Académique Documents
Professionnel Documents
Culture Documents
www.elsevier.com/locate/cattod
Abstract
Pure hydrogen was obtained by catalytic methane decomposition over thermally stable nickel-based catalysts prepared from hydrotalcite-like
precursors. Two different Ni–Mg–Al catalyst compositions were tested in the catalytic pyrolysis reaction of methane. The catalytic results were
compared with those of a catalyst prepared by conventional impregnation of Ni–Mg phases on an alumina support. The high thermal stability of
these catalysts was demonstrated in several reaction–regeneration cycles. The regeneration processes were performed with CO2 and O2 as
regenerating agents. Using either CO2 or O2 gas, hydrogen production after regeneration increased with regard to the fresh catalyst. Upon catalyst
regeneration, larger amounts of hydrogen were obtained on catalysts regenerated with CO2 than with O2. The high thermal stability of the Ni
particles of the catalysts from hydrotalcite-like precursors was improved in the presence of Mg. This high Ni-stability is a key factor for increasing
hydrogen production after consecutive reaction–regeneration cycles.
# 2006 Elsevier B.V. All rights reserved.
Keywords: Methane pyrolysis; Nickel catalysts; Hydrotalcite precursors; Regeneration
1. Introduction specifically of natural gas [3,4], is currently the largest and most
economic way to produce hydrogen. Alternative ways include
The environmental pollution produced by the delivery of the partial oxidation of methane [3,5], the autothermal methane
massive energy through the combustion of fossil fuels has made oxi-reforming reaction [6] and the dry reforming reaction of
it necessary to develop new options based on sustainable methane [3,7].
sources. As an energy carrier, hydrogen (H2) is an attractive A different alternative route for hydrogen production is
choice, since it affords a clean energy when used either directly catalytic methane decomposition: so-called methane pyrolysis
or as fuel in H2-air fuel cells. Both concepts are considered as or methane decarbonization [3,8] (1).
offering the major sources of clean energy in future decades [1].
Apart from these advantages, there are currently two major CH4 ! C þ 2H2 ; DH 0 ¼ þ90:1 kJ=mol (1)
hurdles to be overcome in order to give hydrogen a major role in This reaction has received much attention in recent years [9],
the energy sector: (i) existing hydrocarbon fuels are readily because its most important characteristic is the production of
available and cheap, (ii) hydrogen, which is not really a primary H2 free of CO or CO2 and the only by-product of this reaction is
fuel, must be produced from other primary sources with no a carbon material or a carbonaceous species [10]. The CO–
significant cost in energy or pollution. In order to meet H2 CO2-free H2 produced is optimal for direct supply to a fuel cell,
energy needs worldwide, its production must be considered whose electrocatalysts are highly sensitive to the presence of
within the framework of all resources available, including fossil CO at ppm levels in the hydrogen stream [11].
fuels, in the near future of the so-called ‘‘hydrogen economy’’ It is well documented that nickel catalysts catalyze the
[2]. In this way, the steam reforming of hydrocarbons [3], decomposition of methane into hydrogen and carbon as
C-nanofibers [12,13]. The main problem of using Ni-catalysts
is deactivation of the catalyst through the carbon deposits on the
* Corresponding author. Tel.: +34 91 585 4759; fax: +34 91 585 4760. catalyst surface, over the Ni active centre [14]. Catalyst
E-mail address: rut.guil@icp.csic.es (R. Guil-López). deactivation may be overcome by regeneration of the catalyst.
0920-5861/$ – see front matter # 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2006.05.078
290 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 290
This carbon is in fact highly reactive to O2 [15] (2) or CO2 [16] 2. Experimental
(3):
2.1. Catalyst preparation
C þ O2 ! CO2 ; DH 0 ¼ 394:7 kJ=mol (2)
The hydrotalcite-like precursors described in this study were
C þ CO2 ! 2CO; DH 0 ¼ þ174:5 kJ=mol (3) prepared by co-precipitation by urea hydrolysis [21]. An
aqueous solution (0.5 M) containing appropriate amounts of
The use of CO2 as a regenerating agent (reaction (3)) allows NiCl2 6H2O (active metal), MgCl2 6H2O (modifying divalent
collection of the CO product, which can be considered as a cation), and AlCl
P 3 6H2O in the presence of urea (urea
useful intermediate chemical for carbonylation reactions of quantity = 3 mol of cations) as the precipitating agent
methanol and acetylene, the hydroformylation of alkene, the were heated, controlling the temperature to obtain the desired
synthesis of phosgene or the formation of metal carbonyls [16]. precipitation pH value. The precipitate was aged in suspension
On the other hand, the need for a regeneration process makes at 90 8C for 48 h under stirring. The resulting solid was filtered,
it necessary to use Ni-catalysts in which Ni crystallites have thoroughly washed with distilled water – until the filtered water
been stabilized in order to avoid Ni sintering during the pH was around 6 – and dried overnight at 80 8C. All precursors
regeneration process. An alternative option for stabilising contained both chloride (Cl ) and carbonate (CO32 ) as charge
active metal particles is the use of different precursors. The so- compensating anions. Chemicals and reagents were purchased
called hydrotalcite-like compounds seem to be good precursor from Aldrich.
