Catalytic Decomposition of Nitric Oxide over Promoted Copper-Ion

Chapter 2

Catalytic Decomposition of Nitric Oxide over Promoted Copper-Ion-Exchanged ZSM-5 Zeolites Downloaded by EAST CAROLINA UNIV on September 17, 2015 | http://pubs.acs.org Publication Date: February 23, 1994 | doi: 10.1021/bk-1994-0552.ch002

Yanping Zhang and Maria Flytzani-Stephanopoulos Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge,MA02139

Alkaline earth and transition metal cocation effects are reported in this paper for Cu ion-exchanged ZSM-5 zeolites used for the catalytic decomposition of nitric oxide in oxygen -rich gases. The effect, manifested under a specific mode of ion exchange, enhances the catalytic activity at high temperatures (450-600°C) and appears to be due to stabilization of the active copper sites. The conversion of NO to N is invariant to oxygen in the high temperature region for dilute NO gases. The coexistence of rare earth ions with Cu ions in ZSM-5 produced a markedly different effect promoting the activity of copper ions for the NO decomposition at low temperatures (300-400°C). 2

The direct catalytic decomposition of nitric oxide to nitrogen and oxygen in oxygen-rich post-combustion gas streams would greatly benefit the economics of post-combustion NOx control in power plants, industrial boilers and engine systems. The initial report by Iwamoto (1) that Cu ionexchanged ZSM-5 (Cu-Z) zeolite has stable steady-state activity for the direct decomposition of nitric oxide has been followed by many recent studies (2-9) of this catalyst system addressing pertinent issues potentially affecting its activity including copper exchange level, S i / A l ratio, oxygen effect, poisoning by SO2, H2O effect, etc. The conversion of N O to N2 over the Cu-Z materials is not a linear function of the copper ion-exchange level. Rather, very low activity exists for up to 40% exchange level, above which conversion increases rapidly with the exchange level (2). Over-exchanged( >100%) Cu-Z are active either

0097-6156/94/0552-0007$08.00/0 © 1994 American Chemical Society In Environmental Catalysis; Armor, J.; ACS Symposium Series; American Chemical Society: Washington, DC, 1994.

ENVIRONMENTAL CATALYSIS

Downloaded by EAST CAROLINA UNIV on September 17, 2015 | http://pubs.acs.org Publication Date: February 23, 1994 | doi: 10.1021/bk-1994-0552.ch002

8

because copper ions adequately populate sites accessible to N O ( i.e., in the 10-membered rings of the ZSM-5) or because catalytic activity only emerges if enough adjacent copper active sites exist (6). In recent work, Kagawa, et al (8) reported that the incorporation of cocations into Cu-Z promoted the catalytic activity of Cu ion for the direct N O decomposition in 02-free gas streams at temperatures above 450°C. Transition metals or alkaline earths were equally effective cocations in Cu-Z at 550°C. The mode of ion exchange was important, i.e., the cocation had to be exchanged first followed by copper ion exchange, for the effect to show. The positive cocation effect was more pronounced for low Cu ion exchange levels. The effect of oxygen on the conversion of N O to N2 was not examined in that work. Oxygen has been reported to inhibit the reaction over unpromoted Cu-Z catalysts, but the inhibition decreases with temperature (7). Iwamoto, et al (6) have reported that the conversion of N O to N2 in the presence of oxygen is a function of both the Cu-exchange level and the ratio of PNO/P02In this paper we examine the cocation effect on the N O decomposition reaction in oxygen-containing gases over M / C u - Z catalysts, where M is an alkaline earth or transition metal ion. The importance of the preparation method is discussed in terms of active ( Cu ion) site stabilization. Rare earth metal-modified Cu-Z catalysts were also prepared and tested in this work both in oxygen-free and oxygen-containing gases. Experimental Catalyst Synthesis and Characterization. Catalysts were prepared by incorporating metal cations into ZSM-5 zeolite supports according to ion exchange procedures widely used in preparation of metal/zeolite catalysts. The starting materials were the Na+ form of ZSM-5 zeolites with S i / A l ratio of 21.5 synthesized by the Davison Chemical Division of W.R. Grace Co. In catalysts containing copper and a cocation, the ZSM-5 zeolites were first ion-exchanged with the cocation in nitrate form in dilute aqueous solution with concentration of 0.007M. The exchanges were made either at room temperature for 10 hours or at 85°C for 2 hours. After filtration, the metal ion-exchanged ZSM-5 zeolites were dried at 100°C for 10 hours, and some of them were further calcined in a muffle furnace in air at 500°C for 2 hours. The reason for calcining the catalysts was to stabilize cations in the zeolite. The catalysts were further ion-exchanged with Cu^+ in an aqueous solution of cupric acetate of concentration 0.007M at room temperature overnight. This was repeated several times, depending on the desired Cu exchange levels. Finally, the catalysts were washed with deionized water at room temperature and dried at 100°C overnight. For simplicity, a catalyst with intermittent air calcination of cocation exchanged ZSM-5 is designated as a precalcined M / C u - Z catalyst in the text.

