114 Annesini

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Carbon Formation in the Steam Reforming Process: a

Thermodynamic Analysis Based on the Elemental


Composition
Annesini, M. C., Piemonte, V., Turchetti, L.
Dipartimento di Ingegneria Chimica, “Sapienza” Università di Roma
Via Eudossiana, 18 – 00184 Roma - Italy

A thermodynamic analysis of the conditions leading to carbon formation on the


catalytic bed of methane steam reformers is presented. In the first part of the analysis, a
closed system in which steam reforming reactions take place is considered, and
elemental compositions leading to carbon formation are determined. In the second part,
results of an industrial steam reformer simulation are used to locate, inside the reactor,
regions with a positive affinity for coke formation.

1. Introduction
Carbon formation on the reactor’s catalytic bed is a major problem in the methane steam
reforming process. Depending on the system conditions, side reactions such as methane
cracking or Boudouard reaction may occur at the catalyst surface, leading to carbon
deposition in different forms. Carbon may then encapsulate the catalyst surface or
dissolve and diffuse inside the catalyst pellet, reducing its activity. The Whisker
filamentous form of carbon is particularly noxious, because it can grow inside the
catalyst pellet and eventually break it, producing finer particles that alter the
fluidodynamics of the reactor.
In order to reduce the risk of carbon formation, industrial steam reformers are fed with
gas streams with a steam-to-carbon ratio of about 3, but important economical benefits
would be obtained if a lower steam-to-carbon ratio could be used. Therefore,
mechanisms and kinetics of carbon formation reactions have been thoroughly
investigated (Trimm, 1997; Snoek et al. 2002 ; Ginsburg et al., 2005), with the aim of
finding new process configurations that allow for the use of a lower steam content in
the feed.
The gas phase inside the reformer is, in general, made of five components (methane,
water, hydrogen, carbon monoxide, carbon dioxide), involved in many reactions, thus, it
is not easy to define synthetically the conditions that would favour carbon deposition,
even limiting the analysis to equilibrium conditions.
In this paper, a thermodynamic analysis based on the elemental composition of the
reacting system is presented. In the first part of the analysis, a closed system in which
steam reforming reactions take place is considered, and elemental compositions leading
to carbon formation are determined. In the second part, results of an industrial steam
reformer simulation are used to locate, inside the reactor, regions with a positive affinity
for carbon formation.
2. Thermodynamic analysis
2.1 Reactions
Several chemical reactions can take place during the steam reforming process. The most
important and thermodynamically probable ones are (Snoek et al., 2003):

CH 4 + H 2 O ↔ CO + 3H 2 (SR)
CH 4 + CO 2 ↔ 2CO + 2H 2 (DR)
CO + H 2 O ↔ CO 2 + H 2 (WGS)
CH 4 + 2H 2 O ↔ CO + 4H 2 (OSR)

and side reactions leading to carbon formation

CH 4 ↔ C + 2H 2 (MC)
2CO ↔ C + CO 2 (B)
CO + H 2 ↔ C + H 2 O (RG)

It can be easily shown that only 3 out of these 7 reactions are independent, thus, for the
following thermodynamic analysis only reactions (SR), (WGS) and (B) will be
considered.
Thermodynamic constants for the reactions as a function of temperature have been
calulated by ΔH and ΔG of reaction data at 298 K and specific heat c p (T ) of the
reactants. Graphite thermodynamic properties have been assumed for carbon.

2.2 Elemental composition


Since the 6 compounds considered contain only C, H and O atoms, the composition of
the reacting mixture can be conveniently represented in terms of only 2 independent
atomic fractions, defined as the ratio between the number of atoms of one type and the
total number of atoms in the mixture. I.e., the hydrogen atomic fraction in the gas
mixture is

4 yCH 4 + 2 y H 2 O + 2 y H 2
χH = (1)
5 yCH 4 + 3 y H 2 O + 2 y H 2 + 2 yCO + 3 yCO 2

where yi is the mole fraction of compound i in the mixture.


Compared to molar fractions, atomic fractions provide less, but still enough information
about the system composition to carry out a thermodynamical analysis of the carbon
formation conditions. On the other hand, atomic fractions offer many advantages in
terms of simplicity, because they don’t vary with the extent of reaction and allow the
composition range corresponding to carbon formation to be synthetically defined.
Fig. 1 Carbon formation conditions on the triangular diagram. The shaded area is the
carbon formation zone at 800 K. Limiting curves at 1000K and 1500K are also reported

2.3 Carbon formation conditions


When carbon is formed, at equilibrium the system is composed of 2 phases, with 6
reactants involved in 3 independent reactions. For such a system, the Gibbs phase rule
gives a variance v=3, that is, at given temperature and pressure, the molar fraction of
only one compound in the gas phase is independent, while the others may be calculated
by the thermodynamic constants of the reactions. Therefore, in the C-H-O plane, the
possible equilibrium gas phase compositions lie on a parametric curve of the type

⎧ χ H = χ H ( yi )
⎨ (2)
⎩ χ O = χ O ( yi )

where yi is the molar fraction of one of the gaseous reactants. Fig. 1 shows the
equilibrium curves obtained at different temperatures for a pressure of 15 atm. At high
temperature, all the reactions are shifted towards the formation of CO and H2, thus, the
equilibrium curve tends to the line joining the points representing these compounds.
The equilibrium curve divides the triangular diagram in two regions (fig. 1):

• CARBON+GAS: systems in this region (i.e., point A), at equilibrium are


constituted of a carbon solid phase (point C) and a gas phase (point B).
• GAS: systems in this region are homogeneous at equilibrium and no carbon is
formed
Fig. 2 Carbon formation conditions for a steam–methane system. The shaded area is the
carbon formation zone at 30 atm. Limiting curves at different pressures are also
reported.

