Graphitic nanofilaments as novel support of Ru-Ba catalysts for ammonia synthesis

Article abstract of DOI:10.1016/S0021-9517(02)93724-3

A novel graphitic-nanofilament- (GNF-) supported Ru-Ba catalyst is prepared and used in ammonia synthesis reaction. The Ru-Ba/GNFs catalyst shows remarkably high activity and stability for ammonia synthesis, which can be attributed to high purity and graphitization of GNFs with unique structure. TEM micrographs of the Ru-Ba/GNFs catalysts show that Ru metal particles uniformly disperse on the outer wall of GNFs, and the particles become bigger than that before ammonia synthesis reaction after 50 h of operation at 500°C and 7.0 MPa, probably due to the Ru crystals covered by promoter and support materials and/or sintering of Ru crystals.

Full text of DOI:10.1016/S0021-9517(02)93724-3

Journal of Catalysis 211, 278–282 (2002)
doi:10.1006/jcat.2002.3724
RESEARCH NOTE
Graphitic Nanofilaments as Novel Support of Ru–Ba Catalysts
for Ammonia Synthesis
∗
∗
Changhai Liang,∗,1 Zhonglai Li, Jieshan Qiu,† and Can Li
∗
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, People’s Republic of China;
and †Department of Materials and Chemical Engineering, Dalian University of Technology, Dalian 116012, People’s Republic of China
Received April 2, 2002; revised May 22, 2002; accepted June 13, 2002
with various supports and promoters have been studied in
both fundamental understanding and industrial application
by some groups, including Aika’s group (5, 6) in Japan,
Tennison’s group (7) at British Petroleum, Muhler’s group
(8, 9) in Germany, Kowalczyk’s group (10, 11) in Poland,
Forni’s group (12) in Italy, groups (13, 14) in Denmark, and
others (15–20). Finally, the promoted ruthenium supported
on graphite-containing carbon with high-surface-area cata-
lyst was introduced to produce ammonia from dinitrogen
and dihydrogen on an industrial scale (21).
A novel graphitic-nanofilament- (GNF-) supported Ru–Ba cata-
lyst is prepared and used in ammonia synthesis reaction. The Ru–
Ba/GNFs catalyst shows remarkably high activity and stability for
ammonia synthesis, which can be attributed to high purity and
graphitization of GNFs with unique structure. TEM micrographs
of the Ru–Ba/GNFs catalysts show that Ru metal particles uni-
formly disperse on the outer wall of GNFs, and the particles be-
come bigger than that before ammonia synthesis reaction after 50 h
of operation at 500◦C and 7.0 MPa, probably due to the Ru crystals
covered by promoter and support materials and/or sintering of Ru
A number of studies have shown that carbon-supported
catalysts are more active than the oxide-supported cata-
lysts. High-temperature treatment of carbon support leads
to graphitization, which was considered to be responsible
for the high activity of carbon-supported Ru catalysts (10–
12). At the same time, high-temperature treatment can also
decrease the methanation reaction even at above 700◦C and
100 bar (12), which is more severe than the conventional
reaction conditions on Ru-based catalysts. Recently, Chen
et al. have tested the activity of K–Ru supported on mul-
tiwalled nanotubes for ammonia synthesis and found the
catalyst is more active than other supports (20). In this pa-
per, graphitic nanofilaments (GNFs) with high surface area
are used as support of Ru–Ba catalysts for ammonia syn-
thesis. It is found that the Ru–Ba/GNF catalyst is more
active than the activated-carbon- (AC-) supported Ru–Ba
catalyst. Furthermore, the Ru–Ba/GNF catalyst does not
need high Ba/Ru atomic ratio, which is associated with the
AC-supported catalysts.
c
crystals. ꢀ 2002 Elsevier Science (USA)
KeyWords:graphiticnanofilaments;barium;ammoniasynthesis;
ruthenium.
1. INTRODUCTION
Catalytic synthesis of ammonia is one of the most impor-
tant industrial processes in which a promoted iron cata-
lyst is widely used at high temperatures and high pres-
sures (1). The process under such reaction conditions is
energy-intensive as well as capital-intensive although some
advances have been made. Therefore, ammonia synthesis
process with low energy consumption has been a longstand-
ing target in chemical industry. One of the keys to reaching
this target is to develop catalysts with high activity under
milder conditions. The mechanism of the reaction on the
promoted iron catalysts has been studied extensively, and a
dissociative mechanism where dinitrogen and dihydrogen
are broken up into adsorbed nitrogen and hydrogen with
successive recombination steps is accepted (2). Ru-based
catalysts for ammonia synthesis have also been studied in
the early investigation carried out by Mittasch (3), but supe-
rior activity was not found. Ru-based catalysts were found
to be the most active catalyst for ammonia synthesis by
Aika et al. (4). Since then, a number of ruthenium catalysts
2. EXPERIMENTAL
GNFs were synthesized by CH4 decomposition over a
nickel-based catalyst, then the catalyst was removed by dis-
solution typically in aqueous solution of HNO3. The Ru–
Ba/GNF and Ru–Ba/AC catalysts were prepared by using a
two-step impregnation method (18). Namely, The Ru pre-
cursor salt, RuCl3, was dissolved into acetone and the GNFs
1 To whom correspondence should be addressed. Fax: 86-411-4694447.
E-mail: chliang@dicp.ac.cn.
278
0021-9517/02 $35.00
c
ꢀ 2002 Elsevier Science (USA)
All rights reserved.