In Environmental Catalysis; Armor, J.; ACS Symposium Series; American Chemical Society: Washington, DC, 1994.

2.

ZHANG & FLYTZANI-STEPHANOPOULOS

Decomposition of Nitric Oxide

9

The mode of exchange described here was the evolution of several different preparation methods. A summary of observations made during this process when using Mg2+ as the cocation is as follows: a) exchanging copper ions first or co-exchanging copper and magnesium ions for N a in the ZSM-5 did not succeed in high-exchange levels of Mg2+ in the zeolite; b) even when Mg + was exchanged first followed by copper ion exchange, we would observe loss of Mg2+ in the solution. Both a) and b) are the result of a more favorable exchange equilibrium for C u ^ / N a (10); c) when the M g ion solution was heated at 85°C for 2 hours, higher levels of exchange were obtained and better stability in subsequent room temperature copper ion exchanges; d) air calcination of Mg2 -exchanged zeolites at 500°C for 2 hours was very effective in keeping the M g exchange high even after subsequent copper ion exchanges. This procedure was followed in preparing two of the M g / C u - Z catalysts as well as the Sr, N i , Pd, Ce and La/Cu-Z catalysts shown in Table I. The elemental analyses were performed by Inductively Coupled Plasma Emission Spectrometry (ICP, Perkin-Elmer Plasma 40) after catalyst samples were dissolved in HF(48%). It is noted that a small amount of Si and A l was extracted from the zeolite during the ion exchange procedures. A list of catalyst syntheses and properties is shown in Table I. In the text, the catalysts are identified in the following way: cocation type(percent exchange level)/Cu(percent exchange level)-Z, for example, Mg(34)/Cu(86)Z, where 100% ion exchange level is defined as one C u (or Mg +) replacing two N a ( or neutralizing two Al") and the atomic ratio Cu (or Mg)/Al=0.5. +

2

+

+


2 +

+

2

2 +

+

2

+

Conversion Measurements and Kinetic Studies. Conversion and kinetic studies were performed in laboratory-scale packed-bed reactors, consisting of a 60cm long quartz tube with I.D. equal to 1.1 cm( for conversion measurements) or 0.4 cm (for kinetic studies). A porous quartz fritted disk was placed in the middle of the tube to support the catalyst bed. The reactor was placed in a temperature-programmed furnace that was electrically heated and controlled by a temperature controller ( Tetrahedron: Model Wizard). Three mass flow controllers were used to measure flow rates of NO+He, 02+He and He. Certified standard helium and gas mixtures were used from Matheson and Airco. A gas chromatograph ( Hewlett Packard: Model 5890) with a thermal conductivity detector, and a 5A molecular sieve column of 1/8 in. O.D. by 6 ft. long, was used to measure concentrations of nitrogen, oxygen and nitric oxide. A NO-NOx analyzer (Thermo Electron: Model 14A) was also used to measure low N O / N O x concentrations, ranging from O.OOlppm to 2,500 ppm. A n amount of 0.5-1.0 g of catalyst was placed in reactor for conversion measurements, and 30-35 mg for kinetic studies. The catalyst packing density in the reactor was about 0.5 g/cc. Total flowrates of the feed gas were 30-90 cc/min. The contact time,


