Shape-Selective Reactions for Methylamine Synthesis from Methanol and Ammonia

Guczi. L.et al. (Editors), New Frontiers in Catalysis Proceedings of the 10th International Congress on Catalysis, 19-24 July, 1992, Budapest, Hungary 8 1993 Elsevier Science Publishers B.V.All rights reserved

SHAPE-SELECTIVE REACTIONS FOR METHYLAMINE SYNTHESIS FROM METHANOL AND AMMONIA

K Segawa and H. Tachibana Department of Chemistry, Faculty of Science and Technology, Sophia University, 7-1 Kioi-cho, Chiyoda-ku, Tokyo 102, Japan

Abstract Methylamine syntheses from methanol and ammonia on various zeolite catalysts have been studied. The mordenite catalysts were treated with silicon tetrachloride in sodium-form and then ion-exchanged to the acid-form; such catalysts have been shown t o suppress trimethylamine production almost completely, and t o generate enhanced yields of dimethylamine and monomethylamine. The formation of trimethylamine has been extremely retarded, because the pore openings of the catalyst are smaller than trimethylamine molecules. The adsorption data of methylamines suggest that the more bulky base molecules, such as trimethylamine, cannot penetrate into mordenite channels after treatment with silicon tetrachloride. 1. INTRODUCTION

The methylamines are impprtant chemical intermediates in the chemical industry. Solid acids catalyze the sequential reactions of methanol (CH30H) and ammonia (NH3) t o give monomethylamine (MMA), dimethylamine (DMA), and trimethylamine (TMA); the following equations apply:

These reactions generally proceed towards equilibrium, which favors TMA. A t 99.8 % conversion of methanol, the equilibrium composition is 15.1 % for MMA, 22.8 % for DMA, and 62.1 % for TMA (NH3/CH30H=1.9) a t 623 K [l]. However, the industrial market demands lower alkylated methylamines, such as MMA and DMA. Generally TMA is recycled with additional NH3 over the catalyst bed to produce additional MMA and DMA; that is an inefficient and energy intensive process. For the selective synthesis of MMA and DNLA, various zeolite catalysts have been proposed to improve the selectivity of MMA and DMA [l-41.At higher conversions, however, selectivity for DMA was not high enough. For example, the partially alkali metal exchanged mordenite showed

1274

a higher selectivity of MMA plus DMA: ca. 80 % selectivity at 95 % methanol conversion (NH&H30H=1.9, 673 K) [ll. Weigert proposed that Na-mordenite catalysts would show lower selectivity of TMA [31. However, the catalytic activity over Na-mordenite is lower than that of acid type mordenite. Shannon and coworkers found that the small-pore zeolites, H-RHO, H-ZK5, and chabazite, are not only significantly more active but have selectivities much greater than those previously reported for these other zeolites 15-91. The selectivity of MMA plus DMA was ca. 89 % at 90 % methanol conversion (NH&H30H=l.O, 598 K) [9]. We found that the mordenite catalyst which was treated with silicon tetrachloride (SiC14) in sodium-form and then ion-exchanged to hydrogen form (acidic form) shows higher selectivity and activity for MMA and DMA. The selectivity of MMA plus DMA was ca. 99 % a t 90 % methanol conversion (NHdCH30H =1.0,633 K). The sorption data of methylamines suggest that the more bulky base molecules, such as TMA, cannot penetrate into mordenite channels after treatment with SiC14. The shape selectivity for methylamines can be explained by the changes of the pore size of mordenite after treatment with SiCl4.

