Dimethyl Ether Synthesis from CO2–H2 Mixture over Cu ...

388 Journal of the Japan Petroleum Institute, 63, (6), 388-393 (2020)

J. Jpn. Petrol. Inst., Vol. 63, No. 6, 2020

1. Introduction

Production of methanol from CO2 by hydrogenation is a potential method for decreasing CO2 emissions1)～3). Dehydration of the methanol can form dimethyl ether (DME, CH3OCH3), a source of clean energy. DME does not emit particulate matter (PM) since it contains no C_C bond. The chemical properties are very simi-lar to those of liquefied petroleum gas (LPG)3),4). Therefore, DME is a promising fuel that can be synthe-sized from CO2 emissions.

Scheme 1 shows the overall DME synthesis from CO2 and H2 via methanol, which involves methanol synthesis (1) and methanol dehydration (2), whereas reverse water gas shift (RWGS) (3) and methanol decomposition (4) are undesirable side reactions result-ing in CO formation.

CO2 + 3H2 → CH3OH + H2O (1)

2CH3OH → CH3OCH3 + H2O (2)

CO2 + H2 → CO + H2O (3)

CH3OH → CO + 2H2 (4)

Cu_Zn catalysts are widely utilized for methanol syn-thesis via CO2 hydrogenation (1)5)～7). In addition, In2O3 catalysts8),9) and TiO2-based catalysts10),11) have been reported as a catalyst for methanol synthesis in recent years. Solid acid catalysts, such as zeolite, are widely utilized for methanol dehydration (2)12)～14). The main problem encountered in DME synthesis from a CO2

_H2 mixture is insufficient CO2 activation in the methanol synthesis (1). Low temperature and high pressure are preferable for the thermodynamics of methanol synthesis15), so high catalytic activity is required at low temperatures. Another problem is low methanol selectivity, caused by the side reactions of the

[Technical Report]

Dimethyl Ether Synthesis from CO2–H2 Mixture over Cu/Amorphous-ZrO2 Mixed with FER-type Zeolite

Kazumasa OSHIMA†1)＊, Shiori NAKAJIMA†1), Shohei TADA†2), Ryuji KIKUCHI†3), and Shigeo SATOKAWA†1)＊

†1) Dept. of Materials and Life Science, Faculty of Science and Technology, Seikei University, 3-3-1 Kichijoji Kitamachi, Musashino, Tokyo 180-8633, JAPAN

†2) Dept. of Materials Science and Engineering, College of Engineering, Ibaraki University, 4-12-1 Nakanarusawa-cho, Hitachi, Ibaraki 316-8511, JAPAN

†3) Dept. of Chemical System Engineering, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, JAPAN

(Received June 11, 2020)

Combinations of Cu-based catalysts and acidic zeolite were investigated for dimethyl ether (DME) production from a mixture of CO2 and H2. The physical mixture of Cu/a-ZrO2 (a-: amorphous) catalyst and protonated FER-type zeolite produced a higher yield of DME than that of a Cu/ZnO/Al2O3 catalyst and FER-type zeolite at high pressure of 1.0 MPa. Cu/a-ZrO2 catalyst showed relatively low activity in terms of undesirable CO forma-tion, and the FER-type zeolite combined with the Cu/a-ZrO2 catalyst had good activity for methanol dehydration, leading to the high yield of DME. DME selectivity over the combination catalyst of Cu/a-ZrO2 and FER reached almost 40 %.

KeywordsDimethyl ether synthesis, Carbon dioxide hydrogenation, Amorphous zirconia support, Copper catalyst, FER-type zeolite

DOI: doi.org/10.1627/jpi.63.388 ＊ To whom correspondence should be addressed. ＊ E-mail: [email protected] (Kazumasa Oshima)

[email protected] (Shigeo Satokawa)

Scheme 1　 Reaction Scheme of DME Synthesis from CO2 Hydroge-nation

389


reverse water gas shift (3) and methanol decomposition (4). By-product water can cause deactivation of the Cu_Zn and zeolite catalysts16). Therefore, direct syn-thesis of DME from a CO2

_H2 mixture requires high methanol selectivity.

