close

Вход

Забыли?

вход по аккаунту

?

Synthesis Gas Production Using Carbon Dioxide as a Source of Carbon-Current Research and Perspectives.

код для вставкиСкачать
Dev. Chem. Eng. Mineral Process., 7(5/6),pp.443-462, 1999.
Synthesis Gas Production Using Carbon
Dioxide as a Source of Carbon - Current
Research and Perspectives
G.Q. (Max) Lu and S. Wang
Department of Chemical Engineering, The University of Queensland,
Brisbane, Queensland 4072, Australia
C 0 2 is a major greenhouse gas, but it is also an important source of carbon. Catalytic
reforming of methane with CO, to synthesis gas is a promising technology in
utilisation of C02.Thisprocess has attracted increasing attention among the research
communities in the world, and much progress has been made. This paper presents an
overview on the recent advances in catalyst development, reaction mechanism and
kinetics. It is found that catalytic activity and stability strongly depend on the nature
of support, metal-support interaction, metal precursor and addition of promoters.
Further fundamental research on the reaction and coking mechanisms and kinetics
will lead to a high performance catalyst system that will deliver high conversion at
low temperaturesand with long-term coking-pee operation.
Introduction
Carbon dioxide is a principal greenhouse gas. It is estimated about 6 billion tons of
COz (carbon-equivalent mass) is released into the atmosphere every year [Ichikawa et
al., 19931 accounting for 60% of the total greenhouse gas emissions in the world. In
Australia, about 290 Mt of COz is released from fossil fuel combustion, being 1.3-1.5
% of the total world COz emissions [IEA, 19941. Australia is one of the signatories
of the 1992 Rio Treaty on “the Framework for Climate Change” at the Earth
Summit’s United Nations Conference on Environment and Development. In the
recent Kyoto Protocol (1997), Australia was given some special considerations for its
unique energy resources and economical features, and was allowed to limit the
greenhouse gas emission growth to achieve 8% above the 1990 levels by the year
2010. Nonetheless, Australia like many other signatory countries has an obligation
to enforce new policies and measures to significantly reduce C 0 2 emissions, at least
443
G.Q.LU and S. Wang
limit the growth rate to be 0.13% per year instead of 1.22 'YOas in the business-asusual scenario. It is well known that combustion of fossil fuels contributes a large
portion of the C02 emissions. Over the last few years, Australian utilities industries
have started to pay more attention to C02 emissions. Some progresses have been
made in areas of policy formulation and emission reduction by improving energy
efficiency and developing renewable energy resources. In particular, the coal-firing
power industry starts to take it very seriously in promoting clean coal combustion and
gasification technologies to partially reduce C 0 2 emissions.
However, considering the amount of C02 generated in coal-burning power
stations, it is unlikely to meet the emission control target by the year 2010 without
employing new technologies to actively sequester or utilise C02 from flue gases.
With the increasing concern over global warming and climate change, there has been
an exponential growth in research and development on carbon dioxide futation and
utilisation since 1990. Investment in R&D in this area continues to increase rapidly in
Japan, USA and EU countries. This is mainly because utilisation of C 0 2 via a proper
technology not only help reduce C02 emission but also bring about potential profits
from the products. Many processes have been developed to either utilise the C 0 2 for
manufacturing chemicals, fuels and other materials, or to fix it in some forms of nonuseful materials. Technologies for chemical conversion can be realised by catalytic,
electrocatalytic, photocatalytic, and enzyme catalytic reactions. In general chemical
catalytic reaction is more technologically promising and economically viable for
large-scale conversion of C02[Wang et al., 19951.
Carbon dioxide hydrogenation to fuels such as methane, methanol, synthesis
gas and other C1 compounds is one of the effective processes. However, this process
requires hydrogen or expensive reactants, which consume large quantity of energy in
their production. This would limit commercialisation of these processes. Since the
most abundant hydrogen-rich material is methane beside water, utilising methane to
reduce C 0 2 is another promising method. In addition, this approach makes it possible
to reduce emissions of methane as well, particularly attractive for the coal-seam
methane and landfill gas utilisation.
444
Synthesis gar production using carbon dioxide as carbon source
Synthesis gas is an important raw material both as fuel and industrial
feedstock. Currently, most synthesis gas is produced by steam reforming of methane
(Reaction 1). Partial oxidation of C& (Reaction 2) and COz reforming of CH,
(Reaction 3) are newly developed processes. Comparing with the former two
processes, COz reforming of methane has several advantages [Wang et al., 19961. It
produces a synthesis gas with a lower Hz/CO ratio and has higher energy efficiency in
conversion to hydrocarbons. Such syngas produced has a high CO content and is
effective for the synthesis of valuable oxygenated chemicals like methanol, alcohol,
light olefins and gasoline [ h i , 1996
1. Because of
its high endothermicity, the
reaction can be used in energy transfer from solar energy to chemical energy, energy
storage in the form of CO and H2, and transporting nuclear energy. Economic
evaluation on conversion of syngas to acetic acid by applying above three processes
showed that the COz reforming option has the lowest operating cost, about 20% lower
than the other processes [Ross et al., 19961. Many studies have pointed out that C 0 2
reforming of methane into syngas is a promising technology from the viewpoints of
both economics and environment.
