrothermal stability of the catalysts for this reaction system. The hydrothermal stability of the catalysts can be improved by the optimization of the Si/AI ratio of the zeolite catalysts. A significant loss in the deNOx activity is mainly due to the alteration of Cu 2+ ions on the catalyst surface through the collapse of the zeolite structure resulting in a decrease of the catalyst surface area. REFERENCES 1. M. Iwamoto and N. Mizuno, J. Auto. Eng., 207 (1993) 23. 2. Y.Li, P. Battavio and J.N. Armor, J. Catal., 142 (1993) 561. 3. K.C.C. Kharas, H.J. Robota and D.J. Liu, Appl. Catal. B, 2 (1993) 225. 4. M.H. Kim, I.-S. Nam and Y.G. Kim, Appl. Catal. B, 6 (1995) 297. 5. M.H. Kim, I.-S. Nam and Y.G. Kim, Appl. Catal. B, 12 (1997) 125. 6. M.H. Kim, U.-C. Hwang, I.-S. Nam and Y.G. Kim, Catal. Today, 44 (1998) 57.
`
`Umicore AG & Co. KG
`Exhibit 1014
`Page 6 of 6
`
`8
`1516
`Energy