DME synthesis from a mixture of CO2 and H2 using a tandem reactor has been extensively investigated to improve the process efficiency by omitting the dehydra-tion step. A catalyst specific for DME synthesis was prepared by mixing a methanol synthesis catalyst with a methanol dehydration catalyst17)～21). DME synthe-sis rarely uses methanol synthesis catalysts other than Cu_Zn catalysts22),23). Cu_Zn catalysts are optimized for CO hydrogenation (a current methanol synthesis process), but are not sufficient active for CO2 hydroge-nation24). In addition, Cu_Zn catalysts are not suitable for DME synthesis in the tandem reactor because of the significant poisoning by water. Therefore, new cata-lysts must be developed that are specific for CO2 hydro-genation.

Recently, we reported the use of copper nanoparticles supported on amorphous ZrO2 (Cu/a-ZrO2; denoted as CZ) as an active catalyst for CO2 hydrogenation25)～27). The most significant feature of the CZ catalyst is the high selectivity for methanol compared to the Cu_Zn catalyst because methanol decomposition (4) is sup-pressed26). Furthermore, the CZ catalyst is presumably more resistant to water than the Cu_Zn catalyst because the a-ZrO2 support has a large number of hydroxyl groups on the surface. In contrast, DME formation via methanol dehydration (2) depends on the acidity of the solid acid catalyst. Acidic zeolites (FER, MFI, and CHA) and α-Al2O3 are employed for DME formation due to their satisfactory catalytic performances19)～21). FER has the highest activity because of the strong acid sites20). Therefore, the CZ catalyst, in combination with a solid acid catalyst, is a potential catalyst for DME synthesis from the CO2

_H2 mixture.The present study investigated DME synthesis from a

mixture of CO2 and H2 using a mixture of CZ catalyst with FER-type zeolite. For comparison of catalytic activity, we also prepared a Cu/ZnO/Al2O3 catalyst mixed with FER-type zeolite. Evaluation the catalytic performances revealed the advantage of the CZ catalyst for one pass DME synthesis from a CO2

_H2 mixture.

2. Experimental

2. 1. Catalyst PreparationThe CZ catalyst was prepared by an incipient wet-

ness impregnation method using amorphous ZrO2 (NND supplied by Daiichi Kigenso Kagaku Kogyo Co., Ltd.) and copper nitrite monohydrate (Cu(NO3)2, Kanto Chemical Co., Inc.). a-ZrO2 was impregnated with an aqueous solution of Cu(NO3)2. The gel was dried at 110 °C overnight and calcined at 350 °C for 5 h in an elec-

tric furnace. According to our previous research25)～27), copper cations in the catalyst were doped into the amor-phous ZrO2. Cu/ZnO/Al2O3 (MDC-7 supplied by Clariant K.K.; denoted as MDC) was used as a Cu_Zn catalyst in this experiment.

FER-type zeolite (HSZ-720KOA supplied by Tosoh Corp.) was protonated with ammonium nitrate solution and then calcined at 500 °C in air prior to use. The protonated sample was named FER.

For the activity test, the copper-based catalyst and the FER-type zeolite were physically mixed in an agate mortar for 10 min with the same copper content. Therefore, the amount of the mixed catalyst used for the activity test was as follows: (i) 500 mg of CZ, (ii) 500 mg of CZ and 100 mg of FER (denoted as CZ＋FER), (iii) 100 mg of MDC, and (iv) 100 mg of MDC and 100 mg of FER (denoted as MDC＋FER).2. 2. Characterization