CH, + H20
+ CO +3Hz
AH298= 206 kJ/mol
(1)
c& + 0 2 +
2 c 0 +4H2
AHzg8 = -71 kJ/mol
(2)
+
2CO +2H2
AHm8 = 247 kJ/mol
(31
COZ + CH4
Background on CI&/CO2 reforming reaction
Catalytic reforming of CH4 with COz was first proposed by Fischer and Tropsch in
1928. However, there is so far no established commercially industrial process for
COz reforming of C& due to the fatal problem of coking. No effective commercial
catalyst exists which operates without carbon formation.
Thermodynamics
This reaction is highly endothermic, favoured by low pressure but requires high
temperature. A reverse water-gas shift reaction occurs as a side reaction (R4). Under
conditions of stoichiometric COz reforming, carbon forms as in the Boudouard
reaction (R5)and methane cracking (R6).
445
C.Q.Lu and S.Wang
C02+ H2 + H20+ CO
2 c o + c02 + c
AH298 = 41 kJ/mol
AH298= - 172 kJ/mol
(4)
CI&+
A H298= 75 kJ/mol
(6)
C +2H2
(5)
Figure 1 presents CH, and C02conversions at C02:CI&=l:1. As shown, CI&
and C02 conversions increase as temperature increases. At 8OO0C, the conversions
reach over 90%. It is also seen that the C02 conversion is always greater than that of
CH,due to the reverse water-gas shift reaction.
I
400
450
500
550
600
650
700
750
8W
Temperature(%)
Fig.1 CI€, and C02conversions at various temperatures
Calculations indicate that in the temperature range of 557 - 7OO0C, carbon
will be formed fiom methane cracking or the Boudouard reaction [Wang et al., 19961.
Figure 2 shows the lower and upper temperature limits to prevent carbon formation
and carbide formation at 1 atm. With a feed ratio of CO2/CI&=l:1, the optimal
temperature is between 870-1040 OC. Hence, methane reforming with C02 must
operate at high temperature so as to avoid coking. The energy required for this
reaction can be waste process heat or solar energy in practical applications and heat
pipe technology can be incorporated into this process for the energy recovery purpose.
446
Synthesis gas production using carbon dioxide as carbon source
Key issues in catalytic reforming
Coking
Carbon deposition is the major deactivating factor in catalytic reaction, especially for
Ni-based catalysts. The catalysts based on noble metals are reported less sensitive to
coking compared to the nickel-based catalysts for CH.,+C02 reaction. However,
considering the high cost and limited availability of noble metals, it is more practical
to develop improved Ni-supported catalysts with high coking resistance and stable
activity.
1200 I
1100
1
ouu
-
1
I
I
I
I
\
.
1
I
I
,
I
I
1
2
3
4
5
700
0
I
I
6
1
1
1300 y^ 1200
%!
1100
2
al
@OO-
(b)
\
-
Y
=
-
I-
900 -
800 I
I
I
I
I
I
I
0
1
2
3
4
5
6
COJCH,
Fig.2 Effects of feed ratio of C02/CH4on limiting temperatures below which
carbon deposits at various pressures.
G.Q.Lu and S. Wang
To prevent coking several approaches have been attempted. They are (1)
sulfur passivation of catalyst, (2) selection of good support, (3) addition of promoter,
and (4) change of reaction conditions.
Carbon suppression on Ni catalysts by sulfur passivation has been
commercialised as the SPARG process [Dibbern et al., 19861. Sulfur may block the
nickel catalyst surface, which means that nickel cannot catalyse the formation of
carbon because of the ensemble effect. However, the passivation process suffers fiom
lower catalytic activity and high operating temperatures.
Ruckenstein and Hu [1995] reported that NiOMgO catalysts with a solid
solution formed, exhibited carbon suppression, higher stability and yields. A novel
N i b 2 0 3 catalyst was developed by Zhang and Verykios [ 19951, which exhibits high
activity and excellent long-term stability for C 0 2 reforming of methane. This catalyst
is stable at temperatures as low as 823K without losing its catalytic activity. We also
found that a Ni/y-A1203catalyst showed high conversion and long life during 140 h
test.
In our study, we found that promoters Na20, MgO, La203 and Ce02
exhibited the ability of carbon suppression. On the contrary, CaO increased carbon
formation. Figure 3 presents the dynamic coking process over unpromoted and
promoted Ni/A1203catalysts at 70OoC. It is seen that carbon on Ni/AI2O3is about
20%. Carbon on Ni/Ca0-AI2O3 can reach 35 % after 2h reaction while the coke on
Ni/Na2O3-AI2O3is only about 3%.