10


W / F , defined as the ratio of catalyst weight in the reactor to the total flowrate of the feed gas was 0.03- l g s/cc( STP). The total gas pressure in the reactor was 1.5 atm during conversion measurements and about 2 atm in the kinetic studies. N O concentrations varied from 0.2%- 4% in the feed gas stream, O2 from 0%-5%, balance He. The 02+He stream was heated to 150°C before it was mixed with NO+He at the inlet of the reactor to avoid N O 2 formation (4), A l l measurements were made after steady state had been reached. Downloaded by EAST CAROLINA UNIV on September 17, 2015 | http://pubs.acs.org Publication Date: February 23, 1994 | doi: 10.1021/bk-1994-0552.ch002

Results and Discussion Nitric Oxide Decomposition in the Absence of Oxygen. The Cu-Z and M g / C u - Z catalysts shown in Table I were evaluated in a gas containing 2% NO-He, at contact time of l g s/cc (STP) over the temperature range of 350 600°C. Control experiments with Mg-Z materials verified that the activity of M g / C u - Z was exclusively due to Cu ions. Figures 1 and 2 show the N O to N 2 conversion profiles obtained for the catalysts tested under these conditions. For the same copper ion-exchange level (-70%), the Mg(52)/Cu(66)-Z catalyst shows a positive effect, i.e., higher N O to N2 conversion than the Cu(72)-Z material, in the high temperature region ( 450 -600°C ), as can be seen in Figure 1. These results are in agreement with the report by Kagawa, et al (8). While still present, this effect is not as pronounced for M g / C u - Z catalysts with Cu exchange level higher than about 110%. Within the group of the M g / C u - Z catalysts, preparation conditions were important. As mentioned in the previous section, heating of the solution during Mg2+-exchange was necessary in order to preserve the M g ions in the zeolite upon subsequent exchange with copper ion solutions. The effect of intermittent air calcination of Mg-Z material at 500°C for two hours on the catalytic activity has been evaluated. The precalcined catalyst, Mg(34)/Cu(86)-Z, gave higher N O to N2 conversion over the whole temperature range (350-600°C) than the catalyst Mg(40.4)/Cu(91.2)-Z without intermittent air calcination, as shown in Figure 2. At the present time, no consensus exists in the literature on the mechanism of N O decomposition over Cu-Z catalysts. This makes a mechanistic interpretation of the cocation effects reported here premature. However, the importance of ion exchange sequence and catalyst heat treatment on the N O decomposition activity is worth discussing in terms of active site modification on the basis of available information. In ESR studies, Kucherov, et al ( 11,12) have identified two types of isolated C u ions: one in a five-coordinated square pyramidal configuration, the other in a four-coordinated square planar. At low Cu exchange level, the fivecoordinated C u ions were preferably formed (12). Shelef (13) proposed the square planar copper ions as the active sites of Cu-Z for N O decomposition. This explains the negligible activity of Cu-Z catalysts with 2 +

2 +


Decomposition of Nitric Onde

2. ZHANG & FLYTZANI-STEPHANOPOULOS

Table I. Summary of Catalyst Syntheses and Properties


Catalysts Parent zeolite Cu-Z

Si/Al a

21.5 20.3

Cu-Z

19.9

Mg/Cu-Z^

18.0

Mg/Cu-Zd,e

17.1

Mg/Cu-ZC/β

19.5

Ni/Cu-ZC>e

20.8

Sr/Cu-Z^e

20.8

Pd/Cu-Z^

20.7

Ce/Cu-ZC/e

19.5

La/Cu-ZC/e

19.4

Cu/Al

0.705 (141%) 0.36 (72%) 0.456 (91%) 0.430 (86%) 0.33 (66%) 0.478 (96%) 0.524 (105%) 0.45 (90%) 0.596 (119%) 0.627 (123%)

b

b

Cocation/Al N a / A l

-

1.0 ~0

b

Cu exchange thrice, RT

0.25 (25%) ~0

once, RT

-0

twice, RT

~0

twice, RT

~0

twice, RT

0.194 (39%) 0.44 (88%)

-0

twice, RT

~0

twice, RT

0.036 (11%) 0.06 (18%)

