Hydrogasification of Loy Yang Brown Coal by Ion-Exchanged Nickel

The hydrogasification behavior of Loy Yang brown coal with ion-exchanged nickel ... the amount of exchanged nickel species, comparable to a broader pe...
0 downloads 0 Views 53KB Size
1240

Energy & Fuels 2000, 14, 1240-1244

Hydrogasification of Loy Yang Brown Coal by Ion-Exchanged Nickel Species Kenji Murakami, Masahiko Arai, and Masayuki Shirai* Department of Materials-Process Engineering & Applied Chemistry for Environments, Faculty of Engineering and Resource Science, Akita University, Akita, 010-8502 Japan, and Division of Materials Science and Engineering, Graduate School of Engineering, Hokkaido University, Sapporo 060-8628, Japan, and Institute for Chemical Reaction Science, Tohoku University, Katahira, Sendai, 980-8577 Japan Received May 17, 2000. Revised Manuscript Received August 30, 2000

The hydrogasification behavior of Loy Yang brown coal with ion-exchanged nickel species was studied. The C1 gas evolution profile was affected by nickel species loaded. The peak of CH4 evolution appeared at 860-1010 K, depending on the amount of exchanged nickel species, became larger with increasing nickel loading. The total amounts of CO evolved were not enhanced by the presence of nickel, however, the maximum temperature shifted to 640-700 K, depending on the amount of exchanged nickel species, comparable to a broader peak at 810 K for an acid washed coal. The CO2 evolution peak at 650 K became slightly greater with nickel loadings, but the peak top temperature did not change. The agglomeration temperatures of the 1.1 wt %-, 6.6 wt %-, and 12.7 wt %-nickel-loaded samples were 700, 620, and 600 K, respectively. Large nickel metal particles catalyzed the CH4 formation from the Loy Yang coal under hydrogen atmosphere.

Introduction Highly dispersed catalysts are more effective for coal conversion processes (e.g., coal gasification, liquefaction and pyrolysis) than conventionally mixed catalyst powder. An ion-exchanging method is useful to support welldispersed catalysts, which have large surface areas. Many studies have been performed in order to clarify the catalysis of ion-exchanged metal species.1-22 Al* Corresponding author.Tel. and Fax.: +81-22-217-5631. E-mail: [email protected]. (1) Hatswell, M. R.; Jackson, W. R.; Larkins, F. P.; Marshall, M.; Rash, D.; Rogers, D. E. Fuel 1983, 62, 336-341. (2) Hengel, T. D.; Walker, P. L., Jr. Fuel 1984, 63, 1214-1220. (3) Agnew, J. B.; Jackson, W. R.; Larkins, F. P.; Rash, D.; Rogers, D. E.; Thewlis, P.; White, R. Fuel 1984, 63, 147-152. (4) Nabatame, T.; Ohtsuka, Y.; Takarada, Y.; Tomita, A. J. Fuel Soc. Jpn. 1986, 65, 53-58. (5) Takarada, T.; Nabatame, T.; Ohtsuka, Y.; Tomita, A. Energy Fuels 1987, 1, 308-309. (6) Takarada, Y.; Ohtsuka, Y.; Tomita, A. J. Fuel Soc. Jpn. 1988, 67, 683-692. (7) Takarada, T.; Nabatame, T.; Ohtsuka, Y.; Tomita, A. Ind. Eng. Chem. Res. 1989, 28, 505-510. (8) Salinas-Martinez de Lecea, C.; Almela-Alarcon, M.; LinaresSolano, A. Fuel 1990, 69, 21-27. (9) Kyotani, T.; Hayashi, S.; Tomita, A. Energy Fuels 1991, 5, 683688. (10) Ohtsuka, Y.; Asami, K. Ind. Eng. Chem. Res. 1991, 30, 19211926. (11) Joseph, J. T.; Forrai, T. R. Fuel 1992, 71, 75-80. (12) Miki, K.; Yamamoto, Y.; Inaba, A.; Sato, Y. Fuel 1992, 71, 825829. (13) Takarada, T.; Ichinose, S.; Kato, K. Fuel 1992, 71, 883-887. (14) Asami, K.; Ohtsuka, Y. Ind. Eng. Chem. Res. 1993, 32, 16311636. (15) Yamashita, H.; Tomita, A. Ind. Eng. Chem. Res. 1993, 32, 409415. (16) Taghiei, M. M.; Huggins, F. E.; Ganguly, B.; Huffman, G. P. Energy Fuels 1993, 7, 399-405. (17) Otake, Y.; Walker, P. L., Jr. Fuel 1993, 72, 75-80. (18) Taghiei, M. M.; Huggins, F. E.; Mahajan, V.; Huffman, G. P. Energy Fuels 1994, 8, 31-37.