candidates for preparing stabilized Ni-catalysts because they Two different samples, depending on nominal composition,
can afford highly thermally stabilized mixed oxides upon were obtained with two different Ni-loadings: (i) high Ni-
treatment at high temperature (calcination treatment) [17]. loading (nominal composition in the molar ratio of cations:
Hydrotalcite-like compounds are layered double hydroxides 58.0% Ni–15.0% Mg) NMA-I, (ii) low Ni-loading (18.0% Ni–
(LDHs) derived from brucite layers with an ideal formula of 46.0% Mg) NMA-II. The two LDH-precursors were calcined to
2+
[M(1 x) Mx3+ (OH)2] [A n] x/n m H2O; in which M2+ and M3+ obtain the corresponding mixed oxides under a flow of air in a
can be any divalent and trivalent metal cation capable of prepared by classical impregnation of an alumina substrate
occupying the octahedral interstitial sites of a brucite-like sheet, modified by magnesium.
which are positively charged due to partial substitution of the
magnesium by trivalent cations. A n is almost all classical
balancing exchange anion (NO3 , CO32 , Cl ), amongst the
others [18]. Upon heating, these highly ordered hydrotalcite-
like compounds form an amorphous mixture of metal oxides –
M2+M3+(O) – with a very small crystal size that are stable under
thermal treatment. These mixed oxides show interesting
properties, such as a high specific surface area, high active
metal dispersion, a strong interaction between different metals
in mixed oxides, and a controllable basic character of the
calcined mixed oxides [17]. Moreover, they form small,
thermally stable metal crystallites under reduction conditions
[19]. Thus, very stable nickel-containing material may be
prepared by using the hydrotalcite-type structure as the starting
material. In this sense, reduced Ni–Al(O) mixed oxides from
hydrotalcite-like precursors were successfully tested as
catalysts in the methane decomposition reaction by Li et al.
[12,20]. These authors studied Ni deactivation by carbonaceous
deposits and the lifetime of the Ni–Al(O) mixed oxides from
the hydrotalcite-like precursors during methane decomposition.
In this work, we studied different Ni, Mg–Al catalysts from
LDH precursors used in two cycles of the methane decom-
position reaction to produce hydrogen and also the regeneration
of the deactivated catalyst. Thus, this contribution explores the
influence of Mg on Ni-stability in the mixed oxide matrix
during two severe cycles of methane decomposition after
regeneration process of the catalyst using CO2 as an effective
regenerating agent. Comparison of these catalysts from
hydrotalcite-like precursors is also made with a Ni-catalyst
291 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 291
broadening (linewidth at half maximum, FWHM) of the three steps: (i) reduction, (ii) catalytic methane decomposition,
characteristic Ni reflection using the Debye–Scherrer equation. (iii) catalyst regeneration—with CO2 or O2. For each catalyst,
X-ray photoelectron spectra (XPS) were acquired with a VG two cycles of three steps were completed. The main (m/z)
Escalab 200R spectrometer equipped with a hemispherical signals relative to H2, H2O, CH4, CO, CO2 and Ar were
electron analyzer and an Mg Ka (hn = 1253.6 eV) X-ray recorded during the three steps of each of the two cycles of
source. The peaks were fitted by a non-linear least square fitting experiments. To eliminate errors due to changes in pressure
program using a properly weighted sum of Lorentzian and within the ionization chamber of the spectrometer, all signals
Gaussian component curves after background subtraction were referred to the Ar signal.
according to Shirley and Sherwood. The constant charging
of the samples was corrected by referencing all energies to the 3. Results and discussion
Al 2p peak at 74.5 eV.