GRAPHITIC NANOFILAMENTS AS SUPPORT OF CATALYSTS
279
were added to form slurry. The sample was dried overnight ters were in the range of 5–10 nm. The BET surface area of
at 110◦C, reduced at 400◦C (until no chlorine ions were the GNFs, determined by nitrogen adsorption at 77 K, was
detected with the solution of AgNO3), and passivated in about 140 m2/g.
1% O2 for 2 h at room temperature. The as-prepared sam-
Figure 2 shows the ammonia synthesis activity of the
ple was impregnated with the solution of Ba(NO3)2, then aforementioned catalysts changing with Ba/Ru atomic
dried overnight at 110◦C. Ru–Ba/AC catalysts were pre- ratio. The improved activity of ammonia synthesis on Ru–
pared with the same methods.
Ba/GNF relative to that on Ru–Ba/AC is evident. The
Catalytic reactions of ammonia synthesis were carried ammonia synthesis activity of the Ru–Ba/GNF catalyst in-
out in a fixed-bed stainless-steel microreactor with a stoi- creases when the Ba/Ru atomic ratio is below 0.25, but de-
chiometric H2 and N2 mixture (H2/N2 = 3) flow at 3.0 MPa. creases with further increasing of the amount of promoter,
The purity of H2 or N2 gases is over 99.99%, and the mix- while that of the Ru–Ba/AC catalyst increases gradually in
ture gases were further purified before the reaction by self- the range of the Ba/Ru atomic ratio studied. This behav-
designed guard containers packed with palladium catalyst ior may be attributed to various reasons, including (a) de-
and molecular sieves. The catalyst in the reactor was acti- creased impurity in GNF support, (b) increased electronic
vated in the stream of N2 + 3H2 according to the following conductance of GNFs when compared to the activated car-
temperature program: heating to 450◦C in 100 min, stand- bon, and (c) the attainment of the sites with high activity
ing at 450◦C for 240 min, and then cooling to the reaction because of specific interaction between Ru and GNFs.
temperature in 30 min. The amount of ammonia in the ef-
It is suggested that the surface impurities from carbon
fluent was determined by a chemical titration method. A materials such as S, N, O, and Cl can suppress the catalytic
well-dispersion 4.0 wt% Ru/AC catalyst was used as the activity and consume part of promoter (18, 22). Element
reference catalyst.
analysis shows that AC contains some impurities such as
Transmission electron microscopy (TEM) studies were S (0.15%), N (0.04%), O (7.75%), and Cl (0.10%), while
carried out using a JEOL JEM-2000EX electron micro- GNFs only contain traces of oxygen (1.22%) and hydrogen
scope with 100 kV.
(0.02%) besides carbon (98.76%). The highest rate of am-
monia synthesis is reached when the Ba/Ru atomic ratio is
about 0.25, which is much lower than the optimum Ba/Ru
on graphite support. This indicates that the nature of GNFs,
3. RESULTS AND DISCUSSION
TEM micrographs of the GNF material are shown in such as structural and electronic properties, plays an impor-
Fig. 1. It can be seen that the GNF sample does not contain tant role on the catalytic activity for ammonia synthesis.
the residues such as Ni catalyst and its support, and few car-
FortheammoniasynthesisontheRucatalysts, it hasbeen
bonaceous impurities agglomerate. The outer diameters of shown that the B5-type sites dominate the activity of the
the GNFs were in the range of 20–60 nm, and inner diame- catalyst (13, 14). In the case of the Ru–Ba/GNF catalyst, the
FIG. 1. TEM micrographs of the GNF samples.