The specific surface areas (SSAs) of the copper based catalyst and FER zeolite were measured by N2 adsorp-tion at 77 K (BELSORP-mini II, MicrotracBEL Corp.) after a heat treatment to remove surface moisture. The copper content of the CZ and the MDC, and SiO2/Al2O3 molar ratio of FER, were measured by X-ray fluores-cence (XRF, Epsilon 1, Malvern PANalytical). Trans-mission electron microscopy (TEM) images of the mixed catalysts of CZ＋FER and MDC＋FER were collected using a JEM-2100F (JEOL Ltd.) operated at 200 kV. The exposed surface areas of metallic copper of the CZ and the MDC were measured by an N2O titra-tion technique, according to our previous study27). We previously reported that the interaction between copper and a-ZrO2 prevents N2O molecules from oxidizing the metallic copper surface, which hinders the accurate measurement of the surface area of metallic copper27). Therefore, the space velocity and catalytic activity for the reaction test were normalized by the copper amount rather than by the copper surface area. Furthermore, the amount of acid site on FER was evaluated by NH3-temperature-programmed desorption (TPD) mea-surement after removing moisture adsorbed on the sur-face at 500 °C. H2O titration and NH3-TPD were mea-sured using a BELCAT (MicrotracBEL Corp.).2. 3. Activity Test

Catalytic activity tests were conducted using a high-pressure fixed-bed flow tubular reactor (PID Eng&Tech, Microactivity Effi reactor). The catalyst was charged into the reactor with 5.0 g of quartz sand as a diluent. After H2 reduction (H2/N2＝10/50 mL min–1) at 300 °C for 2 h, stabilizing treatment was conducted at 270 °C for 12 h to evaluate the stable activity. The feed gas was supplied at 42 mLSTP min–1 with composi-tion of CO2/H2/N2＝1/3/1. Thus, the feed gas volume per unit time per unit weight (denoted as F/W) was ca. 70 LSTP h–1 g-Cu

–1. The reaction temperature and pres-sure were 180-230 °C and 1.0 MPa, respectively. The

390


products were analyzed using gas chromatographs equipped with a flame ionization detector (FID, for the detection of CH3OH and DME) and a thermal conduc-tivity detector (TCD, for the detection of N2, CO, and CO2). The conversion and selectivity were calculated based on the carbon products.

3. Results and Discussion

3. 1. Catalyst PropertiesThe SSAs of the CZ and MDC catalysts were 221 m2

g–1 and 61 m2 g–1, respectively, as shown in Table 1. The SSA of the CZ catalyst was larger than that of the MDC catalyst due to the amorphous support. The copper contents of the CZ and MDC catalysts were 7 wt% and 38 wt%, respectively (Table 1). The cop-per content of the MDC catalyst was five-fold higher than that of the CZ catalyst. The exposed surface areas of metallic copper of the CZ and MDC catalysts were 7.8 μmol g–1 and 118 μmol g–1, respectively (Table 1). The metal surface area for the MDC cata-lyst was larger than that for the CZ catalyst. There-fore, the amount of surface copper acting as the active sites of the CZ catalyst was much smaller than that of the MDC catalyst, although the CZ catalyst had a larger SSA. The SiO2/Al2O3 molar ratio, SSA, and amount of acid sites of the FER catalyst were 18.1, 384 m2 g–1, and 0.25 mmol-NH3 g–1, respectively.

Figure 1 shows the TEM images of the mixed cata-lysts of CZ＋FER and MDC＋FER. The FER parti-cles of 400-500 nm size were partially covered with copper-based catalyst particles of ＜50 nm size. The particle sizes were almost the same in the copper-based

and FER catalysts, and contact between these catalysts was confirmed.3. 2. Comparison with Cu–Zn Catalyst

First, the activities of the CZ, CZ＋FER, MDC, and MDC＋FER catalysts were compared. Figure 2 presents the formation rates of DME, CH3OH, and CO in DME synthesis from CO2

_H2 mixture over these cat-alysts at 230 °C, and the carbon-based selectivity based on the total of CH3OH and DME. Since neither CZ nor MDC produced any DME, the presence of FER was essential for methanol dehydration to form DME (2). CO2 conversions over CZ, CZ＋FER, MDC, and MDC＋FER catalysts were 2.4, 3.0, 4.3, and 1.9 %, respec-tively. Since the formation rate of CO via the sequen-tial reaction (4) is substantially dependent on CO2 con-version, the CO formation rate over the CZ＋FER catalyst was higher than that over the CZ catalyst, indi-cating that the sequential reaction (4) proceeded over both CZ and FER catalysts. The methanol yield of CZ (9 mmol h–1 g-Cu