The TPR analyses were performed on a Micromeretics 3.1. General characterization
2900 TPD/TPR instrument. The reducing agent was
10 vol.% H2/Ar, gas flow 50 mL min 1. The samples Chemical analysis of hydrotalcite-like precursors and the
(35 mg) were heated from room temperature to 1000 8C at a corresponding mixed oxides revealed no significant differences
rate of 10 8C min 1. This heating program was chosen based on with the nominal composition. In order to determine whether or
the literature to optimize the resolution of the curves. not cations were lost along the synthesis, filtered mother-liquor
Calibration of the instrument was carried out with a CuO waters were also analysed. The result showed a small loss of
standard (from Merck). Several cycles of reduction-regenera- Mg2+ (always less than 3%) in filtered mother-liquor waters
tion cycles were conducted with fresh catalysts to study Ni (Table 1). Table 1 also shows some textural properties of the
stabilization. mixed oxides derived from LDH. The high values of the specific
Two different thermogravimetric measurements were surface area of ex-LDH mixed oxides (NMA-I and NMA-II)
carried out on a Mettler Toledo TGA/SDTA 851 thermal point to the high thermal stability of the specific area in these
analyser. The first type of experiment consisted of 10 cycles of mixed oxides pretreated at such a high temperature (850 8C for
reduction–reoxidation of fresh catalysts under conditions 7 h, under air flow) during the calcination process. In contrast,
analogous to those of TPR analysis. The reduction process the supported catalyst (imp-NMA) displayed a very low surface
consisted of heating the samples (10 mg) from room area, characteristic of the alumina a-phase.
temperature to 1000 8C under a 20 vol.% H2/N2 flow. After The powder X-ray diffraction patterns of the two non-
reduction, the sample was purged in N2 to room temperature calcined hydrotalcite-like precursors (Fig. 1) reveal a single
and was then re-heated up to 1000 8C under a 20 vol.% O2/N2 phase typical of the hydrotalcite structure [17,23]. The basal
flow. The second type of thermogravimetric experiment reflections (0 0 3), (0 0 6) and (0 0 9) are easily recognized in
consisted of testing the regeneration of the used catalyst the patterns of the hydrotalcite-like materials, although the
efficiency. The regenerated catalysts – after reaction – were (0 0 9) line overlaps the (0 1 2) and (0 1 5) reflections, resulting
oxidized to quantify the amount of carbonaceous deposits by in a broad signal between 408 and 458 2u. No excess phase was
heating the used and regenerated catalysts from 50 to 1100 8C detected, suggesting that Ni2+ had isomorphically replaced
under an O2/N2 flow. Subsequently, the oxidized catalysts were Mg2+ cations in the brucite layers. The relatively high intensities
reduced under analogous conditions to those of the TPR combined with broad lineshapes indicate a relatively high
analysis. crystallinity of the hydrotalcites. LDH crystallinity increased
with Ni content.
2.3. Methane decomposition and catalyst regeneration Thermal decompositions of LDH precursors were com-
pleted from the layered structure (Fig. 1) to oxide structures
Temperature-programmed CH4 decomposition was per- (Fig. 2) at 850 8C. The XRD pattern of Ni–Mg–Al mixed
formed on 60 mg samples loaded in a U-shaped quartz reactor oxides from LDH (Fig. 2a and b) exhibited three broad
(4 mm i.d.) at atmospheric pressure on-line with a Baltzer reflections at about 378, 448 and 638 2u, characteristic of the
Prisma QMS 200TM quadrupole mass spectrometer. Acceler- periclase MgO (JCPDS file no. 43–1022) and/or bunsenite NiO
ated catalyst deactivation tests were chosen to probe catalyst
stability. Prior to the reaction, the catalyst was reduced in a 10%
Table 1
H2/Ar flow at 900 8C for 30 min. After reduction, the sample
Chemical composition and textural properties of the calcined materials
2 1
deactivated catalysts were regenerated using either CO2 or O2 600 8C, for CO2 and O2, respectively, maintaining the
as the regenerating agent. A flow of 10% of regenerating agent/ maximum temperature for 2 h. Thus, each cycle included
Ar (100 mL min 1) was passed while heating to 700 8C or
293 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 293
catalysts display a rather small crystal size of the NiO Calcined-fresha Reducedc Used in
(NiO) b (Ni) b reactiond (Ni) b
crystallites (cf. Table 2); however, a much larger crystallite size
is derived from the imp-NMA sample, whose nickel-loading is O2e CO2 f
similar to that of NMA-II. The large particle size of NiO elements and the relative surface proportions in the catalysts.
crystallites in the imp-NMA sample is due to the low specific For each sample, Ni 2p3/2, Al 2p, Mg 2p, O 1s and C 1s core-
surface area of the a-alumina substrate, which causes a low
interface to disperse nickel species. The crystallite size of the
metallic nickel particles in the reduced samples and those used
in the methane reaction are also shown in Table 2 (vide infra).