though atomically dispersed catalysts can be obtained by ion-exchanging methods, the catalysts aggregate to large particles with high-temperature treatments during coal conversion processes. Loadings of catalysts, types of species, treatment conditions, and coal structures are related to the aggregation behavior of catalysts. To understand the role of the ion-exchanged metal species in coal conversion processes, it is important to determine the active structure of catalysts under in-situ conditions. X-ray absorption fine structure (XAFS) is a powerful technique to determine the local structure of a specific element in complex systems under any sample conditions. XAFS has been used to examine the chemical structure of catalysts in coal for liquefaction,23-28 gasification,27-31 and pyrolysis.26,27,32 We have reported (19) Murakami, K.; Shirato, H.; Ozaki, J.; Nishiyama, Y. Fuel Process. Technol. 1996, 46, 180-194. (20) Murakami, K.; Shirato, H.; Nishiyama, Y. Fuel 1997, 76, 655661. (21) Murakami, K.; Shirato, H.; Hanada, N.; Nishiyama, Y. Energy Fuels 1998, 12, 843-848. (22) Murakami, K.; Yamada, T.; Fuda, K.; Matsunaga, T.; Nishiyama, Y. Fuel 1997, 76, 1085-1090. (23) Sandstorm, D. R.; Filby, R. H.; Lytle, F. W.; Greegor, R. B. Fuel 1982, 61, 195-197. (24) Zhao, J.; Huggins, F. E.; Zhen, F.; Lu, F.; Shah, N.; Huffman, G. P. J. Catal. 1993, 143, 499-509. (25) Huffman, G. P.; Ganguly, B. G.; Rao, K. R. P. M.; Shah, N.; Feng, Z.; Huggins, F. E.; Taghiei, M. M.; Lu, F.; Wender, I.; Pradham, V. R.; Tierney, J. W.; Seehra, M. S.; Ibrahim, M. M.; Shabtai, J.; Eyring, E. M. Energy Fuels 1993, 7, 285-296. (26) Rao, V. U. S. Energy Fuels 1994, 8, 44-47. (27) Wasserman, S. R.; Winans R. E.; McBeth R. Energy Fuels 1996, 10, 392-400. (28) Shah, N.; Zhao J.; Huggins F. E.; Huffman G. P. Energy Fuels 1996, 10, 417-420. (29) Yamashita, H.; Yoshida, S.; Tomita, A. Ind. Eng. Chem. Res. 1991, 30, 1651-1655. (30) Yamashita, H.; Nomura, M.; Tomita, A. Energy Fuels 1992, 6, 656-661.

10.1021/ef000099v CCC: $19.00 © 2000 American Chemical Society Published on Web 10/21/2000

Hydrogasification of Loy Yang Brown Coal

Energy & Fuels, Vol. 14, No. 6, 2000 1241

Figure 1. Fourier transforms for EXAFS oscillations of nickel-loaded brown coals hydrogasificated at various temperatures; (a) 1.1 wt % nickel-loaded coal, (b) 6.6 wt % nickel-loaded coal, and (c) 12.7 wt % nickel-loaded coal.