From Fig. 2 it is also seen that the temperature limit for carbon deposition
increases as the pressure increases for the same feed ratio. Clearly, carbon formation
is thermodynamically possible for a C02/CH, reforming feed ratio of 1:l at
temperature up to 87OoC at 1 atmosphere, and 103OoC at 10 atmospheres. In
addition, at a given pressure, the temperature limit increases as the C02/CH, feed
ratio decreases. This means that using excess C 0 2 in the feed may avoid carbon
formation at lower temperatures. Normally, C 0 2 reforming of methane is best
operated at 1 atm. Higher temperature operation will increase cost. Therefore, Using
C 0 2 rich feed will help to reach maximum conversion and suppress carbon
deposition.
448
Synthesis gas production using carbon dioxide as carbon source
Reaction Mechanisms
Understanding of the reaction mechanism and kinetics helps to model an effective
catalyst. Some researchers believe that the reaction of COzwith CH4 is expected to
proceed by the following steps: (])dehydrogenation of methane to form surface
carbon and hydrogen; (2)dissociative adsorption of COzand H2 ,and (3) reduction of
C 0 2 to CO. However, others proposed that C 0 2 participate in the reaction in gas
phase. C 0 2 is directly reacted with hydrogen fi-om C& dissociation. The kinetics for
C 0 2 reforming of methane depends on the type of catalyst and reaction conditions. So
far, no generally expression has been derived. More comprehensive studies are
desirable to elucidate the mechanism.
0.40 I
I
0.35
I
I
I
NUCaO-AI,O,
I
I
0.30
0
5m 0.25
g
1
0.20
a
8 0.15
0.05
0.00
0.0
0.5
1.o
1.5
2.0
Time (h)
Fig.3 Effect of promoters in Ni/y-A1203catalyst on carbon deposition
Recent Advances in Catalytic Reforming for C& with CO2
Active metal
Nickel-based catalysts and noble metal-supported catalysts (Rh, Ru, Pd, Pt, Ir) have
been found to have high activity and selectivity for this reaction. Table 1 illustrates
the activity order for metals dispersed on various supports. It is seen that the
449
G.Q.LU and S.
Wang
combination of metal and support affects the resultant catalyst activity. Even for the
same support there are conflicting conclusions reported by various researchers. This
may be due to the different operation conditions in those studies. Among the first row
of VIII metals (Fe, Co, and Ni), Ni shows the highest activity. Rh is a superior
component with high activity and coking resistance with A1203as a support. Ru may
be better than Rh if MgO, SiOz are used as supports. Other noble metals such as Pd
and Pt showed high activity supported on AlzO3, while Ni catalysts with MgO or
zeolite support demonstrated less coking ability.
Loadings of metals on supports also affect the activity of a catalyst. Low
loadings (about 1-5wt %) of the noble metals are usually sufficient because of their
effectiveness. Higher loadings are required for Ni and Co catalysts in which the
metal-support interaction is stronger. However, there is an optimum loading beyond
which an increase in nickel content did not produce any further significant increase in
conversion. The effect of Ni loading on CH4 conversion over Ni/y-Alz03 catalyst
obtained in our laboratory is shown Figure 4.
1
0 '
0
5
10
15
I
20
N i loading ( w t Y )
Fig.4 Effect of Ni loading on C& conversion over Ni/y-Al2O3catalyst at different
temperatures
It is seen that the optimum Ni loading on Ni/y-A1203was approximately 12-
15 wt%. Apart from group VIII metals, other metals have recently been used for dry
methane reforming with COz. A Manganese-based catalyst gave high yield of
synthesis gas without carbon deposition detected at 1200 K. This catalyst, however,
450
Synthesis gas production using carbon dioxide as carbon source
had significantly lower activity than the group VIII metals [Mirzabekova et al., 19921.
Rhenium supported on alumina was also tested as a catalyst for dry reforming of
methane. At temperatures above 973K, Re was actually more active than Ir for the dry
reforming reaction at stoichiometric C&:C02 ratios. However, the activity of Re
catalyst decreased dramatically at lower temperatures. At about 873K, methane
conversion was found less than 5% [Claridge et al., 19941. Bimetal catalysts are also
effective and show synergistic effect for this reaction. A Ni-based three-component
catalyst such as Ni-La203-Ru or Ni-CeOz-Pt exhibited high activity.
Table 1. Catalytic activities of metals on various supports
Metal Activity
Metal
Temperature Reference
loading
(K)
(wt"/.)