XP analyses of fresh catalysts and reduced NMA-I sample
were performed to determine both the chemical state of the
295 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 295
NMA-I 8 13 – –
NMA-II 6 11 8 7
imp-NMA 29 – – –
a
Performed prior to reduction.
b
Nature of the Ni particle measured.
c
Performed on calcined and later reduced samples.
d
Samples used in the CH4 decomposition reaction.
e
Regenerating agent used in the regeneration process: O2. Last step: reaction
process.
f
Regenerating agent used in the regeneration process: CO2. Last step:
regeneration process.
296 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 296
Table 3
Binding energies (eV) of core electrons of the HT samples and surface atomic
ratios of NMA-I and NMA-II fresh catalysts and reduced NMA-I
Catalyst Ni 2p3/2 Ni/Al Mg/Al
(%) (nominal value) (nominal value)
NMA-I fresh 856.5 (100) 2.50 (2.15) 0.38 (0.55)
NMA-II fresh 856.1 (100) 0.35 (0.5) 0.48 (1.3)
NMA-I reduced a 853.0 (31) 0.34 0.15
855.9 (69)
a
Reduction process in the TPR equipment: 10% H2/Ar (50 mL min 1),
heated from RT to 1000 8C at 10 8C min 1.
Table 4
Reduction temperature (8C) and percentage (%) of the deconvoluted peaks of the reduction profiles of the NMA-I and NMA-II samples and the reference samples
imp-NMA and Ni-spinel
Catalyst T1 T2 T3 T4 T0 T5 T6 T00
NMA-I fresh 470 (19%) 580 (40%) 650 (30%) 720 (6%) – 850 (3%) 950 (2%) –
NMA-II fresh – 570 (7%) – – 770 (93%) – – –
imp-MNA – – – – 775a – >1100a
Ni-spinel – – – – – 840 (97%) – –
a
Final reduction temperature up to 1100 8C; no determination of percentage of Ni reduced.
297 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 297
the last cycles are shown in Table 5. The most stable catalyst The results for catalytic methane pyrolysis on the NMA-II
was the one with low Ni-loading (NMA-II), which maintained catalyst are shown in Fig. 5. The results corresponding to CH4
its reduction profile during the 10 reduction–reoxidation cycles consumption and H2 production are expressed in terms of the
(Fig. 4b). In this catalyst, the Ni2+ environment stabilized corresponding (m/z) mass spectra as a function of temperature.
reduced Ni species at high temperatures through interaction of The only products detected were hydrogen and traces of carbon
the reduced Ni particles with the mixed oxide matrix. The only oxides. Fig. 5 (inset) shows the low amount of COx produced
change in NMA-II is the disappearance of the shoulder present (note the 100-fold smaller scale of the COx signal). The COx
at lower temperature (T2 peak, Table 5). The disappearance of produced comes from the initial C deposits produced at low
the T2 peak may due to the incorporation of Ni2+ inside the temperature (cf. Fig. 5: H2 begins to appear about 315 8C),
mixed oxide matrices. which react with the most reactive oxygen from the catalyst
For the NMA-I sample, no change in the reduction surface. It is known that increasing support basicity can lead to
temperature was observed although Ni distribution was altered the promotion of carbon gasification [25]. When this small
to a certain extent (cf. Fig. 4a and Table 5). The lower quantity of highly reactive oxygen is fully depleted, no more
temperature peak (T1) increased at the expense of the T2 and COx is detected. A good correlation between CH4 consumption
T3 peaks, but T3 maintained its Ni-population at the expense of and H2 production was found. Methane consumption and
Table 5
Reduction temperature (in 8C) and percentage (%) of the deconvoluted peaks of the reduction profiles of the fresh NMA-I and NMA-II samples, and the same samples
after 10 cycles of reduction–reoxidation derived from TGA
Sample T1 T2 T3 T4 T0 T5 T6
NMA-I fresh 460 (19%) 550 (39%) 595 (24%) 660 (8%) – 840 (7%) 940 (3%)
NMA-I after 10 cycles 460 (47%) 540 (19%) 630a (22%) – – 855 (12%) –
NMA-II fresh – 570 (10%) – 750 (90%) – –
NMA-II 10 cycles – – – – 760 100% – –
a
The FWHM of this peak was 350 8C vs. an average of 70 8C of other peaks, indicating that several species reducing in the large temperature range were present.