Preparation of catalysts. Zeolites (JRC-Z) and silica-alumina (JRC-SAL-2, Si/A1=5.3) samples were supplied by the Catalysis Society of Japan (JRC: Japan Reference Catalysts). Three different types of zeolites were studied: H-Y (JRC-Z-HY5.6,Si/A1=2.8), H-M (H-mordenite: JRC-HM20, Si/Al=9.7), and HZSM5 (JRC-Z5-25H, Si/A1=12.5). Some Na-M samples; Na-mordenite (JRCM10, Si/AI=5), were treated with Sic14 at 973 K for 3 h, followed by washing with boiling distilled water and drying at 373 K, and were then calcined at 773 K (Na-SC-M). 10 kPa of Sic14 vapor with N2 carrier gas was passed over the Na-M. Acid-type mordenites, after treatment with Sic14 (H-SC-M), were prepared by ion-exchange of Na-form (Na-SC-M) with aqueous solution of NH4N03; ion-exchanged samples were dried a t 373 K for 24 h and then calcined in the furnace at a constant temperature increase (1K min-1)from 373 K to 773 K and kept at 773 K for 5 h. Catalytic reactions. Methylamine (MA) synthesis from CH3OH and NH3 was carried out at 633 K by using a flowing reaction system under atmospheric pressures with a mixture of NH3, CHsOH, and N2 in the ratio of 1/1/31. The flow rate of mixture gas was 100 cm3 rnin-1. Prior t o the reaction, the catalyst (500 mg) was calcined a t 773 K under 02 flow for 2 h. The reaction products were analyzed by a gas chromatograph (TCD) which was equipped with a 4.6m column of PEG-400 (25 %) + KOH (2.5 %) on acid-washed Chromosorb W at 358 K [3]. The activities and selectivities were calculated in carbon base from the chromatograms. Adsorption measurements. Chemisorption measurements of base molecules were performed under vacuum conditions. A McBain-type of quartz spiral spring was hung down to the sample basket. The catalyst was evacuated at 773 K for 2 h. After evacuation, 0.7 kPa of base molecules (MMA, DMA, TMA, NH3) were admitted at 373 K for several hours until adsorption had equilibriated, then evacuated at 373 K for 12 h. The amount of chemisorption of

1275

base molecules was determined by the change in length of the spring, which was equipped with a displacement meter (Type 2U, Shinko Electronic). Hightemperature calorimetry of NH3 on zeolite catalyst was obtained by the calorimeter (HAC-450G, Tokyo Rikou). 1.5 g of sample was charged to the calorimeter, and evacuated at 673 K for 4 h. 15 pmol g-1 of NH3 were admitted doze after doze at 473 K. IR spectroscopy. A vacuum-tight IR cell with KBr windows was designed to fit an infrared spectrometer (270-30, Hitachi) and to be attached to a vacuum system (10-4 Pa). The calcined aliquots of the zeolites were pressed into platelets having a thickness of about 10 mg cm-2. These were mounted on a quartz rack which positioned them between the KBr windows (optical path length: 5 mm) of a vacuum-tight spectroscopic cell. The cell was arranged such that the zeolite wafer could be lowered into slots between the optical windows, and withdrawn upward by the action of a magnet into the heated portion for the pretreatment and adsorption of NH3 and pyridine (C5H5N). After evacuation at 773 K for lh, the zeolite sample was cooled to 423 K before adsorption of the base molecules to be studied. XPS and MASNMR spectroscopy and XRF analysis. For the determination of surface SUAl atomic ratio, x-ray photoelectron spectroscopy (XPS) measurements of the zeolite samples before or after treatment with Sic14 on an XPS spectrometer (SSX-100, Surface Science Laboratory) with a monochromatic AlKa radiation (1486.6 eV). For the determination of framework Si/Al atomic ratio of zeolite sample, the MASNMR has been employed for zeolite samples. The 29Si-MASNMR spectra were obtained at 53.7 MHz on a Fourier transform pulsed NMR spectrometer (JNM-GX270, JEOL) which was equipped with a CP/MAS unit (NM-GSH27MU JEOL). All NMR spectra combined with magic angle spinning (MAS) spectra were measured with proton decoupling during data acquisition. The chemical shifts are calibrated by TMS for 2%-MASNMR measurements. In order to determine the bulk Si/Al atomic ratio of zeolite, x-ray fluorescence analyses (SXF-1200, Shimadzu: Rh target, 40 kV, 70 mA) have been employed. 3. RESULTS AND DISCUSSION

3.1 Catalytic activity and selectivityfor methylamine synthesis Acid types of zeolite catalysts tested in this work include faujasite (H-Y), mordenite (H-M, H-SC-M, and Na-SC-M), and ZSM5 (H-ZSM5). Amorphous Si02-AI203 (JRC-SAL-2)was also tested for this reaction. The catalytic activities and selectivities over the catalysts are summarized in Figure 1. On H-Y, Si02-Al203, and H-M catalyst, the conversion of methanol is higher than on the other zeolite catalysts, but the selectivities of TMA and dimethyl ether (DME) are rather high; the selectivity is about 32-53 % for TMA and 7-26 % for DME. The higher catalytic activity may be due to a higher concentration of aluminum sites (lower Si/Al ratios), which gives higher numbers of acidic sites. Therefore, the reaction rate on H-ZSM5 is lower than that of H-Y and HM. In addition, the acid type of zeolite catalysts shows higher activity for the reaction than the Na-form of zeolites. On Na-SC-M (Na-mordenite treated with SiCl4), the activity is about one or two orders of magnitude lower than on H-M.