1 AlzOj
1
823
Solymosi et al., 1991
Rh > Pd>Ru>Pt>Ir
0.5-1
823-973
Sakai et al., 1984
Rh>Pd>Pt>>Ru
Ashcroft et al., 1991
1
1050
Ir>Rh>Pd>Ru
Xu et al., 1992
9
773-973
Ni>Co>>Fe
Tokunaga et al., 1989
Ni>Co>>Fe
10
1023
0.5
873
Richardson et al., 1990
Ru>Rh
0.5
923-1073
Richardson et al., 1990
Rh>Ru
2 SiOz
1
973
Nakamura, 1993
Ru>Rh>Ni>PPPd
893
Nakamura, 1993
Ni>Ru>Rh>PPPd>>Co 0.5
3 MgO
Qin & Lapszewicz, 1994
Rh>Ru>Ir>PoPd
0.5
1073
1
973
Nakamura, 1993
Ru>Rh>Ni>Pd>Pt
1
823
Rostrup-Nielsen, 1993
Ru>Rh-Ni>Ir>PPPd
Ru>Rh>Pt>Pd
1
913
Takayasu et al., 1991
4 Eu203
1-5
873-973
Perera et al., 1991
RuXr
5 NaY
2
873
Kim et al., 1994
Ni >Pd>Pt
Even at a low concentration, the precious metal enhanced the reaction rate
greatly and this synergistic effect was ascribed to the hydrogen spillover effect
through the part of precious metal resulting in a more reduced surface of the main
catalyst component. In particular, a marked enhancement was observed by the
451
G.Q. Lu and S.Wung
combination of a low concentration Rh with Ni-Ce02-Pt catalyst [Inui et al., 19951.
Chen et al. 119961 also found that addition of noble metals to Ni-based catalysts
produced promoting effects in catalytic activity and carbon resistance. A small
amount of noble metals can promote the reducibility of based metal and stabilise their
degree of reduction during the catalytic process.
Catalyst support
It is generally believed that metal is the active site for C02-C& reaction. For
supported metal catalysts, the nature of support greatly affects the catalyst activity due
to the metal-support interaction and acid-base property, which will change the active
surface area. Carbon dioxide reforming involves the adsorption and dissociation of
C02on catalysts. Since C 0 2 is well known as an acid gas, adsorption and dissociation
of C 0 2may be improved with a basic catalyst.
Metal-support interaction plays a role in catalytic activity depending on its
strength. Weak metal-support interaction favours the dispersion of metal ions.
However, sintering of metal particles may be pronounced with weak metal-support
interaction. Strong metal-support interaction (SMSI) can induce the migration of
support species to metal surface covering the active sites resulting in lower activity.
We found that Ni/Ti02 and Ni/Ce02 catalysts showed much less activity due to
SMSI. Another interaction is the reaction between metal oxide and support to form
solid solution which generally happens in Ni/y-A1203 and Ni/MgO catalysts. This
effect helps to stabilise the nickel particle and thus makes them exhibit a long-term
stability.
Table 2 summarises the order of catalytic activity over supported catalysts in
the literature. It is seen that A1203 is much better as a support than Si02 and MgO for
C02/CI-&reaction. We have studied the catalytic behaviour of various Ni catalysts
supported on metal oxide, activated carbon, clay and zeolite for this reaction and
found that Ni-oxide catalysts are generally better than other supported Ni catalysts.
W A C catalysts showed the least activity. Among the Ni-oxide catalysts, Ni/AI2O3,
Nina203 and Ni/MgO showed not only high activity but also high stability [Wang,
19981. The stability of various Ni-oxide catalysts at 7OO0C is shown in Fig. 5 .
452
Synthesis gas production using carbon dioxide as carbon source
Catalystpreparation
There are many techniques for preparing supported catalysts. The primary difference
lies in the manner in which the support material and precursor salts are brought
together. Therefore, the active size can be varied depending on the textural structure
of support, technique employed, and active species precursor. We have prepared
Ni/Si02 and Ni/A1203catalysts with different porous structure and found that porous
supports favour the metal dispersion and the contact between reactants and the active
sites, resulting in the higher conversion [Wang and Lu, 19971. In another research on
Ni/clay catalysts for C02 reforming reaction of methane, CH4 conversion showed
positive relation with external surface area of catalyst [Fig. 61.
100
I
I
I
I
!
4
20
I
I
I
I
!I
0
5
10
15
20
25
0
5
10
0'
15
20
25
Time (h)
Fig.5 Stability of various Ni-oxide catalysts at 70OoC.
453
G.Q.Lu and S. Wang
Chang et a1.[1994] reported the different activity of pentasil zeolitesupported nickel catalysts synthesised by two different methods, namely, (i) solidstate reaction and (ii) incipient wetness method. KNiCdZSI catalyst prepared by
incipient wetness method had much less activity than the one prepared by solid-state
reaction technique. KNiCdZSI(1) catalyst showed over 90% CH., conversion at
8OO0C. It maintained its high performance for over 140 h without deactivation. They
attributed it to the melting effect of metallic precursor mixtures via solid-state
reaction different fiom impregnation. We prepared two Ni/MgO catalysts by
impregnation and coprecipitation methods. It is found that Ni/MgO prepared by
impregnation method showed high CH, and C02 conversions whereas the one
prepared by coprecipitation technique had little activity [Wang and Lu, 19971.