–1) was higher than that of MDC (7 mmol h–1 g-Cu

–1), and the methanol selectivity of CZ

Table 1　Properties of the Catalysts

Catalyst SSA [m2 g–1]Cu content

[wt%]N2O conc. [μmol g–1]

CZ 221 7 7.8MDC 61 38 118FER 384 - -

Fig. 1　TEM Images of the Mixed Catalysts of CZ＋FER (a) and MDC＋FER (b)

Percentages in the figure correspond to the CH3OH＋DME selectiv-ity. F/W: ca. 70 LSTP h–1 g-Cu

–1.

Fig. 2　 Catalytic Activity of the Prepared Catalysts for Conversion of the CO2 and H2 Mixture at 230 °C and 1.0 MPa

391


(64.0 %) was higher than that of MDC (29.6 %). Therefore, CZ catalyst was more specific for CO2-to-methanol hydrogenation than MDC, consistent with our previous report27). The formation rates of CH3OH and DME over CZ＋FER (6 mmol h–1 g-Cu

–1) were approximately three-fold higher than those over MDC＋FER (2 mmol h–1 g-Cu

–1) despite the same F/W (ca. 70 LSTP h–1 g-Cu

–1). Such improvement may be significant for the manufacturing processes of chemical products from CO2

_H2 mixture.Next, the relationship between CO2 conversion and

product selectivity over CZ＋FER and MDC＋FER catalysts was investigated (Fig. 3). CO2 conversion was controlled by changing the F/W: CO2 conversion increased with lower F/W from 20 to 120 LSTP h–1 g-Cu

–1. CO selectivity increased with higher CO2 conversion, whereas methanol and DME selectivity simultaneously

decreased. Therefore, methanol decomposition (4) is inevitable at high CO2 conversion over both catalysts. This study focused on the trendline between CO2 con-version and CO selectivity. The trendline for the CZ＋FER catalyst was below 60 C-%, whereas that of MDC＋FER is above 60 C-%. CZ catalyst shows rel-atively low activity for the undesirable methanol decomposition (4), leading to high selectivity for DME and methanol. Therefore, the catalytic activity of DME formation over the mixture of Cu-based catalyst and FER-type zeolite could be increased by changing the Cu/ZnO/Al2O3 catalyst to the CZ catalyst.3. 3. Performance of CZ Catalyst

Figure 4 shows the effect of the amount of zeolite addition on the activity of CZ＋FER catalyst. CO2 conversion was almost the same due to the introduction of the same amount of CZ catalyst into the catalyst bed, which converts CO2 to methanol. The CZ＋FER cata-lyst showed sufficient DME synthesis activity: DME selectivity was ca. 30 % with 500 mg of CZ mixed with a small amount of FER (25 mg). However, further increase in the amount of FER from 25 to 500 mg did not affect the DME selectivity, probably due to the extent of the equilibrium between CH3OH and DME (2).

Figure 5 shows the temperature dependence of CO2 conversion and DME selectivity over the CZ＋FER and MDC＋FER catalysts. The CZ＋FER catalyst showed higher activity than the MDC＋FER catalyst at all tem-peratures. In particular, the difference in catalytic activity was remarkable at high temperatures at which large amounts of water were produced. Therefore, the catalytic activity of CZ＋FER was superior for DME synthesis from CO2

_H2 mixture. The high DME yield over our CZ＋FER system

probably results from the superior catalytic perfor-

F/W: ca. 120 to 20 LSTP h–1 g-Cu–1.