~0

twice, RT

-0

twice, RT

_

0.202 (40%) 0.170 (34%) 0.26 (52%) -

twice, RT

a. ZSM-5: SMR-2670-1191, as received from Davison Div.,W.R. Grace. Co. b. The values in parentheses are ion exchange levels, on the basis of A l content as measured by ICP. c. Cocations exchanged once with Na/ZSM-5 at 85°C for 2 hours. d. Cocation exchanged once with Na/ZSM-5 overnight at RT. e. The cocation-exchanged ZSM-5 catalysts were dried in air at 100°C overnight, and calcined at 500°C for 2 hours.


11



350

400

450

500

550

650

Reaction Temperature (°C) Fig. 1. Comparison of N O conversion to N2 over (a)Mg(52)/Cu(66)Z; (b) Cu(72)-Z at 2%NO and W/F=lg s/cc (STP).

CM

2 2

G Ο

• ιΜ

2 Ο

U Ο Ζ 350

400 450 Reaction Temperature ( Ό

650

Fig. 2. Effect of intermittent Mg-Z calcination on Mg/Cu-Z catalyst activity for N O conversion to N2 at 2%NO and W/F=lg s/cc (STP); (a) calcined Mg(34)/Cu(86)-Z; (b) non-calcined Mg(40.4)/Cu(91.2)-Z.


2.



low C u ion-exchange level ( c ο

υ

ι 600

650

Reaction Temperature (°C) Fig. 5. Effect of O2 and N O concentrations on the Mg(34)/Cu(86)-Z activity for N O decomposition at W/F=lg s/cc (STP); (a) 2%NO-0%O2; Ob) 2%ΝΟ-5%θ2; (c) 0.2%NO-0%O2; (d) 0.2%NO-5%O2.

Ζ

fi 2 8 Ο

I ° .001 1.60 1000/Τ (Κ) Fig. 6. Arrhenius plots for N O decomposition over (a) Cu(141)-Z; (b) Mg(34)/Cu(86)-Z at 4%NO and 1.7 atm of total gas pressure.



2.



17

the corresponding activation energy. As can be seen in Figure 6, the apparent activation energy for N O decomposition over the Mg(34)/Cu(86)Z catalyst is 12.3 Kcal/mole in the low temperature region, changing over to a negative value of -8.2 Kcal/mole in the high temperature region (>500°C). Over the Cu(141)-Z material, similar values of apparent activation energies, 12.6 Kcal/mole for the low temperature region and -7.6 Kcal/mole for the high temperature region, were obtained. A changing reaction mechanism or loss of copper active sites at high temperature may explain the observed reaction rate maximum. A similar study in the presence of oxygen is warranted, based on the conversion plots discussed in the previous section. Detailed kinetic studies are presently underway in our laboratory with over-exchanged Cu-Z and Mg/Cu-Z catalysts. One interesting series of tests is reported here as it provides support to the data of Figure 5. A gas containing 0.2% N O in He and oxygen in the range of 0-5% was reacted over the M g (34)/Cu (86)-Z catalyst at contact time of 0.03g s/cc( STP) and reaction temperature of 600°C. Conversion below 30% was used to measure TOF. The results, plotted in Figure 7, show the specific catalytic activity to be invariant to oxygen, i.e., a reaction order of zero in O2 was obtained. Rare Earth Metal Cation Modified Cu-ZSM-5. Two rare earth metal modified Cu-Z catalysts, namely Ce(ll)/Cu(119)-Z and La(18)/Cu(123)-Z shown in Table I, were evaluated and compared to Cu(141)-Z and Mg(34)/Cu(86)-Z catalysts in a gas containing 2% NO-He at contact time of l g s/cc (STP) over the temperature range of 300-600°C. Figure 8 shows the N O conversion to N2 profiles obtained for these catalysts under the above conditions. The coexistence of rare earth ions with copper in the ZSM-5 produced a markedly different effect promoting the activity of copper ions for the N O decomposition at low temperatures (300-400°C). The Ce(ll)-Z material has very low (but measurable) activity, less than 10% conversion of N O to Ν 2 , over the temperature region of 300-500°C. However, the Ce(ll)/Cu(119)-Z catalyst showed higher conversion of N O to N2 than the over exchanged Cu(141)-Z catalyst at temperatures in the range of 300400°C, and higher than the Mg(34)/Cu(86)-Z catalyst at temperatures below 450°C, as seen in Figure 8. When a 50 - 50 mixture (by weight) of Ce(ll)/Cu(119)-Z and Cu(141)-Z was tested, its catalytic activity was between those of Cu(141)-Z and Ce(ll)/Cu(119)-Z( not shown in Figure 8). Effect of Contact Time on Catalyst Activity. Contact time tests showed that the activity of Cu(141)-Z and Mg(34)/Cu(86)-Z catalysts did not decrease much at low N O concentration, oxygen-rich gas and low contact time. This is very important for practical application of the catalysts. Figure 9 shows profiles of N O conversion to N2 over the Mg(34)/Cu(86)-Z and Cu(141)-Z