that, during coal pyrolysis, ion-exchanged nickel species are reduced to metallic state and that the transition temperature from divalent cations to metal particles changes with the metal loading of a brown coal, Loy Yang coal, by XAFS.33-35 Our objective of the present work is to examine the behavior and properties of nickel species supported on brown coal during hydrogasification. Experimental Section Sample Preparation. Loy Yang brown coal from Victoria, Australia, was used in this study. Details of sample preparation were described previously.36 Raw coal was ground below 250 µm particle size, washed with deionized water, dried at 323 K under vacuum and stored in a desiccator. The analyses for this coal are as follows: C 67.6% (daf), H 5.2% (daf), N 0.8% (daf), O 26.4% (diff), and ash 0.2% (dry). Acid-washed coal was prepared by stirring the raw coal into 0.5 M hydrochloric acid for 24 h. For ion-exchange, a coal sample (10 g) was stirred in an aqueous solution of nickel chloride (500 mL) at a concentration of 10000 ppm and the pH of the solution was adjusted by adding ammonia or hydrochloric acid. After the pH stopped changing and remained unchanged for 5 h, the exchange reaction was judged to be at equilibrium. The quantity of exchanged nickel cations was determined by extracting the cations from the sample by hydrochloric acid. The nickel cation-exchanged samples are identified by the amount of metal loaded (wt %) and gasification temperature, e.g., 1.1 wt %/673 K for 1.1 wt % nickel-loaded coal treated at 673 K. Hydrogasification. Hydrogasification experiment was carried out in a fixed bed type reactor. After samples (1 g) were heated at 373 K for 1 h under 50 mL/min of helium flow, the samples were gasificated under 50 mL/min hydrogen flow at a heating rate of 5 K/min from 373 to 1173 K. The C1 gases evolved were analyzed every 5 min by a gas chromatograph attached to the reactor. XAFS. Details of XAFS analysis were described previously.33 After samples were gasified under 50 mL/min hydrogen at a heating rate of 5 K/min from 373 K to desired temperature, the samples were cooled to room temperature and transferred into EXAFS cells with Kapton windows under flowing hydrogen. The samples were sealed in the cells under vacuum. We measured EXAFS spectra at room temperature to reduce the effect of Debye-Waller factor (thermal disorder) for the exact estimation of coordination numbers of the X-ray absorbing atoms. Data were collected at beam lines BL-7C, 10B, and 12C at the Photon Factory of National Laboratory for High Energy Physics, Tsukuba. The storage ring was operated with an electron energy of 2.5 GeV. Data were recorded in the

transmission mode in the region of the Ni K edge (8331.7 eV). The transmission spectra were collected using ion chambers that were filled with nitrogen gas. EXAFS oscillation was extracted from the EXAFS raw data by using a cubic spline method and normalized with the edge height. The k3-weighted EXAFS spectra were Fourier transformed to R space. The inversely Fourier filtering data were analyzed by a curve-fitting technique on the basis of the singlescattering plane-wave theory.37 Experimentally determined phase shifts and backscattering amplitudes for Ni-Ni and Ni-O were obtained from EXAFS data for Ni foil and NiO, respectively.

Results and Discussion Structure Determination of Nickel Species by XAFS. Figure 1shows several results of EXAFS Fourier transforms of nickel-loaded Loy Yang brown coals treated with hydrogen at different temperatures. All data are presented without correction for phase shift. The Fourier transform for the EXAFS spectra of the 1.1 wt %/673 K sample exhibits only one peak between 1 and 2 Å, which is ascribed to the Ni-O bond (Figure 1a). No peaks were found for the Ni-O-Ni and Ni-Ni bonds. This result shows that atomically dispersed nickel atoms existed in this sample. After gasification at 723 K, a peak assigned to Ni-Ni metal bonds at 2.1 Å was observed in addition to the Ni-O bond peak. With increasing temperature of hydrogasification, the peak intensity of Ni-Ni bond increased, indicating that the nickel species agglomerated. Similar results were also observed with the 6.6 wt %- and 12.7 wt %-nickel-loaded samples as shown in Figure 1b,c. However, the three samples were different in the temperature at which a new peak ascribed to Ni-Ni bond appeared. In other words, the aggregation of nickel species began at dif(31) Cazorla-Amoros, D.; Linares-Solano, A.; Salinas-Martines, C.; Yamashita, H.; Kyotani, T.; Tomita. A.; Nomura, M. Energy Fuels 1993, 7, 139-145. (32) Huggins, F. E.; Shah, N.; Huffman, G. P.; Shoenberger, R. W.; Walker, J. S.; Lytle, F. W.; Greegor, R. B. Fuels 1986, 65, 621-632. (33) Shirai, M.; Murakami, K.; Nishiyama, Y. Energy Fuels 1997, 11, 1012-1018. (34) Shirai, M.; Arai, M.; Murakami, K. Energy Fuels 1999, 13, 465470. (35) Shirai, M.; Murakami, K.; Arai, M. Jpn. J. Appl. Phys. 1999, 38, 77-80. (36) Murakami, K.; Ozaki, J.; Nishiyama, Y. Fuel Process. Technol. 1995, 43, 95-110. (37) Yokoyama, T.; Hamamatsu, H.; Ohta, T. Program EXAFSH, version 2.1; The University of Tokyo: Tokyo, 1994.