Fig.3 Carbon formation conditions for a steam-methane-carbon dioxide system at 30


atm. The shaded area is the carbon formation zone when no carbon dioxide is
present. Limiting curves at different yCO2/(yCO2+yCH4) are also reported.
2.4 Steam - methane systems
Steam reformer feed is mainly composed of steam and methane, thus a particular focus
must be put on elemental compositions lying on the line that joins H2O and CH4 points.
The equilibrium curve intersects this line in a point whose position depends on the
system temperature and pressure (fig. 1). This point corresponds to a minimum value of
the yH2O/yCH4 ratio (steam-to-carbon ratio), below which, at equilibrium, carbon will be
produced.
In fig. 2, the limiting conditions are plotted in a temperature vs. steam-to-carbon ratio
plane, for different system pressures: points on the left of the curve represent conditions
in which carbon is formed. It can be noted that higher pressures slightly reduce the
possibility of carbon formation at all temperatures and that for high temperatures the
limiting steam-to-carbon ratio tends to 1, that is far below the value of 3 are used in the
industrial practice.

2.5 Steam - methane - carbon dioxide systems


Actual steam reformer feed contain other components such as carbon monoxide and
carbon dioxide, that, even if present in small quantities, increase the carbon content of
the system and can change significantly the coking limits. Furthermore, the analysis
could also be applied to the methane dry (CO2) reforming process, in which carbon
dioxide is one of the main feed components.
Therefore, in order to give a first evaluation of the effect of other carbon sources, the
feed was assumed as a mixture of water, methane and carbon dioxide. Fig. 2 shows how
the CO2 affects the coking limits. The effect is different depending on the temperature:
at high temperatures (>1000 K), the presence of CO2 reduces the carbon formation
zone, while for low temperatures the opposite effect is observed.
Again, a steam-to-carbon ratio of 3 is far away from the carbon formation conditions.

2. Affinity for coke formation in a fixed bed reactor


An exhaustive study on the carbon formation conditions can not be limited to the
previous analysis of equilibrium conditions. Considering a fixed bed reactor, local
compositions must also be checked to see if, in some points of the reactor, any side
reactions can thermodynamically proceed toward carbon formation.
To this aim, it is useful to define the affinity of each carbon forming reaction as follows
πi
ai = 1 − (3)
Ki
ν
where Ki is the thermodynamic constant of the i-th reaction and π i = Π j p j ji , where
p j is the partial pressure of component j in the reacting mixture and ν ji is its
stoichiometric coefficient in reaction i. It can be stated that, when ai is defined in a
point of the reactor and 0 < ai ≤ 1 , carbon will be produced by reaction i in that point.
In order to locate the critical regions for carbon deposition, simulations of a steam
reformer (length 12 m, internal diameter 12.6 cm) were carried out using a steady-state,
bidimensional model (De Falco et al., 2007). The temperature, pressure and
composition distributions obtained were subsequently used for affinity calculation.
Inlet stream conditions:

Temperature: 793 K
Pressure: 3.03 MPa
Molar flow-rate: 25.7 kmol/h
Molar composition:
CH4 31.15 %
H2O 62.30 %
H2 4.68 %
CO 0.31 %
CO2 1.56 %

Fig. 4 Affinity (eq. 3) for carbon formation by (MC) reaction in a fixed bed steam
reformer. Inlet stream conditions are reported next to the figure.

It must be pointed out that the model used doesn’t take into account carbon forming
reactions: this approach is based on the reasonable hypothesis, also assumed by Snoek
et al. (2003), that carbon formation has a negligible effect on mass and energy balances
inside the reactor.
Fig. 4 shows a shade plot of carbon formation affinity of reaction (MC), which, at the
conditions of the simulation, is the most important carbon formation reaction among
those considered. For the sake of clarity, only positive values of affinity have been
plotted, while negative affinities are plotted as zero. The plot is limited to the first 2 m
after the catalytic tube entrance, because a positive affinity for carbon formation is
obtained only in the first layers of the catalytic bed.

Acknowledgements
The authors whish to thank Prof. Luigi Marrelli and ing. Marcello De Falco for their
useful help. This work was financially supported by MIUR (FISR DM 17/12/2002).

References
De Falco M., L. Di Paola, L. Marrelli and P. Nardella, Chem. Eng. J., in Press
Ginsburg J. M., J. Pina, T. El Solh, and H. I. de Lasa, H. I., 2005, Ind. Eng. Chem. Res.,
44(14):4846-4854.
Snoek J., G. F. Froment, M. Fowles, 2002, Ind. Eng. Chem. Res., 41(17):4252-4265
Snoek J., G. Froment and M. Fowles, 2003, Int. J. Chem. Reac. Eng., 1:A7
Trimm, D. L., 1997, Catal. Today, 37, 233-238

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy