Decomposition of methane over a Ni-Cu-MgO catalyst to produce hydrogen and carbon nanofibers

Article abstract of DOI:10.1021/jp040496x

We have found that a Ni-Cu-MgO catalyst maintained its activity for the decomposition of methane at high levels for substantially long periods of time at 665-725?°C, being capable of generating large amounts of CO-free H ₂ and solid carbon. TEM

Full text of DOI:10.1021/jp040496x

J. Phys. Chem. B 2004, 108, 20273-20277
20273
Decomposition of Methane over a Ni-Cu-MgO Catalyst to Produce Hydrogen and Carbon
Nanofibers
Haiyou Wang and R. Terry K. Baker*
Catalytic Materials LLC, 1750 Washington St., Holliston, Massachusetts 01746
ReceiVed: July 22, 2004; In Final Form: October 13, 2004
We have found that a Ni-Cu-MgO catalyst maintained its activity for the decomposition of methane at
high levels for substantially long periods of time at 665-725 °C, being capable of generating large amounts
of CO-free H2 and solid carbon. TEM examinations revealed that the solid carbon consisted exclusively of
nanofibers possessing a “platelet” structure, in which the graphite layers are aligned in a direction perpendicular
to the axis of the fiber. In sharp contrast, the Ni-MgO-methane system exhibited negligible activity at
temperatures g650 °C. XRD and H2 chemisorption characterizations indicated that after reduction a Ni-Cu
alloy was formed in the Ni-Cu-MgO catalyst and the surface of the particles was enriched in Cu0. A possible
explanation for the observed superiority of the Ni-Cu-MgO catalyst over that of the Ni-MgO system is
presented.
Introduction
that Cu is capable of enhancing the catalytic activity of Ni, the
precise mechanism whereby this phenomenon occurs is still
open to considerable debate. Two major rationales have been
considered: (a) the modification in the electronic structure that
results in surface properties of the bimetallic being different
than those of the pure metal and (b) the geometric effect in
which ensembles of more than one atom of a given component
are less frequently encountered in the surface of a bimetallic
than in the pure metal.
Recently, the direct catalyzed decomposition of methane has
received great attention as a potential process for the production
of COx (x ) 1 or 2)-free hydrogen.1-14 High-purity hydrogen
is extremely desirable for use in proton exchange membrane
fuel cells (PEMFC). In addition, methane decomposition (reac-
tion 1)
CH4(g) f C(s) + 2H2(g) ∆H298K ) 74.52 kJ/mol (1)
A number of studies have focused on the interaction of
methane with Cu-Ni catalysts.19-21 Bernardo and co-workers19
examined methane decomposition and steam re-forming over
silica supported Cu-Ni catalysts over the range 400-600 °C.
They reported that in the absence of steam, carbon formation
became a major reaction and the deposit took the form of
nanofibers. It was found that the morphology of the nanofibers
changed from “whiskerlike” (a single fiber associated with a
given metal particle) to “octopus” (many fibers from a single
particle) as the Cu content of the catalyst was increased from
10 to 80 atom %. In contrast, Kim and co-workers21 claimed
that methane showed no tendency to decompose when passed
over unsupported Cu-Ni powders at temperatures below 900
°C. It would appear, therefore, that the presence and nature of
the support plays a key role in the Cu-Ni-catalyzed decomposi-
tion of methane.
is mildly endothermic and emits no CO2 to the atmosphere.
Hence, it is more energy efficient and more environmentally
benign than the currently used industrial process, namely, the
steam re-forming of methane, followed by CO shift reac-
tion, CO2 scrubbing, and CO methanation. Furthermore, the
direct decomposition of methane also results in the generation
of a very important byproduct, carbon nanofibers, a material
that has an array of potential applications as electronic com-
ponents, as polymer additives, and as catalysts/catalyst sup-
ports.15
To achieve a high concentration of hydrogen in the effluent
stream, it is necessary to carry out the reaction at an elevated
temperature due to the endothermic nature of methane decom-
position. Unfortunately, rapid deactivation of Ni-based catalysts
ensues at temperatures in excess of 650 °C, leading to a low
yield of hydrogen. It is, therefore, a major challenge to develop
a catalyst system that sustains its activity at high temperatures.
In the present paper, we report highly promising results on
methane decomposition to hydrogen and solid carbon via the
use of a Ni-Cu-MgO catalyst, which maintains its activity at
high temperatures (665-725 °C) and operates under high space
velocity conditions (72 000 mL/h/g of catalyst in terms of the
amount of methane flowing through the reactor).
Experimental Section
Materials. The powdered Ni-Cu substrate catalysts were
prepared from a thick paste formed by vaporizing most of the
solvents included in a slurry consisting of substrate powders
and ethanol solutions of nickel and copper nitrates, which was
dried overnight at 110 °C and then calcined in an air flow (100
mL/min) for 4 h at 500 °C. The oxide mixture was ground to
a fine powder prior to use. The gases used in this work, methane
(99.99%), argon (99.999%), and hydrogen (99.99%), were
obtained from Air Products Inc. Both the hydrogen and the
carrier gas, argon, were purified with Hydro-Purge II and Oxy-
Trap columns prior to use. Reagent grade cupric nitrate [Cu-
(NO3)2·2.5H2O], nickel nitrate [Ni(NO3)2‚6H2O], MgO (surface
The Cu-Ni system has been the most frequently studied
bimetallic catalyst for a variety of reactions and has been the
subject of a number of reviews.16-18 While it is well-established
* Corresponding author. Phone: 508-893-9560. Fax: 508-893-9562.
E-mail: baker@catalyticmaterials.com.
10.1021/jp040496x CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/26/2004

20274 J. Phys. Chem. B, Vol. 108, No. 52, 2004
Wang and Baker
area 106 m2/g), and SiO2 (surface area 200 m2/g) were purchased
from Fisher Scientific for the catalyst preparations.
Apparatus and Procedures. The catalytic reactions were
carried out in a quartz reactor (i.d. 40 mm) heated by a Lindberg
horizontal tube furnace. In a typical experiment, 50.0 mg of
powdered catalyst sample was spread along the base of a
ceramic boat, which was afterward placed at the center of reactor
tube in the furnace. The system was flushed with argon (180
mL/min) for 0.5 h and the catalyst was then reduced in 10%
H2/Ar (200 mL/min) at a temperature between 500 and 1000
°C for 1 h. After reduction, the catalyst was once again flushed
with argon (180 mL/min) for 0.5 h. Following this pretreatment,
pure methane (60 mL/min, which for 50.0 mg of catalyst
corresponds to a space velocity of 72 000 mL/h/g of catalyst)
was introduced into the reactor and allowed to interact with
the catalyst at a set temperature under atmospheric pressure
conditions. The residual feedstock together with gaseous
products was analyzed with on-line gas chromatography to
determine methane conversion and hydrogen concentration in
the effluent. The C and H mass balances were within 5%. The
total amount of solid carbon deposited during the reaction was
determined gravimetrically after the system had been cooled to
room temperature.
Figure 1. Hydrogen (b) and carbon (O) yields as a function of Cu:Ni
molar ratio during methane decomposition over the NixCuyMgzO (x:z
) 2.4:1) catalysts at 665 °C (Treduction ) 850 °C).
TABLE 1: Effect of Reduction Temperature on the
Production of Hydrogen and Solid Carbon during Methane
Decomposition over the NixCuyMgzO (x:y:z ) 2.4:0.6:1)
Catalyst at Different Temperatures
yieldc
(mol/g catal)
Treduction
(°C)
Treaction
(°C)
CH4 conva
(%)
lifetimeb
(h)
H2
carbon
The characteristics of the solid carbon products were deter-
mined from a combination of transmission electron microscopy
(TEM), X-ray diffraction (XRD), and surface area determina-
tions. TEM examinations of the materials were performed in a
JEOL 2000EXII instrument (point-to-point resolution of 0.18
nm). Transmission specimens were prepared by ultrasonic
dispersion of a small quantity of a given sample in isobutanol
and then application of a drop of the suspension onto a holey
carbon film. Size distribution profiles were constructed from
measurements of over 300 particles in each system. A Rigaku
Multiflex X-ray diffractometer with monochromatized Cu KR
radiation was used for the X-ray determinations. The calcined
and reduced catalysts were examined at a scanning rate of 1°/
min within the 2θ values from 34° to 84°. BET surface areas,
calculated from nitrogen adsorption at -196 °C, were carried
out on various materials using a Micromeritics TriStar unit.
Hydrogen chemisorption experiments were carried out at room
temperature in a pulse reactor. During measurement, H2 was
pulsed over the prereduced catalyst until no further adsorption
of H2 was detected. The remaining chemisorbed H2 was
determined quantitatively with a thermal conductivity detector
(TCD).
600
700
850
665
665
665
665
30.7
30.5
30.6
30.3
33
34
43
45
50.2
54.5
71.2
76.0
25.1
27.2
35.6
38.0
1000
600
750
850
700
700
700
700
39.7
39.5
39.6
39.0
24
26
29
33
51.7
54.5
60.8
70.0
25.8
27.2
30.4
35.0
1000
500
600
700
850
1000
725
725
725
725
725
46.4
46.5
45.5
44.5
1.1
19
19
18
18
<1
42.0
43.8
42.2
38.7
∼0
21.0
21.9
21.1
19.3
∼0
600
750
52.8
11
27.8
13.9
a Data obtained after 1 h. b Reaction was terminated when the
methane conversion was below 5%. c Yields were calculated on the
basis of the weight of catalyst before reduction.
while the initial methane conversion (after 1 h on stream)
remained almost constant. At higher reaction temperatures,
g725 °C, a catalyst reduction temperature of 600 °C is required
in order to achieve the highest product yields. Reduction at
temperatures in excess of 600 °C appears to cause a high
deactivation rate (a shorter lifetime). By plotting the initial
methane conversion vs reaction temperature for a constant
reduction temperature (600 or 850 °C), it was found that the
initial conversion exhibited a linear relationship with increasing
reaction temperature. The highest yields of hydrogen (mol of
H2/g of catalyst) and solid carbon (mol of carbon/g of catalyst)
obtained at 665, 700, 725, and 750 °C were 76.0 and 38.0, 70.0
and 35.0, 43.8 and 21.9, and 27.8 and 13.9, respectively. Parts
a and b of Figure 2 show the methane conversion and hydrogen
concentration in the effluent stream, respectively, as a function
of time on stream at different temperatures. It can be seen that,
with exception of reactions performed at 750 °C, both methane
conversion and hydrogen production remained at high levels
for substantially long periods of time. Methane conversion and
H2 concentration were maintained above 30% and 46%,
respectively, for 31 h at 665 °C; above 38% and 55%,
respectively, for 22 h at 700 °C; and above 44% and 61%,
respectively, for 11 h at 725 °C. As a consequence, in addition
to H2, large amounts of solid carbon (its size is significantly
Results
The optimal Ni-Cu-MgO catalyst was found to have a bulk
Ni:Cu:Mg molar ratio of 2.4:0.6:1. Figure 1 shows the yields
of hydrogen and solid carbon as a function of Cu:Ni molar ratio
during methane decomposition over the reduced (Treduction ) 850
°C) NixCuyMgzO (x:z ) 2.4:1) catalysts at 665 °C. In the absence
of Cu, the Ni-MgO catalyst exhibited negligible activity under
these conditions. On the other hand, the yields of both products
increased dramatically as the Cu:Ni ratio was progressively
increased from 1/19 to 1/9 and at a more moderate level from
1/9 to1/4. At higher Cu levels the catalytic activity of the system
exhibited a sharp decline.
The effect of reduction temperature on the activity and
lifetime of the catalyst as well as the yields of hydrogen and
solid carbon over the optimal catalyst composition at various
reaction temperatures is presented in Table 1. For reaction
temperatures e700 °C, both the lifetime and the yields increased
as the reduction temperature was raised from 600 to 1000 °C,

Decomposition of CH4 over a Ni-Cu-MgO Catalyst
J. Phys. Chem. B, Vol. 108, No. 52, 2004 20275
a reduction at 850 °C, the amounts of H2 chemisorbed on the
NixCuyMgzO (x:y:z ) 2.4:0.6:1) and the NixMgzO (x:z ) 2.4:1)
systems were 12.8 and 291.5 µL (STP)/g of catalyst, respec-
tively. XRD patterns of the calcined and reduced NixCuyMgzO
(x:y:z ) 2.4:0.6:1) catalyst samples are presented in Figure 4.
The XRD peaks of the calcined catalyst can be assigned to MgO
and any of the following compounds: NiO, (Ni, Cu)O, (Ni,
Mg)O, and (Ni, Cu, Mg)O (Figure 4a). For the reduced sample,
a Ni-Cu alloy rather than independent Ni and Cu species was
identified and the intensities of the peaks from the Ni-Cu alloy
were seen to strengthen with increasing reduction temperature
(Figure 4b-d). The estimated crystallite size of the Ni-Cu alloy
in the NixCuyMgzO (x:y:z ) 2.4:0.6:1) catalyst, based on the
width of the XRD peak of the (200) face, was found to increase
from about 17 to 35 nm as the reduction temperature was
progressively raised from 600 to 1000 °C (30 nm at 850 °C).
The estimated crystallite size of Ni in the NixMgzO (x:z ) 2.4:
1) catalyst sample after reduction at 850 °C was found to be
about 28 nm. It is significant to find that after a reduction at
850 °C the crystallite sizes of metal particles in both catalysts
are very close, whereas the amount of H2 adsorbed over the
NixMgzO (x:z ) 2.4:1) catalyst is about 23 times greater than
that over the NixCuyMgzO (x:y:z ) 2.4:0.6:1) sample. This
finding suggests that the surface of the latter catalyst is enriched
in Cu0, a claim that is consistent with previous work.18
Discussion
Figure 2. Methane conversion (a) and hydrogen concentration in
effluent stream (b) as a function of time on stream during methane
decomposition over the NixCuyMgzO (x:y:z ) 2.4:0.6:1) catalyst at 665
°C (0), 700 °C (9), 725 °C (O), and 750 °C (b) (Treduction ) 1000 °C
for 665 and 700 °C; Treduction ) 600 °C for 725 and 750 °C).
During the catalytic hydrocarbon decomposition reaction,
methane molecules are initially adsorbed and decomposed on
the metal surface of the catalyst particle, resulting in the
formation of chemisorbed carbon species and the release of
gaseous hydrogen; the carbon species proceed to dissolve and
diffuse through the bulk of the metal particle, ultimately leading
to precipitation of carbon at the rear of the metal particle and
the formation of a nanofiber with an exposed metal particle
remaining at its tip. Deactivation occurs when the rate of carbon
diffusion through the metal catalyst particles is slower than that
of the formation of carbon at the surface Ni0 sites.23 Under these
circumstances, carbon builds up at the catalyst surface and
eventually encapsulates the particle and the growth of nanofibers
ceases.
When the Ni-Cu-MgO catalyst is exposed to a methane
atmosphere, the initial carbon formation rate is not expected to
be rapid, since at this stage there are relatively few exposed
Ni0 sites present at the particle surfaces. As a consequence, the
catalyst can survive during the early stages of the reaction. As
the reaction proceeds, there is an enhancement in the number
of surface Ni0 sites due to the selective and strong interaction
that exists between methane and nickel (copper is inactive for
methane adsorption and decomposition), and a balance is
established between the generation of carbon at these regions
and its bulk diffusion through the particle. As a consequence,
the catalyst is able to maintain its high activity for a substantially
long period of time at a high reaction temperature, and
substantial yields of hydrogen and solid carbon can be realized
(Figures 1 and 2). In sharp contrast, the Ni-MgO catalyst
system undergoes rapid deactivation upon interaction of the
particles with methane due to the presence of a large number
of exposed Ni0 sites, resulting in negligible yields of hydrogen
and solid carbon (Figure 1).
Figure 3. High-resolution TEM photograph of a carbon nanofiber
formed during methane decomposition over the NixCuyMgzO (x:y:z )
2.4:0.6:1) catalyst at 665 °C (Treduction ) 1000 °C).
greater than that of the charged catalyst) were generated during
the steady-state period at temperatures between 665 and 725
°C.
TEM examinations revealed that the solid carbon product
consisted exclusively of nanofibers when methane decomposi-
tion was carried out over the Ni-Cu-MgO catalyst. From
measurements of over 300 nanofibers it was possible to deter-
mine that these materials had an average width of 40 nm and
typical lengths of 5-10 µm. Inspection of the high-resolution
TEM micrograph (Figure 3) shows that the nanofibers possess
a “platelet” structure, in which the graphite layers are aligned
in a direction perpendicular to the growth axis of the fiber.22
In an attempt to gain an insight into the observed superiority
of the Ni-Cu-MgO catalyst over that of the Ni-MgO system,
H2 chemisorption and XRD experiments were carried out. After
It is significant to find that if methane is passed over silica-
supported Cu-Ni or Ni particles at temperatures of about 500
°C, then both systems generate carbon nanofibers.19,24 To
understand the difference in behavior of Cu-Ni and Ni when

20276 J. Phys. Chem. B, Vol. 108, No. 52, 2004
Wang and Baker
Figure 4. XRD patterns of the NixCuyMgzO (x:y:z ) 2.4:0.6:1) catalysts after calcination at 500 °C (a), after reduction at 600 °C (b), after reduction
at 850 °C (c), and after reduction at 1000 °C (d).
TABLE 2: Effect of Substrate on the Production of
Hydrogen and Solid Carbon during Methane Decomposition
over the Ni-Cu (4:1)-Based Catalysts Reduced at 600 °C
and Reacted at 725 °C
the Ni-Cu-MgO system during the calcination step, which
subsequently creates metallic particles of smaller size over MgO
following reduction. We have observed that at a reaction
temperature of 725 °C the Ni-Cu-MgO catalyst reduced at a
temperature higher than 600 °C exhibited shorter lifetimes
(Table 1) and the crystallite size of the Ni-Cu alloy was found
to increase with increasing reduction temperature. Indeed,
estimates based on the width of the XRD peak of the (200)
face indicate that the crystallite sizes of the Ni-Cu alloys in
the Ni-Cu-SiO2 and Ni-Cu-MgO catalysts were 35 and 17
nm, respectively, after a reduction at 600 °C.
yieldc
(mol/g catal)
CH4 conva
(%)
lifetimeb
(h)
substrate
H2
carbon
none
SiO2
MgO
0
38.5
46.5
0
9
19
0
17.2
43.8
0
8.6
21.9
a Data obtained after 1 h. b Reaction was terminated when methane
conversion was below 5%. c Yields were calculated on the basis of the
weight of catalyst before reduction.
The metal crystallite size is a further factor that exerts an
impact on the catalyst stability. It is generally agreed that
diffusion of carbon through the catalyst particle is the rate-
determining step in the growth of carbon nanofibers.23 At a given
temperature, the rate of nanofiber elongation is inversely
proportional to the square of the particle size.25 On the other
hand, for a given sized metal particle, the rate of carbon diffusion
increases exponentially with increasing reaction temperature.
Consequently, the diffusion of carbon through the metal particle
is affected not only by the reaction temperature but also by the
size of metal particles and so is the stability (lifetime) of the
catalyst. It is of interest to contrast this behavior with that found
for the growth of carbon nanotubes, where the surface diffusion
of carbon around the catalyst particle is purported to be the
rate-controlling step.26 In this context, it should be appreciated
that the latter structures are believed to be hollow, whereas the
“platelet” graphite nanofibers are completely solid materials.
Indeed, as shown in Table 1, for a constant reduction temper-
supported on SiO2 or MgO with regard to the reaction with
methane, it is necessary to examine the nature of respective
metal-support interactions.
The choice of the MgO component in the catalyst was another
critical aspect. As seen from the data presented in Table 2, at
a reaction temperature of 725 °C, in the absence of MgO, the
NixCuy (x:y ) 4:1) catalyst provided no activity for methane
conversion, indicating the necessity of the presence of MgO as
both a chemical component and dispersing agent in the catalyst.
Furthermore, as seen in Table 2, the substitution of MgO by
SiO2 resulted in formation of a catalyst that exhibited a lower
initial methane conversion level and a shorter lifetime with the
concomitant generation of lower amounts of hydrogen and solid
carbon. The superior performance obtained with a catalyst
containing MgO is most likely related to the tendency to form
less reducible (Ni, Mg)O and (Ni, Cu, Mg)O solid solutions in
TABLE 3: Comparison of the Physical and Structural Characteristics of “Platelet” Carbon Nanofibers Produced from
Different Catalytic Reactionsa
space velocity
(mL/h/g of catalyst)
GNF yield
(g of C/g of catalyst)
average width
(nm)
d-spacing
(nm)
surface area
(m2/g)
catalyst
feedstock
CH4
NixCuyMgzO
72 000
381
38.0
0.3389
216
(x:y:z ) 2.4:0.8:1)
Cu-Fe (3:7)
CO/H2 (4:1)
240 000
42
110.0
0.3359
76
a Treaction ) 665 °C.

Decomposition of CH4 over a Ni-Cu-MgO Catalyst
J. Phys. Chem. B, Vol. 108, No. 52, 2004 20277
ature the lifetime of the Ni-Cu-MgO catalyst was strongly
dependent upon the reaction temperature. In addition, XRD
measurements indicated that the crystallite size of Ni-Cu alloy
particles in the Ni-Cu-MgO catalyst increased as a function
of reduction temperature. This aspect accounts for the finding
that for a constant reaction temperature the stability of the
catalyst was also strongly influenced by the reduction temper-
ature (Table 1). Furthermore, the fraction of surface Ni0 sites
in the Ni-Cu-MgO catalyst is enhanced during reaction and
the final composition of the particle surface is dependent upon
the reaction temperature. Therefore, methane conversion is not
only affected by the reaction temperature but also by the ultimate
number of surface Ni0 sites that the hydrocarbon encounters
during adsorption on the particle surface. This rationale accounts
for why the initial methane conversion did not exhibit an
exponential increase with reaction temperature for a constant
reduction temperature (Table 1).
A further fascinating aspect that emerges from this investiga-
tion concerns the structural characteristics of the carbon nano-
fibers generated from the Ni-Cu-MgO/CH4 reaction. The
finding that the nanofibers adopt a “platelet” conformation is
totally unexpected, since the interaction of this particular
bimetallic composition with other hydrocarbons always leads
to the formation of “herringbone” structures.21,27 Indeed, the
formation of “platelet” nanofiber structures has generally been
reported from reactions where CO has been passed over an iron-
containing catalyst.23-30 It would appear that the chemistry
involved in the interaction of the Ni-Cu-MgO system creates
metal particles that adopt specific crystallographic orientations
that favor the precipitation of carbon in the form of graphite
sheets that are aligned in a direction perpendicular to the
longitudinal axis of the fiber.
adjacent graphite layers increases. The smaller dimensions of
the nanofibers generated from the new catalyst system is also
reflected in a larger value of the surface area.
References and Notes
(1) Pourier, M. G.; Sapundzhiev, C. Int. J. Hydrogen Energy 1997,
22, 429.
(2) Zhang, T.; Amiridis, M. D. Appl. Catal. A: Gen. 1998, 167, 161.
(3) Muradov, N. Z. Energy Fuels 1998, 12, 41.
(4) Aiello, R.; Fiscus, J. E.; zur Loye, H.-C.; Amiridis, M. D. Appl.
Catal. A: Gen. 2000, 192, 227.
(5) Ermakova, M. A.; Ermakov, D. Yu.; Kuvshinov, G. G. Appl. Catal.
A: Gen. 2000, 201, 61.
(6) Li, Y.; Chen, J.; Qin, Y.; Chang, L. Energy Fuels 2000, 14, 1188.
(7) Muradov, N. Int. J. Hydrogen Energy 2001, 26, 1165.
(8) Choudhary, T. V.; Sivadinarayana, C.; Chusuei, C. C.; Klinghoffer,
A.; Goodman, D. W. J. Catal. 2001, 199, 9.
(9) Takenaka, S.; Ogihara, H.; Yamanaka, I.; Otsuka, K. Appl. Catal.
A: Gen. 2001, 217, 101.
(10) Shah, N.; Panjala, D.; Huffman, G. P. Energy Fuels 2001, 15, 1528.
(11) Piao, L.; Li, Y.; Chen, J.; Chang, L. J.; Lin, Y. S. Catal. Today
2002, 74, 145.
(12) Ermakova, M. A.; Ermakov, D. Yu. Catal. Today 2002, 77, 225.
(13) Takenaka, S.; Shigeta, Y.; Otsuka, K. Chem. Lett. 2003, 32, 26.
(14) Villacampa, J. I.; Royo, C.; Romeo, E.; Montoya, J. A.; Del Angel,
P.; MonzOÄ n, A. Appl. Catal. A: Gen. 2003, 252, 363.
(15) De Jong, K. P.; Geus, J. W. Catal. ReV.-Sci. Eng. 2000, 42, 481.
(16) Khulbe, K. C.; Mann, R. S. Catal. ReV.-Sci. Eng. 1982, 24, 311.
(17) Ponec, V. Int. Quantum Chem. 1977, 12, 1.
(18) Sinfelt, J. H. Acc. Chem. Res. 1977, 10, 15.
(19) Bernardo, C. A.; Altrup. I.; Rostrup-Nielsen, J. R. J. Catal. 1985,
96, 517.
(20) Tavares, M. T.; Bernardo, C. A.; Altrup, I.; Rostrup-Nielsen, J. R.
J. Catal. 1986, 100, 545.
(21) Kim, M. S.; Rodriguez, N. M.; Baker, R. T. K. J. Catal. 1991,
131, 60.
(22) Rodriguez, N. M.; Chambers, A.; Baker, R. T. K. Langmuir 1995,
11, 3862.
(23) Baker, R. T. K.; Barber, M. A.; Harris, P. S.; Feates, F. S.; Waite,
R. J. J. Catal. 1972, 26, 51.
(24) Rostrup-Nielsen, J. R. J. Catal. 1972, 10, 221.
(25) Baker, R. T. K. In Gas Chemistry in Nuclear Reactors and Large
Industrial Plant; Dyer, A., Ed.; Heyden: 1980; p 18.
(26) Helveg, S.; Lopez-Cartes, C.; Sehested, J.; Hansen, P. L.; Clausen,
B. S.; Rostrup-Nielsen, J. R.; Abild-Pedersen, F.; Norskov, J. K. Nature
2004, 427, 426.
(27) Marotta, C. L.; Baker, R. T. K. J. Mol. Catal. A: Chem. 2003,
195, 209.
(28) Murayama, H.; Maeda, T. Nature 1990, 345, 791.
(29) Rodriguez, N. M.; Kim, M. S.; Baker, R. T. K. J. Catal. 1993,
144, 93.
A comparison is given in Table 3 of the respective yields
and physical and structural characteristics of “platelet” graphite
nanofibers grown from the decomposition of CH4 over NiX-
CuyMgzO (x:y:z ) 2.4:0.8:1) at 665 °C with similar materials
synthesized from the interaction of Cu-Fe (3:7) with CO/H2
at the same temperature.30 Inspection of these data reveals that
by using the new catalyst system one can synthesize “platelet”
graphite nanofibers having a significantly narrower width than
those grown from the traditional catalyst. It is also evident that
the van der Waals forces are weaker as the width of the
structures decreases, and as a consequence, the spacing between
(30) Carneiro, O. C.; Kim, M. S.; Yim, J. B.; Rodriguez, N. M.; Baker,
R. T. K. J. Phys. Chem. B 2003, 107, 4237.