Activity order
Table 2. Effect of support on catalyst activity
Tem
Metal loading Reference
(K)
Ru
A12O3>TiOpSiO2
Ti02>A1203>Si02
Pd
TiO2>AI2O3>NaY>SiO2>MgO
>Na-ZSM-5
TiO2>Al2O3>SiO2>MgO
Rh
YSZ>Al2O3>Ti02>SiO2>>MgO
AI2O3>SiO2>TiOpMgO
NaY>NaZSM-S>A1203>SiOz
893
893
0.5
0.5
Nakamura., 1994
Nakamura, 1994
773
5
Masai et al., 1988
773
1
Erdohelyi et al., 1994
923
773
823
0.5
1
1
Tsipouriari, 1994
Erdohelyi et al., 1993
Bhat, 1997
723
0.5
Bitter et al., 1996
8001000
873
873
823
1073
40
10
2
4
5
Takano et al., 1994
Tokunaga et al., 1989
Kim et al., 1994
Swaan et al., 1994
Li et al., 1995
Pt
AI2O3>ZrO2>TiO2
Ni
AI2O3>SiO2
723
454
2-10
Bradford et.al
1996
Synthesis gas production using carbon dioxide as carbon source
Ni/(La)MT
NgAI-PILC
NUB1
I
I
150
200
40
0
50
100
Fig.6 Relationship between CH,conversion and surface area of Ni/clay catalysts
The interaction between nickel precursor and support plays an important role
in determining catalyst behaviour. Ruckenstein and Hu [ 1996al studied the activity of
Ni/La203catalyst for C02 reforming of methane using different precursors, and their
results showed that Ni catalyst based on nickel nitrate had a high initial CO yield but
a low stability. In contrast, the Ni/La203 catalyst based on chloride had a high
stability. They also found that the precursor of support influenced the catalytic
activity [Hu and Ruckenstein, 1997bl. In our work, three nickel catalysts were
prepared fiom nitratewi-N), chloride(Ni-Cl) and acetylacetonate(Ni-AA) salts and
their catalytic behaviours for this reaction were compared. Figure 7 presents the
stability results of the three catalysts at 70OoC. Ni-N is found the most active and
stable catalyst. Ni-Cl and Ni-AA showed deactivation due to coking.
Promoters
Addition of metal oxides may have an influence on the activity of supported metal
catalysts depending on the nature of additives. Catalyst modification can bring about
the blocking of active sites on the metal surface or changes in the electronic character
or geometric structure of the catalyst surface, thus change in reactivity by interacting
with the substrate to alter its mode of adsorption.
455
G.Q. Lu and S.Wang
-
E>s
s
f
0
40
-
20
-
0
0
A
-
Ni-N
Ni-CI
Ni-AA
5
100
-
10
15
I
I
20
25
I
&AA.
01
0
I
I
1
I
5
10
15
20
I
25
Time (h)
Fig.7 Stability of Ni/y-A1203catalysts prepared by different Ni precursor
Swaan et al.[l994] found that Ni-WSiO2 catalyst showed less activity for
C02 reforming of methane because of a decorating effect of the active phase by K'
ions. Halliche et a1.[1996] investigated the activity of Ni/a-A1203modified by Fe, Co,
Cu, and Ce oxides and observed lower activities on the modified catalysts. Zhang and
Verykios [1994] found that addition of CaO promoter to Nily-A1203catalyst resulted
in an increase in reaction rate. Wang et al. [1996] reported that La203 and Ce02
promoted Ni/y-A1203catalysts showed higher CH, and COz conversions compared
456
Synthesis gas production using carbon dioxide as carbon source
with Ni/y-Alz03 catalyst. However, Blom et al. [1994] reported that lanthanummodified Ni/y-A1203catalysts were less active for the reforming reaction due to the
high temperature sintering, which resulted in its lower reducibility. In our
investigation, it was observed that NazO and MgO promoters decreased the activity of
Ni/A1203 catalyst and these promoted catalysts also showed some degree of
deactivation. In contrast, CaO, La203 and Ce02promoters enhanced both activity and
stability of Ni/A1203 catalyst. The lower catalytic activities of Na20 and MgO
promoted catalysts are found due to the blockage of active sites. CaO, La203 and
Ce02 increased catalytic activity because they can enhance and stabilise the
dispersion of active sites.
Several researchers have reported the effects of various promoters on
carbon-supported Ni catalysts.
However, their conclusions vary depending on
reaction conditions. In Ni/MgO-CaO system it was found that smaller amount of
carbon formed compared to Ni/MgO catalyst, due to higher basicity induced by the
added CaO, which is expected to enhance adsorption of C02. Horiuchi et al. [1996]
found that added oxides of Na, K, Mg, and Ca markedly suppressed carbon formation
as a result of a decrease in the ability of Ni catalyst for C& decomposition. Chang et
al.( 1996) reported that K, Ca promoted zeolite-supported Ni catalyst exhibited
superior coke resistance for this reaction.
From researches reported in the literature and our own experimental results,
it is suggested that basic promoters behave to prevent coking on Ni catalysts.