1276

However, the selectivities for TMA and DME are extremely retarded i n comparison with the catalysts without treatment with SiCl4. When Na-SC-M ie ion-exchanged t o protonic form (H-SC-M), the catalytic activity is enhanced to a value similar t o that of H-M, and selectivities of MMA and DMA are much higher than .that on H-M. The formation of TMA and DME on H-SC-M is almost negligible up to 100 8 of methanol conversion. In addition, no deactivation has been observed over the whole range of reaction conditions.

BY

1

H-M

SilAb2.8

S ilA1-5.3 SilAI-9.7

H-SC-M

Si/AI=l0.5

H-ZSM5

SilA1=12.5

0

20

60

40

80

Yield

H DME

MMA DMA TMA

100

Conversion and Yleld /%

Figure 1. Catalytic activities and selectivities for methylamine synthesis on various zeolite catalysts: Reaction temperature=633 K, NHs/CH30H=1.0, W/F=62 g h mol-1.

i

100

,,,...... H-M,,*,........p--'' ..*1-*....

wa'

Conversion

DME

0.0

0.5

1.0 1,5 NHdCH30H

2.0

2.5

,

80 60

"

0.0

0.5

1.0 1.5 NHdCH30H

2.0

2.5

Figure 2. Catalytic activities and selectivities for methylamine synthesis on HM (SUA139.7)and H-SC-M (Si/A1=10.6)as a function of NHs/CH30H: Reaction temperature=633 K, W/F=62 g h mol-1, PMaoH=2.8 kPa.

1277

For industrial applications, a higher partial pressure .of NHs has been required in order to minimize the selectivity of TMA 121 and to achieve a higher conversion level of methanol. As shown in Figure 2, the catalytic activities were increased with increasing partial pressure of NHs, and the selectivities of methylamines are different from H-M and H-SC-M. On H-SC-M, the major product was DMA regardless of NHs pressure, and the selectivities of DME and TMA were much smaller than that on H-M. On the other hand, over H-M catalyst, DME and TMA were the major products at lower partial pressures of NHs, and selectivity of TMA was still high even at higher partial pressures of NHs.

c

m

i' \

1.5 NH3

MMA DMA TMA

1.0

'1

5

0.5

0

0.0

Si02-AI2Q

H-M

H-SC-M

Figure 3. Adsorption of methylamines and NHs over solid acid catalysts: Samples was exposed to 1.3 kPa of base molecules at 373 K and then evacuated at 373 K. SUAl atomic ratios of qamples are the same as in Fig. 1. 3.2 Adsorption ofbase molecuka on catalpta Adsorption uptake of methylamines and NHson the catalysts have been measured by using the vacuum balance. After evacuation a t 773 K, the sample was exposed to 1.3 kPa of base molecules at 373 K. After becoming equilibriated with 1.3 kPa of base molecules, the sample was evacuated at the same temperature of adsorption for 12 h. Chemisorption was determined by the weight increase of the sample from the weight prior to the adsorption. Figure 3 shows the chemisorption uptake of methylamines and NHs on SiOg-A120~,H-M, and H-SC-M. For the adsorption of base molecules, the acidic type of catalysts generally show higher adsorption uptake than on sodium-zeolite (Na-M, NaSC-M). Adsorption of NHs and methylamines on H-M shows higher uptake than on Si02-Al203 and H-SC-M. In addition, after treatment of Na-M by Sic14 (Na-SC-M) and the ion-exchange to protonic form (H-SC-MI, the uptake of TMA is extremely retarded. The results suggest that the diffusion of bulkier base molecules such as TMA into mordenite channels can be especially limited over H-SC-M. These results would correlate the higher selectivities of MMA and DMA on these catalysts. During the adsorption of methylamines,

1278

MMA and DMA are quickly saturated, while TMA needs a longer time t o reach equilibrium on H-M. On the other hand, adsorption of TMA on H-SC-M is completely retarded, only MMA is quickly saturated, and DMA requires a longer time t o be saturated. These sorption profiles on H-SC-M are totally different from the profiles on H-M, since the results suggest that TMA cannot penetrate into the pore channels of H-SC-M, and the size of pore opening of HSC-M can be estimated to be similar t o the kinetic diameter of DMA (0.49nm). H-M