298 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 298
Table 6
Activity results for the decomposition of methane using the NMA-I, NMA-II and imp-NMA catalysts
Catalysts Catalytic activity Temperatures (8C)
a b
CH4 conversion (%) Total H2 produced (a.u.) Start H2 productionc Maximum activity Deactivation
NMA-I 48 30 460 650 680
NMA-II 39 10 480 600 625
imp-MNA 25 6 480 565 590
a
Maximum value.
b
Integral of Ar-normalized H2 mass signal curve.
c
CH4 conversion about 5%.
299 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 299
reoxidation cycles (Table 5, Fig. 4). To check the deactivation catalysts was observed. The oxidation power of CO2 was
of the catalyst, two reaction cycles were carried out without insufficient to oxidize the Ni metallic particles. Nickel particle
performing the regeneration process between the cycles size of the CO2 regenerated sample decreased with respect to
(results not shown). After the first reaction cycle (fresh the sample after the first reduction (Table 2), indicating
catalyst), the NMA-II catalyst became deactivated, and no- re-dispersion of the Ni particles after CO2 regeneration.
activity was observed in the following cycles of methane Despite this, the size of the Ni particles after CO2 or O2
decomposition. Fig. 7 shows the hydrogen produced using: (i) regeneration was almost the same (Table 2). It has been
fresh NMA-II catalyst (1st reaction cycle), (ii) CO2- proposed that deposited C may also be active for methane
regenerated NMA-II catalyst (2nd cycle-CO2), (iii) O2- pyrolysis [27]. On the other hand, dilution of the deposited C in
regenerated NMA-II catalyst (2nd cycle-O2). Good reprodu- the Ni particles produces nickel carbide formation [28]. It is
cibility of the reaction system was seen, as can be observed for possible that nickel carbide might be more active for catalytic
the same evolution of the hydrogen produced during the two methane decomposition than metallic Ni. Unfortunately, the
first cycles of methane decomposition (Fig. 7). In terms of the XRD nickel carbide pattern coincided with the metallic Ni
amount of hydrogen produced, regeneration with CO2 proved pattern. Both hypotheses imply the involvement of carbon,
to be better than with O2. Nonetheless, both regenerated NMA- although oxygen must play an important role in the new active
II catalysts were more active than the fresh counterpart (Fig. 7). phase of the catalyst because deposited C deactivated the
The higher activity of the regenerated catalysts may be catalyst and hence oxygen, from O2 or CO2, must be required to
explained in terms of the re-dispersion of Ni particles during reactivate the catalyst.
the reaction and regeneration processes [26]. Fig. 8 shows the To gain information about the nature of the new active site
reduction profiles of NMA-II during the reaction–regeneration after regeneration processes, the oxidation and subsequent
studies. When O2 was used as regenerating agent (Fig. 8a), a reduction of the CO2-regenerated NMA-II catalyst was carried
decrease in the Ni reduction temperature was observed. This out (Fig. 9). Fig. 9a shows that most of the Ni was present as
decrease indicates Ni particle re-dispersion, as measured by metal in the regenerated catalyst, although there was another
XRD (Table 2) of the used catalyst, and this is a common trend fraction of Ni oxide. When this oxidised-regenerated catalyst
in supported catalysts. When CO2 was used in catalyst was reduced again (Fig. 9b), only 75% of the metallic Ni was
regeneration (Fig. 8b), no reduction of the used-regenerated reduced, and the rest (25%) of the Ni remained unreduced. The
high oxidation temperature (cf. Fig. 9, 1100 8C), which is
sufficient for the oxidation of Ni and nickel carbide in other
nickel-containing catalysts [29], did not allow the reduction of
this fraction of nickel oxide.