However, it may increase carbon deposition when it makes Ni crystallite smaller
enough to induce high rate of C& dissociation .
Mechanisms and kinetics
Although there are different views on the reaction mechanism, it is generally believed
that the main steps are methane activation to form CH,(x=O-3) and the reaction
between CH, species and the oxidant, either in the form of oxygen adatoms originated
from C02 dissociation or C 0 2 itself (including C 0 2 activation). B a n g and
Verykios[l996] found that methane activation is a rate-determining step for the
formation of synthesis gas over the Ni/La203 catalyst, while the reaction between
457
G.Q. Lcr and S.Wang
surface carbon species( CH,, x=O) and the oxidant (including Cot activation) is the
rate-determining step over the Ni/y-A1203catalyst. Wang and Au [1996] studied the
isotopic effects of C&/CD4 in the carbon dioxide reforming of methane to syngas
over Ni/Si02 catalyst. They concluded that CH4 dissociation is rate determining and
C 0 2 dissociation occurs prior to surface reaction of CH, fragments. However,
investigations conducted by Slagtern et a1 [19971 showed that the reaction between
the surface carbon species from methane cracking and oxygen adspecies resulted fiom
Cot dissociation is the rate determining in case of Ni/Si02. Osaki et al. [1996]
employed pulsed surface reaction analysis (PSRA) to probe the mechanism of C02reforming of methane on supported nickel catalysts (Ni/Si02, Ni/MgO, Ni/A1203,and
Ni/Ti02). It was found that two reaction steps are responsible for H2 production, i.e.
dissociative C& adsorption to form CH, species and the subsequent surface reaction
of CH, and C02(or 0) is the rate determining step. Bradford and Vannice [1996b]
reported the reaction kinetics of C02-CH4reforming over various nickel catalysts and
proposed a reaction model based on CH4 activation to form CH, and CH,O
decomposition as the slow kinetic steps.
A simple power-law equation is most frequently employed to describe the
reaction kinetics. Some researchers have derived complex rate expressions. We
studied the reaction mechanisms and kinetics on Ni/y-A1203 and Ni/CeO2-AI2O3
catalysts and kinetics of carbon growth on Nily-AI203 and found that the activation
energies for both catalysts are similar. The mechanism is proposed as follows.
(7)
CH4+* + C b *
CO2+* + c02*
(8)
CH4* + CH3*+H
(9)
CH3* + CH2*+H
(10)
(1 1)
CH2* + CH* + H
CH* + C* + H
(12)
CH,* + Cot* + 2CO + d2H2 + 2*
(13)
H + H + H2
(14)
The rate -determining step is Eq.(13). The kinetic expression can be modelled by a
Langmuir-Hinshelwood type mechanism shown as below.
458
Synthesis gas production using carbon dioxide as carbon source
I ‘CH,
r=
‘CO,
(1+KIPCH4)(1+K?PC02)
(15)
Table 3. Reported Kinetic model for C02-CH4reforming reaction
Model
+ ‘H,O
kpcHI (‘CO,
Catalyst
Reference
Cu/Si02
Lewis et al., 1949
Ni foil
Bodrov et al. 1967
Bodrov et al., 1964
MA1203
kR KC& KCH4 ’CO, ’CH,
r=
(’ + KC02
‘CO,
+ KCH,
Paripatyadar, 1990
‘CH4
klPCH. pco,
r=
-
k _ , K p p(4-1)12
k2
CO H ,
+
.
k
(14-
k,
Richardson and
‘CH, )pCO,
Ni/MgO
Bradford and
Ni/Ti02
Vennice, 1996
Perspective
Carbon deposition mechanism
It is clear that carbon deposition or coking is the major deactivating factor influencing
catalyst activity and stability. Two or three types of carbon were found to form on Ni
catalysts. The growth mechanism and role of each carbon species in catalyst
deactivation are not well understood. It is believed that several factors such as metal
morphology and chemical properties affect the carbon structure. Recently, a structurebased model of carbon deposition on Ni/Si02 catalysts in C02-CH4 reforming has
459
C.Q.Lu and S. Wang
been put forward by Kroll et al. [1996]. This offers a new dimension into the carbon
deposition mechanism and modelling the coking process.
So far most work on carbon filament growth mechanism and kinetics are
centred on studies of methane cracking or CO disproportionation. In CO&I-&
reforming system, the behaviour of carbon filament growth will be expected to be
different. Hence, firher research on mechanism and kinetics of carbon formation in
C02-reforming of methane process is required in order to understand the complex role
of different carbon species formed and their influence on the catalyst activity and
stability for this reforming reaction. It is envisaged that improved catalysts of high
activity and long-term stability could be developed for low temperature coking-l?ee
operation when the carbon growth (or suppression) mechanisms are well understood.
A combination of controlling a number of factors is expected to be effective to
achieve carbon suppression or complete elimination. These include: temperature,
C02:C& ratio, catalyst support,
metal particle size and size distribution
(morphology).