.01

ë =

2 *


b ο Ο

.001H 0

•

" 1

•

1

•

1

2

•

1

3

•

•

1

4

5

1 6

Oxygen Concentration (%) Fig. 7. Effect of oxygen on the N O decomposition rate(TOF) over Mg(34)/Cu(86)-Z at 0.2%NO, 600°C , and W/F=lg s/cc (STP).

0Η ' 1 250 300

1

1

350

·

1

400

1

1

450

>

1

500

>

1

550

1

1

1

600

650

Reaction Temperature (°C) Fig. 8. Comparison of N O conversion to N 2 over (a) Ce(ll)/Cu(119)Z; (b) La(18)/Cu(123)-Z; (c) Mg(34)/Cu(86)-Z; (d) Cu(141)-Z at 2%NO and W/F=lg s/cc (STP).



2.



catalysts at 2,000ppm N O and 2.5% O2 in the feed gas and 0.05g s/cc of contact time( 36,000 hr~l S.V.). Conversions are similar to those in Figures 4 and 5 obtained at higher contact time. Figure 9 also shows that the Ce(ll)/Cu(119)-Z catalyst remained more active than Cu(141)-Z and Mg(34)/Cu(86)-Z below 400°C under these conditions. The catalyst Ce(ll)/Cu(119)-Z presented here gave higher N O conversion to N2, compared with a Ce/Cu-Z material reported in the patent literature (15) (only 11% N O conversion to N2 at 400°C under the conditions of Figure 9), which was prepared by exchanging Cu ions into the ZSM-5 first, followed by Ce ion exchange. Again, this shows that a proper mode of ion exchange is very important for a positive cocation effect. Conclusion Alkaline earth and transition metal ion-exchanged ZSM-5 zeolites further exchanged with copper ions are active catalysts for the N O decomposition reaction in the presence or absence of oxygen and over a wide range of temperatures (350 - 600°C). Proper mode of ion exchange is crucial for a positive cocation effect. A n extensive study of Mg/Cu-Z catalysts was performed in this work. The conversion and kinetic data indicate that the role of M g ions is one of stabilizing the activity of copper ions, potentially by occupying the "hidden" sites of the zeolite. For the same total copper exchange level, the

300

350

650 Reaction Temperature (°C)

Fig. 9. N O decomposition over (a) Ce(ll)/Cu(119)-Z; (b) Mg(34)/Cu(86)-Z; and (c) Cu(141)-Z at 0.2%NO, 2.5%θ2 and W/F=0.05gs/cc (STP).


19


20


M g / C u - Z catalysts show enhanced activity at high temperatures (450600°C). Both the over-exchanged Cu-Z and the Mg/Cu-Z catalysts have low sensitivity to oxygen in dilute N O gases. At 600°C, a reaction order of zero in O2 was found with the Mg(34)/Cu(86)-Z catalyst. Detailed kinetic studies with oxygen-containing gases are underway to further examine these effects. The performance of (Ce, La)/Cu-Z catalysts is markedly different, the cocation effect being promotion of the Cu ion activity. This is displayed by the N O conversion to N2 maximum shifting to lower temperatures (

Catalytic Decomposition of Nitric Oxide over Promoted Copper-Ion

Recommend Documents