I

H-SC-M

4 1442

Pyridine adsorbed

--

1700

1400

1700

Wavenumber /ern-'

1400

Figure 4. IR spectra of adsorption of pyridine (CsH5N) and NH3 on H-M (Si/Al=9.7) and H-SC-M (Si/Al=10.5):C5H5N adsorbed and evacuated at 473 K, and NH3 adsorbed and evacuated a t room temperature. For acidic types of mordenites, either zeolite pore size or the internal and external acidic sites are significantly affected by the treatment with SiC14. The IR spectra of adsorbed C5H5N and NH3 on acid type mordenites after evacuation at 773 K are shown in Figure 4. The major acid sites of H-M are Br~nsted sites (1542 cm-1 for CsHsN ads. and 1442 cm-l for NH3 ads.). However, the acidic sites on H-SC-M do not interact with C~HSN, but interact with NH3. The results can be explained as due to the lower selectivities of TMA and DME on H-SC-M. The kinetic diameter of C5H5N (0.55-0.60nm) is quite similar t o that of TMA (0.61 nm) 121. The results from IR experiments are in good agreement with the results of adsorptian measurements. The adsorption and IR results suggest that the more bulky base molecules, such as CsHsN or TMA, cannot penetrate into or diffuse out of the mordenite channels after treatment with SiC14. In addition, these results suggest that C5H5N cannot adsorb on the outer surfaces of H-SC-M. The results suggest that there are no catalytic

1279

active sites on the external surface of H-SC-M, which will minimize the formation of DME. Very recently, Corbin and coworkers reported chemical modification of surfaces for improved selectivity to dimethylamine in synthesis from methanol and ammonia [lo]. The relative contributions of external and intracrystalline acidic sites of small pore H-RHO zeolite for the reaction have been studied. The H-RHO zeolites typically contain chabazite and pollucite impurities. Nonselective surface reactions which produce predominantly TMA can be eliminated by covering the external acidic sites with trimethylphosphite. In this study, similar behavior of the effect of surface treatments by Sic14 over mordenites has been observed.

140

40 .-

I

I

I

I

1

1 .o

0.0

0.5

1

1.5

2.0

NH3 adsorbed lmmol g-’

Figure 5 . High-temperature microcalorimetry of NH3 on Na-M (Si/Al=lO.O), Na-SC-M (SilAl=10.3), H-M (Si/A1=9.7),and H-SC-M (SilAl=10.5). Vedrine et al., reported microcalorimetry measurements of NH3 at 423 K to characterize the acid sites in H-ZSM5 [ll]. They found that the acid distribution of stronger acid site is more homogeneous in H-ZSM5 than that of H-Y. We preferred the temperature at 473 K for calorimetry measurements to collect the information of the stronger acid sites for methylamine synthesis. The results are summarized in Figure 5. All microcalorimetric curves are decreased with increasing coverage of NH3 on zeolites. H-M and H-SC-M showed higher initial heat of adsorption (ca. 130 k J mol-1) than Na-M and NaSC-M (ca. 70 k J mol-I), that catalysts have higher acid strength than that of Na-form. In addition, H-M and H-SC-M gave a drastic decrease of the heat of adsorption at particular narrow domain of coverage of NH3. The results suggest that H-M and H-SC-M have more homogeneous A1 distribution along the channels of zeolites, and H-M showed a larger amount of stronger acid sites than H-SC-M. The smaller amount of stronger acid sites on H-SC-M than on H-M may be due to some dealumination during the preparation of catalyst.

1280

3.3 Active dtm ofmordenite befbre or after tmatmemt with Sicl, In order to elucidate higher values of activity and selectivity of methylamine synthesis from CHaOH and NHa over mordenite catalysts, which are treated with Sic14 in sodium form and then ion-exchanged to acidic form (H-SC-M), high resolution solid state magic angle spinning nmr (MASNMR) for mordenite samples has been studied. In framework aluminosilicates, there are five possibilities, described by the formula Si(nAl), where n=O, 1, 2, 3 or 4. These five basic units of Si(nAl)express the fact that each silicon atom is linked, via oxygene, to n aluminum neighbors 1121. The 2QSi-MASNMRspectra of Na-M, Na-SC-M, H-SC-M, H-M show three different resonance peaks, which represent three kinds of Si-tetrahedra attached to different numbers of adjacent Altetrahedra: Si(2Al), Si(lA11, and Si(OA1) [131. The distribution of Si(nA1) before treatment with Sic14 is similar to that after treatment with SiCla. 20 I

I

Bulk (SVAI)XRF 10

n "

Framework (SVAI)NMR

Surface (SVAl)xps

H-M

Na-M

NaSC-M

HSC-M

Figure 6. SUAl ratios of mordenites before and after treatment with SiCl4: Bulk SUAl were determined by x-ray fluorescence analysis, framework Si/Al were determined by 2QSi-MASNMR,and surface SVAI were determined by xray photoelectron spectroecopy.