Stability test and scale up
For the model catalyst exhibiting high catalytic activity, long-term stability and good
resistance to coking in laboratory-scale studies, a pilot investigation is necessary to
test the performance of the catalyst under industrial operating conditions. Only in
such a way, more information on product yield and life of the catalyst, and other
kinetic parameters can be obtained. For instance, we have developed a 5%wt Ni/yA1203and Ni/5%wtCeO2-AI2O3catalysts showing a high catalytic activity and a long
life in the microreactor testing. Tests are desired for a much longer time (generally
longer than 1000 hours) in pilot-scale reactor. Heat and mass transfer considerations
must also be incorporated in the pilot scale study. Reactor type and design, as well as
fluid flow in the reactor are other essential factors to be considered at that stage.
Concluding Remarks
In conclusion, there has been an ever-increasing interest in C 0 2reforming of methane
to produce syngas due to the growing concerns over greenhouse gas emissions and
460
Synthesis gas production using carbon dioxide us carbon source
global warming. While other mitigation options such as improving energy efficiency,
and ocean sequestering are widely open for exploration by key industrial and
governmental stakeholders, catalytic reforming of methane with C 0 2 does present a
promising technology which will yield potentially two-fold benefits. The key issue in
this process is catalyst coking at normal operating conditions, which hinders its
possible industrial application. Advances have been made in many respects on this
reaction in the recent literature and our own studies. However, much remain to be
better understood on the mechanisms and growth behaviour of carbon deposition
during the reaction. There is a complex array of factors influencing the catalytic
activity and stability, and coking process. We believe that only through further
fundamental research on the reaction and coking (carbon deposition) mechanisms,
and kinetics we could develop a high performance catalyst system that will deliver
high conversion or product yields at low temperature and with long-term coking-free
operation. We envisage that we are currently at a critical stage of technology in this
area, and a technological breakthrough on the catalyst development could be achieved
in the next five to six years making the process commercially viable within 10 years.
Acknowledgement: Financial support from the Australian Research Council and TIL
program, Department of Education is gratefully acknowledged.
References
Aparicio, L.M., J. Catal.,1997, 165,262-274.
Ashcroft, A.T., Cheethan, A.K., Green, M.L.H., andVemom, P.D.F.,Science, 1991,352, 225.
Bhat, R.N. and Sachtler, W.M.H., Appl. Cutal., 1997, 150,279-296.
Bitter, J.H., Hally, W., Seshan, K., van Ommen, J.G., and Lercher, J.A., Cutul. Toby, 1996,29,349-353.
Blom, R., Dahl, I.M., Slagtern, A., Sortland, B., Spjelkavik, A. and Tangstad, E., Cutal. T-,
1994,21,
535-543.
Bodrov, I.M. and Apel’baum, L.O., Kinet. Catal., 1967,8,379-382.
Bodrov, I.M., Apel’baum, L.O. and Temkin, M.I., Kinef. Catal., 1964,5696-705.
Bradford, M.C.J. and Vannice, M.A., Appl. Caml. 1996% 142,73-96.
Bradford, M.C.J. and Vannice, M.A., Appl. Cutul. 1996b, 142,97-122.
Chen, Y.G., Yamazaki, O., Tomishige, K. and Fujimoto, K., Cutul. Lett., 1996,36,91-95.
Cheng, Z.X., Wu, Q.L., Li, J.L., andZhu, Q.M., Cuiul. T+,
1996,30, 147-155.
Claridge, J.B., Green, M.L.H., and Tsang, S.C., Cutul. T W , 1994,21,455460.
Dibbem, H.C., Olesen, P., Rosbup-Nielsen, J.R., Tottrup, P.B. and Udengaard, N.R., Hydroc. Proc., 1986,
65, 71.
Erdohelyi, A., Cserenyi, J., Papp, E. and Solymosi, F., Appl. Cutul., 1994, 108, 205-219.
Erdohelyi, A., Cserenyi, J., Papp, E. and Solymosi, F., J. Cutul., 1993, 141, 287-299.
461
G.Q.Lu and S.Wang
Goula, M.A., Lemonidou, A.A., and Efstathiou, A.M.,J. Curul., 1996, 161,626-640.
Halliche, D., Bouarab, R, Cherifi, O., and Bettahar, M.M., Curul. Today., 1996,29,373-377.
Horiuchi, T., Sakuma, K., Fukui, T., Kubo, Y., Osaki, T. and Mori, T., Appl. Cuml., 1996, 144, 1 1 1-120.
Hu, Y.H. and Ruckenstein, E., Appl. Curd., 19974 154,185-205.
Ichikawa, S., Miyadera, H., and Nanba, M., Hifcchi Rev., 1993,42,255.
International Energy Agency (IEA), C w l Informulion, 1994.
Inui, T., Curul. To&, 19%,29,329.
Inui, T., Saigo, K., Fuji, Y., and Fujika, K., Cutul. Today., 1995,26,295-302.