280
LIANG ET AL.
FIG. 3. Activity of Ru–Ba/GNF catalyst with the Ba/Ru = 0.25 and
FIG. 2. Effects of Ba/Ru atomic ratio on ammonia synthesis activity
of the Ru–Ba/AC and Ru–Ba/GNF catalysts.
Ru–Ba/AC catalyst with Ba/Ru = 1.0 at temperatures from 350 to 420◦C.
The inlet gas does not contain ammonia and the flow rate is varied to
obtain an exit concentration of ammonia of 2.0%.
B5-type sites are formed more easily than in the Ru–Ba/AC
catalyst as a result of the specific interaction between metal
and support. However, the nature of this interaction is
unclear. In the case of cinnamaldehyde hydrogenation on
the Ru supported on carbon nanotubes catalyst, a specific
interaction between metal and support was suggested to ex-
plainthehighselectivityandhighconversion(23). Recently,
the GNF-supported platinum catalysts showed improved
oxidation activity for methanol when compared to Vulcan
carbon. Theimprovementisbelievedtobeduetometalpar-
ticles adopting specific crystallographic orientations when
dispersed on GNFs (24). Therefore, it is believable that the
support, where the crystallographic orientations of Ru are
determined, may play an important role in the catalytic ac-
tivity of ammonia synthesis.
der those conditions, while the Ru–Ba/AC catalyst shows
severe deactivation under similar conditions. It is evident
that the degree of GNF graphitization is responsible for
the catalyst stability. The graphitization of GNFs is higher
thantheactivatedcarbonfromXRD. Thed(002)diffraction
peak of graphite was observed in the GNF sample, while no
obvious diffraction peak of graphite (002) was observed for
the AC samples. At the same time, GNFs with high elec-
trical conductivity (26) also offer possibility for electron
transfer from promoters to Ru metal since most promoter
and Ru particles are not in contact with each other.
Figure 3 compares the ammonia synthesis activities of
the Ru–Ba/GNF catalyst with Ba/Ru = 0.25 and the Ru–
Ba/AC catalyst with Ba/Ru = 1.0 at temperatures from 350
to 420◦C. The activity of the Ru–Ba/GNF catalyst is re-
markably higher than that of the Ru–Ba/AC catalysts un-
der the same conditions. The activation energy of ammo-
nia synthesis on the Ru–Ba/GNF catalyst at constant flow
is about 84 kJ/mol, which is close to that reported in the
literature (4–6, 8, 9). The pressure dependence of the am-
monia synthesis activity of Ru–Ba/GNF catalysts is shown
in Fig. 4. The reaction kinetics of the Ru–Ba/GNF catalysts
are similar to those reported for Ba-promoted ruthenium
catalysts supported on other carbon materials and BN (11,
18, 25). The stability of the Ru–Ba/GNF catalyst is also
studied under continuous reaction for 50 h at 500◦C and
7.0 MPa with a mixture of 3H2 + N2. During this process,
the ammonia concentration in the exit reaches the equi-
FIG. 4. Pressure dependence of the ammonia synthesis activity of
libration value. No obvious deactivation is observed un- Ru–Ba/GNF catalysts at 400◦C.

GRAPHITIC NANOFILAMENTS AS SUPPORT OF CATALYSTS
281
FIG. 5. Representative transmission electron micrographs of (a) the Ru/GNF sample and (b) the spent Ru–Ba/GNF ammonia synthesis catalyst.
Representative transmission electron micrographs of the Ba/GNF catalyst. The catalyst particles become bigger after
Ru/GNF sample and the Ru–Ba/GNF catalyst after ammo- reaction, probably because the Ru crystals are completely
nia synthesis reaction are shown in Fig. 5. In the case of covered by promoter or support materials and/or sintered.
Ru/GNFs, high dispersion of Ru metal particles with a nar- Therefore, rational design of catalysts and experiments can
rowsizedistributionof2–4nmwereseenasshowninFig. 5a, lead to a high active catalyst based on GNFs as support for
and Ru particles mainly deposited in the outer surface of ammonia synthesis.
the GNFs; the result is similar to that of Pt/GNFs (24, 27).
The Ru crystals dispersed on the GNFs were relatively thin,
with a highly crystalline faceted structure, while Ru parti-
cles supported on the activated carbon adopted a globular
morphology. This difference may be due to the interaction
between Ru and support. It is the interaction that mainly
results to high activity for ammonia synthesis on the Ru–
Ba/GNF catalyst. Fig. 5b is a representative ex situ trans-
mission electron micrograph of a Ru–Ba/GNF catalyst after
50 h of operation at 500◦C and 7.0 MPa. It can be seen that
Ru crystals and Ba compounds are about 4–7 nm and local-
ized primarily in the outer surface of the GNFs. It should
be noted that the particles become bigger after reaction,
probably because the Ru crystals are completely covered
by promoter or support materials, as shown by Hansen et al.
(28), and/or sintered.
REFERENCES
1. Jennings, J. R., “Catalytic Ammonia Synthesis.” Plenum, New York,
1991.
2. Schlo¨gl, R., in “Handbook of Heterogeneous Catalysis” (G. Ertl,
H. Kno¨zinger, and J. Weitkamp, Eds.), pp. 1697–1748. Wiley-VCH,
Weinheim, 1997.
3. Mittasch, A., Adv. Catal. 2, 81 (1950).
4. Aika, K., Hori, H., and Ozaki, A., J. Catal. 27, 424 (1972).
5. Aika, K., Angew. Chem., Int. Ed. Eng. 25, 558 (1986).
6. Izumi, Y., and Aika, K., J. Phys. Chem. 99, 10336 (1995).
7. Tennison, S. R., in “Catalytic Ammonia Synthesis” (J. R. Jennings,
Ed.), Chap. 9. Plenum, New York, 1991.
8. Bielawa, H., Hinrichsen, O., Birkner, A., and Muhler, M., Angew.
Chem., Int. Ed. Eng. 40, 1061 (2001).
9. Hinrichsen, O., Rosowski, F., Hornung, A., Muhler, M., and Ertl, G.,
J. Catal. 165, 33 (1997).
10. Rarog, W., Kowalczyk, Z., Sentek, J., Skladanowski, D., and
Zielinski, J., Catal. Lett. 68, 163 (2000).
4. CONCLUSION
11. Kowalczyk, Z., Jodzis, S., Rarog, W., Zielinski, J., Pielaszek, J., and
Presz, A., Appl. Catal., A 184, 95 (1999).
12. Rossetti, I., Perniconi, N., andForni, R., Appl. Catal., A208, 271(2001).
13. Jacobsen, C. J. H., Dahl, S., Hansen, P. L., To¨rnqvist, E., Jensen, L.,
Topsøe, H., Prip, D. V., Mosenshaug, P. B., and Chorkendorff, I.,
J. Mol. Catal. A 163, 19 (2000).
14. Dahl, S., Logadottir, A., Egeberg, R. C., Larsen, J. H., Chorkendorff, I.,
To¨rnqvist, E., and Nørskov, J. K., Phys. Rev. Lett. 83, 1814 (1999).
The Ru–Ba/GNF catalyst shows remarkably high activ-
ities for ammonia synthesis, which are higher than that on
the Ru–Ba/AC catalyst. The high purity, high graphitiza-
tion, and unique structure of GNFs are likely to be respon-
sible for the high activity for ammonia synthesis on the Ru–

282
LIANG ET AL.
15. McClaine, B. C., Becue, T., Lock, C., and Davis, R. J., J. Mol. Catal. A
163, 105 (2000).
16. Cisneros, M. D., and Lunford, J. H., J. Catal. 141, 191 (1993).
17. Rao, K. S. R., Masthan, S. K., Prasad, P. S. S., and Rao, P. K., Appl.
Catal. 73, L1 (1991).
22. Zhong, Z., and Aika, K., Chem. Commun. 13, 1223 (1997).
23. Planeix, J. M., Coustel, N., Coq, B., Brotons, V., Kumbhar, P. S.,
Dutartre, R., Geneste, P., Bernier, P., and Aiayan, P. M., J. Am. Chem.
Soc. 116, 7935 (1994).
24. Bessel, C. A., Laubernds, K., Rogriguez, N. M., and Baker, R. T. K.,
J. Phys. Chem. B 105, 1115 (2001).
18. Liang, C. H., Wei, Z. B., Xin, Q., and Li, C., Appl. Catal., A 208, 197
(2001).
25. Jacobsen, C. J. H., J. Catal. 200, 1 (2001).
19. Liang, C. H., Wei, Z. B., Luo, M. F., Ying, P. L., Xin, Q., and Li, C.,
Stud. Surf. Sci. Catal. 138, 283 (2001).
20. Chen, H. B., Lin, J. D., Cai, J., Wang, X. Y., Yi, J., Wang, J., Wei, G.,
Lin, Y. Z., and Liao, D. W., Appl. Surf. Sci. 180, 328 (2001).
21. Rhodes, A. K., Oil Gas J. 11, 37 (1996).
26. De Jong, K. P., and Geus, J. W., Catal. Rev. Sci. Eng. 42, 481 (2000).
27. Li, W. Z., Liang, C. H., Qiu, J. S., Zhou, W. J., Han, H. M., Wei, Z. B.,
Sun, G. Q., and Xin, Q., Carbon 40(5), 791 (2002).
28. Hansen, T. W., Wagner, J. B., Hansen, P. L., Dahl, S., Topsøe, H., and
Jacobsen, C. J. H., Science 294, 1508 (2002).