1242

Energy & Fuels, Vol. 14, No. 6, 2000

Murakami et al.

Table 1. Curve-fitting Results for EXAFS Data for Nickel Loaded Loy Yang Brown Coala for N ( 20% and R ( 0.02 Å parameter range sample

FT range of k/Å-1

back FT range of R/Å

NiOc 1.1 wt %/573 K 1.1 wt %/623 K 1.1 wt %/673 K 1.1 wt %/723 K 1.1 wt %/748 K 1.1 wt %/773 K 6.6 wt %/573 K 6.6 wt %/623 K 6.6 wt %/673 K 6.6 wt %/723 K 12.7 wt %/523 K 12.7 wt %/573 K 12.7 wt %/623 K 12.7 wt %/673 K 12.7 wt %/723 K

2.0-13.20 2.0-13.20 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45 2.0-12.45

1.4-2.7 1.0-2.0 1.0-2.0 1.0-2.0 1.0-2.0 1.0-2.7 1.0-2.7 1.4-2.7 1.0-2.0 1.0-2.0 1.0-2.7 1.4-2.7 1.0-2.0 1.0-2.0 1.0-2.7 1.4-2.7 1.4-2.7

Nib

Ni-O shell

Ni-Ni shell

curve-fitting range of k/Å-1

N

R/Å

E0/eV

∆σ2/Å2

3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0 3.0-12.0

6 6.5 6.0 5.8 5.6 4.6 nd 6.1 5.9 4.8 nd 5.7 5.6 4.9 nd nd

2.098 2.03 2.01 2.00 2.00 1.99 nd 2.02 2.02 2.03 nd 2.02 2.02 2.03 nd nd

0 -2.0 -2.0 -1.8 0.3 -0.1 nd -2.3 -2.2 1.1 nd -2.2 -0.4 1.5 nd nd

0 0.0029 0.0030 0.0031 0.0058 0.0154 nd 0.0025 0.0024 0.0111 nd 0.0023 0.0027 -0.0026 nd nd

N 12 nd nd nd 1.0 4.7 7.1 nd nd 4.3 6.0 nd nd 3.8 6.6 8.6

R/Å

E0/eV

2.488

0

nd nd nd 2.51 2.49 2.49 nd nd 2.48 2.47 nd nd 2.49 2.48 2.47

nd nd nd -0.9 -13.8 -14.7 nd nd -13.8 -16.6 nd nd -11.1 -14.4 -2.2

∆σ2/Å2 0 nd nd nd -0.0012 0.0026 0.0020 nd nd 0.0028 0.0014 nd nd 0.0016 0.0008 0.0010

a Notation: N, coordination number for absorber-backscatterer pair; R, distance; E , inner potential correction; ∆σ2, differences of 0 Debye-Waller factor from model compounds; nd, not detected. b Reference 39. c Reference 40.