Fig. 8. Reduction processes during the reduction steps after regeneration with
(a) O2 or (b) CO2, as regenerating agents. First reduction of the fresh NMA-II
and second reduction of the regenerated NMA-II sample are represented by
solid line and squares, respectively. Reduction represented by corresponding Fig. 9. (a) Oxidation of reduced fresh NMA-II (dash line) and CO2-regenerated
(m/z) mass of H2O produced as a function of temperature. All mass signals were used NMA-II catalyst (solid line). (b) Reduction of fresh NMA-II (dashed line)
referred to the Ar signal. and re-oxidized NMA-II after CO2 regeneration (solid line).
300 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 300
Elsevier Todos los derechos reservados. doi: fosgeno o la formación de carbonilos metálicos
10,1016 / j.cattod.2006.05.078 [16].
Por otro lado, la necesidad de un proceso de
específicamente del gas natural [3,4], es regeneración hace
actualmente la forma económica más grande y necesario el uso de catalizadores de Ni en el
más para producir hidrógeno. Formas que los cristalitos de Ni han sido estabilizados
alternativas incluyen la oxidación parcial de con el fin de evitar Ni sinterización durante el
metano [3,5], la reacción de reformado de proceso de regeneración. Una opción
metano oxi-autotérmico [6] y la reacción seco alternativa para la estabilización de las
reformado de metano [3,7]. partículas de metal activo es el uso de
Una ruta alternativa diferente para la diferentes precursores. La llamada compuestos
producción de hidrógeno es la descomposición similares a hidrotalcita parecen ser buenos
catalítica de metano: el llamado pirólisis metano candidatos para la preparación de precursores
o metano descarbonización [3,8] (1). estabilizadas de Ni-catalizadores, ya que
CH4! C þ 2H2; DH0 ¼ TH90: 1 kJ mol = (1) pueden permitirse óxidos mixtos altamente
Esta reacción ha recibido mucha atención en estabilizados térmicamente tras el tratamiento a
los últimos años [9], alta temperatura (tratamiento de calcinación)
debido a que su característica más importante [17]. Compuestos hidrotalcita son hidróxidos
es la producción de H2 libre de CO o CO2 y el dobles en capas (LDHs) derivados de las capas
único subproducto de esta reacción es un de brucita con una fórmula ideal de
material de carbono o de una especie
carbonosos [10]. El H2 libre de CO2 CO- 2. Experimental
producida es óptima para el suministro directo a
una pila de combustible, cuya 2.1. Preparación del catalizador
electrocatalizadores son muy sensibles a la
presencia de CO en niveles de ppm en la Los precursores similares a hidrotalcita
corriente de hidrógeno [11]. descritos en este estudio fueron preparados por
Está bien documentado que los catalizadores co-precipitación mediante la hidrólisis de urea
de níquel catalizan la descomposición del [21]. Una solución acuosa (0,5 M) que contiene
metano en hidrógeno y carbono como C- cantidades apropiadas de NiCl2 6H2O (metal
nanofibras [12,13]. El principal problema de la activo), MgCl2 6H2O (modificación de catión
utilización de catalizadores de Ni es la divalente) y AlCl3 6H2O en presencia de urea
desactivación del catalizador a través de los (urea
depósitos de carbón en la superficie del Cantidad = 3 pmol de cationes) como el agente
catalizador, sobre el centro activo de Ni [14]. La precipitante
desactivación del catalizador puede ser se calentaron, controlando la temperatura para
superada por la regeneración del catalizador. obtener el valor pH de la precipitación deseada.
El precipitado fue envejecido en suspensión a
Este carbono es de hecho altamente reactivo a 90 8C durante 48 h bajo agitación. El sólido
O2 [15] (2) o CO2 [16] (3): resultante se filtró, se lavó a fondo con agua
destilada - hasta que el pH del agua filtrada fue
C þ O2! CO2; DH0 ¼ 394: 7 kJ mol = (2) de alrededor de 6 - y se secó durante la noche
a 80 8C. Todos los precursores contenidos
CO2 C þ! 2CO; DH0 ¼ þ174: 5 kJ = mol (3) El tanto de cloruro (Cl) y carbonato (CO32) como
uso de CO2 como agente de regeneración (de aniones de carga de compensación. Productos
reacción (3)) permite químicos y reactivos fueron adquiridos de
recogida del producto CO, que puede ser Aldrich.
considerado como una Dos muestras diferentes, dependiendo de la
químico intermedio útil para la carbonilación de composición nominal, se obtuvieron con dos
metanol reacciones y acetileno, la diferentes cargas de Ni: (i) alta carga Ni-
hidroformilación de alqueno, la síntesis de (composición nominal en la relación molar de
303 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 303
(JCPDS archivo no. 22-1189) fases. Debido a la Tabla 2); sin embargo, un tamaño de cristalito
coincidencia casi completa de las líneas de mucho más grande se deriva de la muestra imp-
difracción de las fases de MgO y NiO, la NMA, cuya carga es de níquel-
discriminación entre estas fases no era posible.