Kim, G.J., Cho, D.S.,
Kim., K.H., and Kim, J.H., Curul. Lerr., 1994.28, 41-52.
Kroll, V.C.H., Swaan, H.M. and Mirodatos, C., J. Cuful.,1996, 161,409-422.
Lewis, W.K. Gilliland, E.R and Reed, W.AId. Eng.Chem., 1949,41,1227-1237.
Li, W.Y., Xie, K.C. and Xie, S.C., 1995,24,727-730.
Mark, M.F. and Maier, W.F., J. Cufal., 1996, 164, 122-130.
Masai, M.; Kado, H.; Miyake, A.; Nishiyama, S.; Tsuruya, S. Slud SUP$Sci. Cutul. 1988, 36,67-71.
Michel, S., Lind, Ag. Hjdroc. Proc., 1989,68(4), 37.
Mirzabekova, S.R.,Mamedov, A.Kh., Aliev, V.S., and Krylov, O.V., Kiner. Kurd., 1992, 33, 591-596.
Nakamura, J. and Uchijima, T., Shohbui, 1993,35,478-484.
Osaki, T., Horiuchi, T., Suzuki, K. and Mori, T., Appl. Curul., 1997, 155,229.
Osaki, T., Horiuchi, T., Suzuki, K. and Mori, T., J. Chem. Soc. Furuuiay Truns., 1996,92,1627-163 1.
Perera, J.S.H.Q., Couves, J.W., Sankar, G. and Thomas, J.M., Curd. Len., 1991, 11,219-225.
Qin, D. and Lapszewin, J., Curul. Today, 1994, 21, 551-560.
Reitmeier, R.E., Atwood, K., Bennett, H.A.Jr., and Baugh, H.M., Id. Eng. Chem., 1948, 40, 620.
Richardson, J.T. and Paripatyadar, S.A., Appl. Curul., 1990, 61, 293.
Ross, J.R.H., van Keulen, A.N.J., Hegarty, M.E.S., and Seshan, K., Curul. Today, 1996,30, 193.
Ruckenstein, E. and Hu, Y.H., J. Cuful., 1996, 161.55-61.
Ruckenstein, E. and Hu, Y.H., Appl. Cuml., 1995, 133, 149.
Sakai, Y., Saito, T., Sodesawa, T. and Nozaki, F., Reuct. Kinef. Curul. Lerr., 1984,24, 253.
Slagtem, A., Schuurman, Y., Leclercg, C., Verykios, X., and Mirodatos, C., J. Curd., 1997, 172, 118-126.
Solymosi, F., Kutsan, Gy.,Erdohelyi, A., Catal. Lett., 1991, 1 1 , 149.
Swaan, H.M., Kroll, V.C.H., Martin, G.A., andMirodatos, C., Curul. Todoy, 1994,21, 571-578.
Takano, A., Tagawa, T. and Goto, S., J. Chem.Eng.Jpn., 1994, 27, 123-731.
Takayasu, O., Hirose, E., Matsuda, N. and Matsuura, I., Chem.Express, 1991,6,447-450.
Tang, S.B.,Qiu, F.L. and Lu, S.J., Cutal. Today, 1995,22,253-255.
Teuner, S., Make CO from COz, Hydroc.Proc., 1985,64(5), 106.
Tokunaga, 0.and Ogasawam S., Reuct. Kinet. Cutul. Left.. 1989, 39, 69.
Tsipouriari, V.A., Efstathiou, A.M., and Verykios, X.E., Curul. Today, 1994, 21, 579-587.
Wang, H.Y and Au, C.T., CU&l. Lett., 1996,38,77-79.
Wang, S . Ph.D thesis, The University of Queensland, 1998.
Wang, S, Lu, G.Q., Tang, H.S., Proc. 23”’Austr. Chem. Eng. Con$ 1995,2,42.
Wang, S. and Lu, G.Q., Appl. Catal. 9 , 1997, in press.
Wang, S. Lu,G.Q., and Millar, G.J., Energy & Fuels, 1996, 10,896.
Xu, H.Y., Su, X.X., Fan, Y.M., Xu, G.L.,Liu, J.X., Yu,W.G., and Zhou, P.H., Perrol. Chem. Eng. 1992,
21, 147.
Yamazaki, O., Nozaki, T., Omat& K. and Fujimoto, K., Chem. Lerr., 1992, 1953.
Zhang, Z. L. and Verykios, X.E., Cutul. Todoy, 1994, 21,589.
Zhang, Z. L. and Verykios, X.E., J. Chem. Soc.,Chem. Commun. 1995,71.
Zhang, Z.L., Tsipouriari, V.A., Efstathiou, A.M., and Verykios, X.E., J. Curd., 1996, 158.51-63.
462
Документ
Категория
Без категории
Просмотров
1
Размер файла
824 Кб
Теги
production, using, current, research, synthesis, dioxide, source, gas, perspectives, carbon
1/--страниц
Пожаловаться на содержимое документа