Fig. 3　 Relationship between CO2 Conversion and Selectivity for DME, CH3OH, and CO over CZ＋FER Catalyst (a) and MDC＋FER Catalyst (b) at 230 °C and 1.0 MPa

Fig. 4　 Effect of Amount of Zeolite Addition on Catalytic Activity at 230 °C and 1.0 MPa

392


mance of the CZ catalyst for CO2-to-methanol hydroge-nation. This study used the CZ catalyst (7 wt% Cu/a- ZrO2), which showed the best performance for metha-nol production among a series of Cu/a-ZrO2 catalysts25). The CZ catalyst produced more methanol than the MDC catalyst even with F/W of ca. 8 LSTP h–1 g-cat

–1 27) (contact time normalized by catalyst weight) and ca. 70 LSTP h–1 g-Cu

–1 (contact time normalized by Cu load-ing, present Fig. 2). However, the low conversion of CO2 is one of the drawbacks of the CZ＋FER system for industrial applications, due to the catalytic perfor-mance limitations of the CZ catalyst. In the future, we intend to further improve the activity of the CZ catalyst for DME formation.

4. Conclusion

The CZ＋FER catalyst has potential for DME syn-thesis from CO2

_H2 mixture at 230 °C. The CZ can convert CO2 selectively to methanol and suppress unde-sirable CO formation. The present study revealed that using a catalyst with high methanol selectivity from CO2

_H2 mixture, such as the Cu/a-ZrO2 catalyst, com-pared to the Cu_Zn catalyst, can directly synthesize chemicals such as DME.

AcknowledgmentWe are grateful to Dr. Takashi Toyao and Mr. Ting

Kah Wei, Hokkaido University, for their kind help with the TEM measurements. This research was supported by the Cooperative Research Program of Institute for Catalysis, Hokkaido University (Grant, 19B1012), Japan, and Unexplored Challenge 2050, New Energy and Industrial Technology Development Organization (NEDO), Japan.

References

1) Olah, G. A., Angew. Chem. Int. Ed., 52, 104 (2013). 2) Gao, P., Li, S., Bu, X., Dang, S., Liu, Z., Wang, H., Zhong, L.,

Qiu, M., Yang, C., Cai, J., Wei, W., Sun, Y., Nat. Chem., 9, 1019 (2017).

3) Alvarez, A., Bansode, A., Urakawa, A., Bavykina, A. V., Wezendonk, T. A., Makkee, M., Gascon, J., Kapteijn, F., Chem. Rev., 17, 9804 (2017).

4) Ogawa, T., Inoue, N., Shikada, T., Ohno, Y., J. Nat. Gas Chem., 12, 219 (2003).

5) Chanchlani, K. G., Hudgins, R. R., Silveston, P. L., J. Catal., 136, 59 (1992).

6) Ostrovskii, V. E., Catal. Today, 77, (3), 141 (2002). 7) Pontzen, F., Liebner, W., Gronemann, V., Rothaemel, M.,

Ahlers, B., Catal. Today, 171, (1), 242 (2011). 8) Ye, J., Liu, C., Mei, D., Ge, Q., ACS Catal., 3, (6), 1296 (2013). 9) Martin, O., Martin, A. J., Mondelli, C., Mitchell, S., Segawa, T.

F., Hauert, R., Drouilly, C., Curulla-Ferre, D., Perez-Ramirez, J., Angew. Chem. Int. Ed., 55, (21), 6261 (2016).

10) Ting, K. W., Toyao, T., Siddiki, S. M. A. H., Shimizu, K., ACS Catal., 9, (4), 3685 (2019).

11) Toyao, T., Kayamori, S., Maeno, S., Siddiki, S. M. A. H., Shimizu, K., ACS Catal., 9, (9), 8187 (2019).

12) Xu, M., Lunsford, J. H., Goodman, D. W., Bhattacharyya, A., Appl. Catal. A: Gen., 149, (2), 289 (1997).

13) Ramos, F. S., Farias, A. M. D., Borges, L. E. P., Monteiro, J. L., Fraga, M. A., Sousa-Aguiar, E. F., Appel, L. G., Catal. Today, 101, (1), 39 (2005).

14) Catizzone, E., Aloise, A., Migliori, M., Giordano, G., Appl. Catal. A: Gen., 502, 215 (2015).

15) Ateka, A., Perez-Uriarte, P., Gamero, M., Erena, J., Aguayo, A. T., Bilbao, J., Energy, 120, 796 (2017).