Picos débiles adicionales acerca de 198, 328 y -Fresha calcinado
458 2u se pueden atribuir a la formación de (NIO) b
fases de espinela MgxNi1 xAl2O4, por ejemplo,
NiAl2O4 JCPDS archivo no. 78-552. La Reducedc
cantidad de fases de espinela aumentó con el (Ni) b
aumento de carga de magnesio, a partir de la
Fig. 2a (15% Mg) a la Fig. 2b (46% Mg), pero Usado en
no fue posible determinar la composición reactiond (Ni) b
química de la fase espinela por XRD.
El tamaño de los cristalitos de la fase de óxido O2e CO2f
de níquel segregada se obtuvo de la ampliación similar a la de NMA-II. El gran tamaño de
de la línea XRD (Tabla 2). Para evitar el fuerte partícula de los cristalitos de NiO en la muestra
solapamiento de las reflexiones más imp-NMA se debe a la baja superficie específica
característicos de las fases de MgO y NiO, el del sustrato a-alúmina, lo que provoca una
NiO (1 1 1) la reflexión a aproximadamente 378 interfaz bajo para dispersar las especies de
2u se consideró debido a que su intensidad de níquel. El tamaño de los cristalitos de las
68% es mucho más alta que la correspondiente partículas metálicas de níquel en las muestras
reflexión característico de MgO (4% ). Ex-LDH reducidas y los utilizados en la reacción de
metano también se muestran en la Tabla 2
(vide infra).
XP análisis de los catalizadores frescos y redujo
la muestra NMA-I se realizaron para determinar
Fig. 2. patrones de XRD de calcinado Ni tanto el estado químico de los elementos y las
catalizadoras ex-LDH: (a) NMA-I y (b) NMA-II. proporciones de superficie relativas en los
Fase de identificación: (&), NiO; JPDF no. 22- catalizadores. Para cada muestra, Ni 2P3 / 2, Al
1189; (*), MgO; JPDF no. 43 a 1022; (), 2p, 2p Mg, O 1s y C 1s sin núcleo
NiAl2O4; JPDF no. 78-552.
NMA-I 8 13 - - NMA-II 6 11 8 7 imp-NMA 29 - - -
una realizada antes de la reducción.
Se registraron los espectros de nivel. Las b naturaleza de la partícula Ni midió.
energías de unión de los picos principales se c realizados sobre muestras calcinadas y
resumen en la Tabla 3; los porcentajes de las posteriormente reducidos.
diferentes contribuciones de los picos se dan en d muestras utilizadas en la reacción de
paréntesis. Las energías de enlace para el nivel descomposición de CH4.
de Ni 2p3 / 2 en ambos a 856,0 eV en e agente regenerador utiliza en el proceso de
regeneración: O2. Último paso: Reacción
Tabla 2 proceso.
Crystal tamaño de segregada NiO y Ni metálico f agente Regenerador utiliza en el proceso de
calcula utilizando la fórmula de Debye-Scherrer regeneración: CO2. Último paso:
realiza en fresco y catalizador reducido y la proceso de regeneración.
utilizada en el proceso de reacción-
regeneración Tabla 3
Encuadernación energías (eV) de electrones
Catalizador tamaño de cristal de las muestras internos de las muestras HT y superficiales
tratadas (nm) relaciones atómicas de NMA-I y NMA-II
catalizadores muestran más bien un pequeño catalizadores frescos y reducen NMA-I
tamaño de cristal de los cristalitos de NiO (cf.
307 R. R.
Guil-Ló
Guil-Ló
pezpez
et al.
et al.