SitAl atomic ratios of mordenite samples before or after treatment with Sic14 are shown in Figure 6 . Si/Al atomic ratios were determined by x-ray fluorescence analysis (bulk composition of mordenite), by 2QSi-MASNMR (framework composition of mordenite), and by x-ray photoelectron spectroscopy (surface composition of mordenite). In H-M and Na-M, SUAl ratios in bulk, framework, and surface show almost similar numbers that represent very uniform distributions in parent mordenite samples. When Na-M was treated with Sic14 at 973 K (Na-SC-M), only the values of the surface are enhanced due to the deposition of silicious materials on the surfaces. However, no detectable dealumination from mordenite framework has been observed. Such results suggest that only very small amounts of Si02 are deposited on the external surfaces of Na-M. In addition, when Na-SC-M wa8 ion-exchanged t o protonic form (H-SC-M), some dealumination of framework

1281

aluminum has occurred, which indicated by framework Si’/Al ratios, and the values of Si/Al in surface are also higher than the values in bulk or framework. SiOp coatings on the external surfaces of mordenites in sodium form and then ion-exchanged to acidic form are very effective in increasing DMA and MMA and in decreasing TMA and DME selectivities in synthesis from CH30H and NH3. 4. CONCLUSIONS

Methylamine syntheses from methanol and ammonia on various zeolite catalysts have been studied. We found that the mordenite catalyst which was treated with Sic14 in sodium-form and then ion-exchanged to acidic form (HSC-M) shows higher selectivity and activity for MMA and DMA. The selectivity of MMA plus DMA was ca. 99 96 at 90 % methanol conversion (NHdCH30H 40,633K). Over this catalyst, dealumination in the framework does not occur significantly over Na-M after treatment with SiCl4, but does occur during ion-exchange procedures. However, the stronger acidic sites, which are located along the main Channel of mordenites are still survive even after treatment with SiC14. Those Br~nstedacid sites along the mordenite channels act as catalytic active sites for methylamine synthesis. On the other hand, when H-M was treated with Sic14 instead of Na-M at 973 K, serious dealumination occurred. This catalyst showed non-selective reaction, which favors DME and TMA from the reaction with CHsOH and NH3.

sio7

MMA: 0.4 nm

Mordenite Channels

DMA: 0.49 nm

TMA: 0.61 nm

Scheme 1 Shape selectivity bf methylamine synthesis over H-SC-M. The formation of TMA over H-SC-M has been extremely retarded, because the pore openings of the catalyst are smaller than TMA molecules, which are shown in Scheme 1. Pore structure of H-SC-M is accessible only to MMA and DMA. In addition, the selectivity of DME over H-SC-M also has been retarded. The formation of DME on the catalysts during the reaction may be generated

1282

on weaker acidic sites, which are located on external surfaces of zeolites. However, the external surfaces of H-SC-M are already covered with Si02. The 27A1-MASNMR spectra of Na-M and of Na-SC-M give one sharp resonance peak at 59 ppm from Al(H2O)$+, which is assigned to tetrahedral framework sites occupied by the aluminum 1131. 2"-MASNMR also shows directly that six-coordinated aluminum (Al-octahedra) builds up at the expense of the fourcoordinated aluminum in the framework a t -1.4 ppm, when the Na-SC-M sample is ion-exchanged to protonic form (H-SC-M). If we can minimize the dealumination, such as the formation of six-coordinated aluminum sites, during the ion-exchange procedures, we may be able to obtain a more highly active catalyst for methylamine synthesis.