16) Wang, D., Han, Y., Tan, Y., Tsubaki, N., Fuel Process. Technol., 90, 446 (2009).

17) Garcia-Trenco, A., Martinez, A., Appl. Catal. A: Gen., 411, 170 (2012).

18) Bonura, G., Migliori, M., Frusteri, L., Cannilla, C., Catizzone, E., Giordano, G., Frusteri, F., J. CO2 Util., 24, 398 (2018).

19) Sai Prasad, P. S., Bae, J. W., Kang, S. H., Lee, Y. J., Jun, K. W., Fuel Process. Technol., 89, (12), 1281 (2008).

20) Frusteri, F., Migliori, M., Cannilla, C., Frusteri, L., Catizzone, E., Aloise, A., Giordano, G., Bonura, G., J. CO2 Util., 18, 353 (2017).

21) Sun, J., Yang, G., Yoneyama, Y., Tsubaki, N., ACS Catal., 4, (10), 3346 (2014).

22) Wang, S., Mao, D., Guo, X., Wu, G., Lu, G., Catal. Commun., 10, (10), 1367 (2009).

23) Bahruji, H., Armstrong, R. D., Ruiz Esquius, J., Jones, W., Bowker, M., Hutchings, G. J., Ind. Eng. Chem. Res., 57, (20), 6821 (2018).

24) Martin, O., Perez-Ramirez, J., Catal. Sci. Technol., 3, 3343 (2013).

25) Tada, S., Kayamori, S., Honma, T., Kamei, H., Nariyuki, A., Kon, K., Toyao, T., Shimizu, K., Satokawa, S., ACS Catal., 8, (9), 7809 (2018).

26) Tada, S., Katagiri, A., Kiyota, K., Honma, T., Kamei, H., Nariyuki, A., Uchida, S., Satokawa, S., J. Phys. Chem. C, 122, (10), 5430 (2018).

27) Tada, S., Oshima, K., Noda, Y., Kikuchi, R., Sohmiya, M., Honma, T., Satokawa, S., Ind. Eng. Chem. Res., 58, (42), 19434 (2019).

F/W: ca. 70 LSTP h–1 g-Cu–1.

Fig. 5　 Temperature Dependence of CO2 Conversion and DME Selectivity over CZ＋FER and MDC＋FER Catalysts

393


要　　　旨

非晶質ジルコニア担持銅触媒と FER型ゼオライト複合触媒上での CO2–H2混合ガスからのジメチルエーテル合成

大島　一真†1)，中嶋　栞理†1)，多田　昌平†2)，菊地　隆司†3)，里川　重夫†1)

†1) 成蹊大学理工学部物質生命理工学科，180-8633 東京都武蔵野市吉祥寺北町3-3-1†2) 茨城大学工学部物質科学工学科，316-8511 茨城県日立市中成沢町4-12-1†3) 東京大学工学部化学システム工学科，113-8656 東京都文京区本郷7-3-1

CO2と H2の混合ガスから一段でのジメチルエーテル（DME）合成を目的として，Cu系触媒とゼオライトの混合触媒の触媒性能を評価した。非晶質ジルコニアを担体とした担持銅触媒（Cu/a-ZrO2）と FER型ゼオライトの混合触媒は，CO2水素化に用いられる Cu/ZnO/Al2O3と FER型ゼオライトの混合触媒よりも高い DME収率を示した。非晶質ジルコニアを担体とすることで，副反応である CO生成が抑制されるため，高い DME収

率を示したと考えられる。また，FER型ゼオライトはメタノール脱水に有効な酸点を有しており，これらの混合が CO2から一段での DME合成に有効であることが示された。反応圧力1.0 MPa，反応温度230 ℃の条件での DME選択率は，その条件での平衡組成である40 %に近い値を示しており，Cu/ZnO/

Al2O3との混合触媒よりも約2倍の DME生成量を達成した。

Dimethyl Ether Synthesis from CO2–H2 Mixture over Cu ...

Documents

Transcript of Dimethyl Ether Synthesis from CO2–H2 Mixture over Cu ...