/ Catalysis
/ Catalysis
Today
Today
116116
(2006)
(2006)
289–297
289–297 307
Tabla 4
Temperatura de reducción (8C) y porcentaje
(%) de los picos deconvoluted de los perfiles de Fig. 5. Resultados de la pirólisis catalítica de
reducción de las muestras NMA-I y NMA-II y las metano en la muestra NMA-II. Los resultados
muestras de referencia imp-NMA y Ni-espinela correspondientes a consumo y la producción de
Catalizador T2 T3 T4 T1 T0 T5 T6 T00 CH4 H2 están representados por sus
correspondientes perfiles (m / z) de masa como
NMA-I fresca 470 (19%) 580 (40%) 650 (30%) una función de la temperatura. Los perfiles de
720 (6%) - 850 (3%) 950 (2%) - CO y CO2 se han ampliado para facilitar la
NMA-II fresca - 570 (7%) - - 770 (93%) - - - identificación. Todas las señales de masas se
imp-MNA - - - - 775A -> 1100a refiere a la señal de Ar.
Ni-espinela - - - - - 840 (97%) - -
una temperatura de reducción final hasta 1100
8C; ninguna determinación del porcentaje de Ni el pico T4, que desapareció (cf. Tabla 5). Pico
reduce. T6 también desapareció mientras que el pico T5
aumentó. El aumento en el fondo a alta
temperatura durante el último ciclo indica que la
transformación aún no ha llegado a la
Fig. 4. Los estudios de los ciclos de reducción conclusión (Fig. 4a). Por último, los 10 perfiles
de reoxidación del (a) NMA-I (10 ciclos) y (b) de oxidación también fueron diferentes para
NMA-II (10 ciclos), utilizando la técnica de TG. cada catalizador (Fig. 4), señalando que los
La curva DTG positivo corresponde a procesos diferentes entornos de Ni se mantienen
de oxidación y la curva DTG negativo después de descenso.
corresponde a procesos de reducción. Primer
ciclo (cuadrados), último ciclo (línea continua) y 3.3. La descomposición de metano y la
intermedio ciclos (líneas punteadas). regeneración del catalizador
(b) D. Duprez, K. Fadili, J. Barbier, Ind. Eng. [29] A. Djaidja, S. Liberaciones, A. Kiennemann,
Chem. Res. 36 (1997) A. Barama, Catal. Hoy en día, en prensa.
Desde 3.180 hasta 3.187.
[15] J. I. Villacampa, C. Royo, E. Romero, J. A.
Montoya, P. Del Angel, A.
Monzo' n, Appl. Catal. R: El general 252 (2003)
363 a 383.
[16] K. Otsuka, S. Takenaka, H. Ohtsuki, Appl.
Catal. R: El general 273 (2004) 113-
124.
[17] A. Cavani, F. Trifiro, A. Vaccari, Catal. Hoy
en día 11 (1991) 173-301.
[18] La composición química de hidrotalcita y el
Grupo de hidrotalcita
Minerales, vol. 58, W.F. Foshag, Washington,
1921, pp. 147F-153f.
[19] (a) K. Takehira, T. Shishido, P. Wang, T.
Kosaka, K. Takaki, J. Catal. 221 (2004) 43-54;
(b) S. Murcia-Mascaros, R. M. Navarro, L.
Gómez-Sainero, U. Costan- tino, M. Nocchetti,
JLG Fierro, J. Catal. 198 (2001) 338.
[20] (a) Y. Li, J. Chen, L. Chang, L. Chang,
Appl. Catal. R: El general 163 (1997)
45-57;
(b) J. Chen, Y. Li, Z. Li, X. Zhang, Appl. Catal.
R: El general 269 (2004) 179-
189.
[21] U. Costantino, F. Marmottini, M. Nocchetti,
R. Vivani, Eur. J. Inorg.
Chem. (1998) 1.439 a 1.446.
[22] Comité Conjunto sobre polvo de difracción
estándar, Centro Internacional de
Difracción de datos, Pennsylvania, 1997.
[23] (un) S. Miyata, A. Okada, arcillas arcilla
Miner. 25 (1977) 14-17; (b) S. Miyata, Arcillas
arcilla Miner. 31 (1983) 305.
[24] D. Dissanayake, M.P. Rosynek, K.C.C.
Kharas, J.H. Lunsford, J. Catal.
132 (1991) 117.
[25] L. García, R. francés, S. Czernik, E.
Chornet, Appl. Catal. R: El general 201 (2000)
225.
[26] T. Ueckert, R. Lamber, N.I. Jaeger, U.
Schubert, Appl. Catal. R: El general 155
(1.997) de 75-85.
[27] T.V. Choudhary, E. AKSOYLU, D.W.
Goodman, Catal. Rev. Sci. Ing. 45 (2003) 151-
203.
[28] E. Ruckenstein, Y.H. Hu, Carbono 36
(1998) 269-275.