Figure 2. Coordination numbers of (O) Ni-O and (b) Ni-Ni bonds as a function of hydrogasification temperature: (a) 1.1 wt % nickel-loaded coal, (b) 6.6 wt % nickel-loaded coal, and (c) 12.7 wt % nickel-loaded coal.

ferent temperatures depending on the nickel loading. Table 1 summarizes the EXAFS curve-fitting analysis results. The number of parameters used in our analyses satisfied the Nyquist criterion.38 Figure 2 shows the dependence of Ni-O and Ni-Ni coordination numbers of nickel-loaded samples on the hydrogasification temperature. The agglomeration temperature decreased with increasing amount of nickel species. We have previously reported that nickel carboxylate groups are formed by exchanging protons of carboxyl groups of Loy Yang brown coal with nickel cations and nickel carboxylate groups decomposed and the metal aggregated to particles during pyrolysis.34 EXAFS results in the present report also show that nickel metal particles are (38) Iwasawa, Y. X-ray Absorption Fine Structure for Catalysts and Surfaces; World Scientific: Singapore, 1995. (39) Wyckoff, R. W. G. Crystal Structures, 2nd ed.; Wiley: New York, 1963; Vol. 1, p 10. (40) Rooksby, H. P. Acta Crystallogr. 1948, 1. 226-227.

formed in all the ion-exchanged brown coals during hydrogasification, but the transition temperatures for the appearance of nickel particles are different. The transition temperatures of the 1.1 wt %-, 6.6 wt %-, and 12.7 wt %-nickel-loaded samples are 700, 620, and 600 K, respectively (Figure 2). EXAFS analysis showed that the Ni-O coordination number for the 1.1 wt % nickelloaded sample treated at 573 K is about six and decreased gradually with temperature up to 720 K and the Ni-Ni bonds were observed at 720 K (Figure 2). This indicates that most of the nickel carboxylate groups decompose below 600 K and the nickel species diffuse in the coal matrix at temperatures up to near 700 K. Near 700 K, nickel species coalesce together and become larger nickel particles. The Ni-O coordination number for the 6.6 wt % nickel-loaded sample decreased gradually with temperature up to 623 K, and the Ni-Ni bonds were observed at 673 K. Figure 2a indicates that carboxylate groups decompose and nickel atoms coalesce simultaneously near 620 K in the 6.6 wt % nickel-loaded coal. This means that nickel cations coalesce together when carboxylate groups decompose. We reported that carboxylate of the 6.4 wt % nickel-loaded Loy Yang coal has a bridge-type structure, that is, two separate nickel atoms bound to each oxygen atom of a carboxylate group.34 Bridge-type carboxylate groups would be formed in the 6.6 wt %-nickel-loaded sample. Nickel cations coalesces together when bridge-type carboxylate groups decompose. The concentration of nickel carboxylate groups in the 6.6 wt % sample is higher than that in the 1.1 wt % nickel-loaded one. The lower aggregation temperature of the 6.6 wt % nickel-loaded sample than that of the 1.1 wt %-nickel-loaded one can be explained by the structure and coalescence probability of nickel species in the sample. The nickel-divalent cation species with the amount of 12.7 wt % were loaded from an aqueous solution of nickel chloride at pH 7.1. The amount of nickel species loaded is 8 wt % when all carboxylic groups were changed to bridge-type nickel carboxylate groups. Not only nickel carboxylate groups but also nickel hydroxide clusters would exist for the 12.7 wt %-nickel-loaded coal. We did not observe second shell (Ni-O-Ni bond) of

Hydrogasification of Loy Yang Brown Coal

Energy & Fuels, Vol. 14, No. 6, 2000 1243

Figure 4. Total amounts of C1 gases evolved (CO, b; CO2, 9; CH4, 2) against the amount of nickel loaded.

Figure 3. C1 gas evolution profile (CO, solid line; CO2, dotted line; CH4, broken line); (a) acid-washed coal, (b) 1.1 wt % nickel-loaded coal, (c) 4.5 wt % nickel-loaded coal, (d) 6.6 wt % nickel-loaded coal, (e) 9.9 wt % nickel-loaded coal, and (f) 12.7 wt % nickel-loaded coal.

nickel hydroxide in the EXAFS Fourier transform of the 12.7 wt %-nickel-loaded coal hydrogasified at 573 K, indicating that the amount and size of nickel hydroxide clusters would be very small in the sample. The Ni-O coordination number for the 12.7 wt %-nickel-loaded sample treated at 573 K is six and decreases drastically at 623 K. The Ni-Ni bonds were observed at the same temperature (Figure 2). Not only the decomposition of nickel carboxylate groups with a bridge-type structure but also the reduction of nickel hydroxyl oxide clusters would occur in this sample. Catalysis of Nickel Species with Flowing Hydrogen. C1 gas evolution profiles of the acid washed coal and several nickel-loaded samples are given in Figure 3. CO, CO2, and CH4 gases were evolved from these samples and Figure 4 shows the total amounts of C1 gases evolved. The evolution profile of volatile components was affected by large amounts of nickel loaded. Total amount of C1 gases evolved from the 12.7 wt %-nickel-loaded coal was 8 times larger than that of carboxylic groups of the Loy Yang coal. There are some reports that methane yields in hydrogasification of carbonaceous materials were enhanced by adding nickel species;41-43 however, there is no report about the influence of the extent of nickel

loading on the hydrogasification. The CH4 evolution peaks at 860 and 1010 K were broad for the acid washed coal. The position and amount of CH4 peak did not change in the samples loaded with nickel species less than 4.6 wt %. Large amount of CH4 was evolved at 900 K from the 6.6 wt % nickel-loaded sample. The CH4 peak shifted to 820 K and the amount became greater for the 9.9 wt %- and 12.7 wt %-nickel-loaded samples. In addition to the CH4 peak evolved at 820 K, another CH4 evolution peak appeared around 650 K in the samples with nickel species more than 9.9 wt %. EXAFS analysis showed that nickel metal particles were formed at 700 K in the 1.1 wt %-nickel-loaded sample under hydrogen flow. The amount of CH4 evolved from the 1.1 wt %-nickel-loaded sample was similar to that of the acid washed coal under hydrogen flow. Small nickel particles produced at 700 K would not promote the formation of CH4 in this sample. The CH4 peak appeared at 600 K and had a maximum at 900 K in the 6.6 wt %-nickelloaded sample and the amount of CH4 evolved was three times larger than that of the 1.1 wt %-nickel-loaded sample. EXAFS analysis showed that nickel metal particles were formed at 620 K in the 6.6 wt %-nickelloaded sample. Large nickel metal particles formed above 620 K promote the production of CH4 in the 6.6 wt %-nickel-loaded coal. The peak temperature and amount of CH4 evolved from the 12.7 wt %-nickel-loaded coal became lower and larger compared with the 6.6 wt %-nickel-loaded coal. EXAFS analysis showed that nickel metal particles formed at 600 K and large metal particles were formed higher than 600 K in the 12.7 wt %-nickel-loaded sample. The large nickel metal particles produce CH4 from the 12.7 wt %-nickel-loaded sample. The maximum temperature of CO evolved from nickel-loaded coals shifted to 640-700 K, depending on the amount of exchanged nickel species, comparable to a broader peak at 810 K for an acid washed coal. We reported that the exchanged nickel species on Loy Yang brown coal aggregated to metal particles and the peak top temperature of CO evolved during pyrolysis decreased with increasing amount of exchanged nickel species. The peak top temperature of CO evolved is also closely related to the aggregation behavior of nickel species during hydrogasification. The total amount of CO evolved from nickel-loaded coals were slightly smaller than that from the acid washed coal during (41) McKee, D. W. Carbon 1974, 12, 453-464. (42) Tamai, Y.; Watanabe, H.; Tomita, A. Carbon 1977, 15, 103106. (43) Haga, T.; Nishiyama, Y. Carbon 1983, 21, 219-223

1244

Energy & Fuels, Vol. 14, No. 6, 2000

hydrogasification because the CO formed would be hydrogenated to CH4 on nickel metal particles of the nickel-loaded samples. The CO2 evolution peak around 650 K became slightly greater with nickel loadings, but the peak top temperature did not change.

Murakami et al.

amount of exchanged nickel species, become larger with increasing nickel loading. Two peaks of CO evolution appeared at 640-700 K, depending on the amount of exchanged nickel species, and 870 K. The CO2 evolution peak appearing at 660 K became larger slightly with increasing nickel loading but peak top temperature did not change.

Conclusions Hydrogasification profile of Loy Yang brown coal was affected by nickel loadings. Total amount of CO evolved did not change by loading nickel species. The amount of CH4 evolution at 820-1000 K, depending on the

Acknowledgment. This work was carried out under the approval of PF advisory committee (No. 99G252). EF000099V