1 2

3 4 5

6 7 8 9 10 11

12 13

Y. Ashina, T. Fujita, M. Fukatsu, and J. Yagi, Eur. Patent Appl. 130,407(June 6,1984). Y. Ashina, T.Fujita, M. Fukatu, K. Niwa, and Y. Yagi, in "Proceedings of the 7th International Zeolite Conference, Tokyo, 1986" (Y. Murakami, A. Iijima and J. W.Ward, Eds.), p. 1960. KodanshaElsevier, Tokyo, 1988. F. J. Weigert, J. Catal., 103 (1987)20. I. Mochida, A. Yasutake, H. Fujitsu, and K. Takeshita, J. Catal., 82 (1983)313. R. D. Shannon, M. Keane, Jr., L. Abrams, R. H. Staley, T. E. Gier, D. R. Corbin, and G. Sonnichsen, J. Cutal., 113 (1988)367. R. D. Shannon, M. Keane, Jr., L. Abrams, R. H. Staley, T. E. Gier, D.R. Corbin, and G. C. Sonnichsen, J. Catal., 114 (1988)8. R. D. Shannon, M. Keane, Jr., L. Abrams, R. H. Staley, T. E. Gier, and G. C.Sonnichsen, J. Catal., 115 (1988)79. H.E. Bergna, M. Keane, Jr., D. H. Ralston, G. C. Sonnichsen, L. Abrams, and R. D. Shannon, J. Catal., 116 (19881148. M. Keane, Jr.,G. C. Sonnichsen, L. Abrams, D. R. Corbin, T. E. Gier, and R. D. Shannon, Appl. Catal., 32 (1987)361. D. R. Corbin, M. Keane, Jr., L. Abrams, R. D. Farlee, P. E. Bierstedt, and T. kein, J. Catal., 124 (1984)268. J. Vedrine, A. Auroux, P. Dejave, V. Ducatme, H. Hoser, and S. Zhou, J. Catal., 73 (1982)147. J. M. Thomas and J. Klinowski,Adu. Catal., 33 (1985)199. K.Segawa and H. Tachibana, J. Catal., 131 (1991)482.

1283

DISCUSSION Q: W. Holderich (Germany) Can you make a comment in respect to the life time of your SiCI, treated mordenite compared to that of the Nitto-catalyst (mordenite with carefully adjusted Nacontent) ?

A K. Segawa

No deactivation of the catalyst has observed, if you run the reaction below 633 K. Because of the presence of ammonia, the deactivation by the acidic sites may retarded significantly. Na-form of mordenite catalyst (Nitto-catalyst) shows also higher effect against the deactivation, but it shows less active than that of protonic forms of mordenite.

Q: R. Prins (Switzerland)

Your H-SC-M catalyst has a very nice shape selectivity and makes dimethylamine (DMA). Assuming that dimethylether @ME) is about as large as DMA, how d o you then explain the suppression of the DME formation ? A: K. Segawa According to the sorption experiment of DME, the sorption profiles shows very much like the profile of DME. The results suggest that it is slightly hard to defuse out from the main channel of the catalyst. As far as DME react with ammonia to produce either monomethylamine (MMA) or DMA inside the channel, the products may easier to diffuse out. In addition, the catalyst becomes inactive on the outer surfaces of the catalyst, since it already covered with SiOP

Q: R. F. Howe (Australia) Your reaction is producing water as a reaction product. Do you have any evidence for dealumination as a result of steaming during reaction, which might ultimately limit catalyst life time ?

A. K. Segawa We checked AI distribution of the catalyst by 29Si-MAS Nh4R before or after reaction, the spectra show that it has no significant change over the catalyst. As the ,y = 2.8 kPa), the effect of water reaction media diluted with excess amount of N2 (Po vapor in the reaction media may negligible for dealumination of catalyst. Q: W. 0. Haag (USA) Your results are very interesting and significant. One might be inclined to attribute the low selectivity to MMA to serve diffusion limitation in the SiO modified mordenite, which favors the consecutive reaction products such as DMA a n d h A . Also, the very slow uptake rates in the sorption experiments point towards diffusion limitation. Do you have information whether at the higher temperature of the catalytic conversion the reaction rate is diffusion limited and whether this is the reason for the observed selectivity '? Furthermore, how much SiO, has been added to the catalyst, and does it have any effect on the available pore volume, as measured by N2 or small hydrocarbon sorption ?

A: K. Segawa We measured the initial reaction rate as a function of reaction temperature. The apparent activation energy over H-SC-M at higher temperature re ion (above 633 K) was 37 kJ mol-l, and the energy below 633 K was 90 kJ mol' f . We think at higher temperature region, the reaction may involve the diffusion limitation. During the treatment with SiC14 over Na-mordenite, SiCI, can not penetrate into mordenite channel due to the bulkiness of SiCI4. Therefore, the pore volumes of the catalyst may not change significantly before or after treatments.