Journal of CO2 Utilization 10 (2015) 12–22

Contents lists available at ScienceDirect

Journal of CO2 Utilization

journal homepage: www.elsevier.com/locate/jcou

A model-based analysis of CO2 utilization in methanol synthesis plant

Dia Milani a, Rajab Khalilpour a, Gholamreza Zahedi b, Ali Abbas a,*
a
School of Chemical and Biomolecular Engineering, The University of Sydney, NSW 2006, Australia
b
Ozone Solutions, Huntington, West Virginia Area, USA

A R T I C L E I N F O A B S T R A C T

Article history: Utilizing the greenhouse gas CO2 as a feedstock in chemical processing could offer alternative solutions
Received 9 November 2014 to long-term storage. Large-scale production of light hydrocarbons such as methanol (MeOH) is one of
Received in revised form 9 February 2015 the predominant and sensible schemes for such utilization. This proposal will not only recycle the CO2
Accepted 17 February 2015
gas within methanol synthesis process, but will also reduce the uptake of raw materials such as natural
Available online
gas (NG) and reduce the greenhouse-gas (GHG) emissions of a comparable stand-alone NG-based
methanol synthesis plant.
Keywords:
In this paper, a comprehensive model for CO2 integration in NG-based methanol synthesis plant has
CO2 utilization
been developed. The reformer product (syngas) is mixed with the high-purity CO2 stream that comes out
Methanol synthesis
Carbon capture and sequestration of power-plant carbon capture (PCC) process. It is found that this integration may reduce methane
Process integration uptake by 25.6% and decrease the combined CO2 emissions for both power-plant and MeOH-plant by
Lurgi reactor 21.9%. The energy intensity for this integration is 33.45 GJth/tonneMeOH and 0.64 GJel/tonneMeOH. The
Steam reformer energy efficiency of this integration is 59% and the product to feed ratio is 2.27 tonneMeOH/tonneCH4
higher than 1.69 tonneMeOH/tonneCH4 calculated for a comparable standalone NG-based methanol
synthesis plant with 68% efficiency.
ß 2015 Elsevier Ltd. All rights reserved.

1. Introduction either improving the energy efficiency of various industrial

processes or shifting to non-fossil energy sources (renewable and
Economic and population growth continue to be the most nuclear). Traditionally, most of the world is heavily reliant on fossil
important factors behind increases in CO2 emissions from fossil fuel sources for their energy demand and trillions of dollars have
fuel combustion. The escalation of anthropogenic CO2 emissions already been invested in existing energy infrastructures. Shifting
linked to global warming has attracted global efforts to find new away from fossil fuel energy sources will require a huge investment
and better ways of meeting the world’s growing energy demand, in other resources and will also require long transitional periods.
while simultaneously reducing greenhouse-gas (GHG) emissions Employing CCS to reduce CO2 emissions would allow continued use
[1]. Despite a growing number of climate change mitigation of fossil energy, while buying time to ease the transition to clean
initiatives and policies, global GHG emissions grew on average by energy sources systematically. However, there is a strong debate on
2.2% carbon dioxide equivalent (CO2eq) per year from 2000 to the feasibility of CCS in general and the high energy penalty imposed
2010 compared to 1.3% CO2eq per year from 1970 to 2000 by such processes. This debate is more intensified on CO2
[2]. Carbon dioxide emissions from fossil fuel combustion and sequestration part where the gas is arguably considered as a waste.
industrial processes contributed about 78% of the total GHG Uncertainty is also associated with the eternity of the stored CO2 and
emission increase from 1970 to 2010. In 2010 fossil fuel-related the high cost involved [3–6]. Conversely, CO2 can be captured and
CO2 emissions reached 32 (2.7) Gt/y, and grew further by about utilized (CCU) as a feedstock or commodity which can be converted
3% between 2010 and 2011 and by about 1–2% between 2011 and to a variety of commodity chemicals, fuels and building blocks for
2012 [2]. other products [7].
Large-scale carbon capture and sequestration (CCS) is consid- One of the transitional strategies is to use NG for power
ered as one of the effective emissions avoidance mechanisms. generation. Natural gas is the fastest growing primary energy
Other emissions avoidance mechanisms can also be achieved by source where the global consumption has increased from
118.4 trillion cubic metre (TCM) in 1993 to reach 185.7 TCM at
the end of 2013 [8]. The geographical mismatch between most NG
* Corresponding author. Tel.: þ61 2 9351 3002; fax: þ61 2 9351 2854. resources and markets usually involves higher handling and
E-mail address: ali.abbas@sydney.edu.au (A. Abbas). transportation costs in comparison with transportation of liquid

http://dx.doi.org/10.1016/j.jcou.2015.02.003
2212-9820/ß 2015 Elsevier Ltd. All rights reserved.
 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22 13

resulting from biomass gasification or by hydrogenation of carbon

Abbreviations and symbols
dioxide captured from flue gases of large industrial processes [13].
This paper presents and analyses the strategy of CO2 utilization
ASU air separation unit with methanol synthesis process to generate liquid methanol at
ATR auto-thermal reforming lower energy and GHG emissions intensities. By blending the CO2
BAU business as usual outlet of a carbon capture plant retrofitted at the end of a coal-fired
CCS carbon capture and sequestration power plant (PP), this study shows how the NG feedstock can be
CCU carbon capture and utilization reduced for the equivalent amount of produced MeOH. This
CO2eq carbon dioxide equivalent practice will not only reduce the GHG emissions due to the extra
FT-GTL Fischer–Tropsch gas to liquid NG portion, but also will have the CO2 stream recycled in the
methanol production chain at lower net energy intensity (GJ/
GHG greenhouse-gas
tonneMeOH). For this purpose, a 660 MWel base case power plant is
GTL gas-to-liquid
considered. The flue gas of this PP is treated in post-combustion
KTPA kilo-tonnes per annum carbon capture plant and the outlet CO2 stream is blended with a
LHHW Langmuir–Hinshelwood–Hougon–Watson kinetics syngas stream that is coming from an existing and detached steam-
MEA monoethanolamine methane reforming (SMR) process (Fig. 1). These two streams enter
MeOH methanol a methanol synthesis reactor to generate MeOH.
MTBE methyl tertiary-butyl ether In this paper, an assessment for CCS and CCU technologies is
NG natural gas presented and the diverting point from CCS to CCU is identified. A
PCC power-plant carbon capture comprehensive model of CO2 utilization paired with methanol
POX partial oxidation synthesis process using Aspen Plus V8.4 (Aspentech, USA) is
developed in this work. The flowsheet parameters are explained
PP power-plant
and the operational conditions of the predominant unit operation
RWGS reverse water–gas shift reaction
are analysed. The mass and energy balance for the entire system is
SMR steam-methane reformer
discussed and the results are compared with an equivalent
S/M steam to methane ratio standalone NG-based methanol synthesis plant. The overall
TCM trillion cubic metre efficiency, the energy intensity and the CO2 emissions for both
WGS water–gas shift reaction cases are highlighted and discussed.
E activation energy (kJ/kmol)
k pre-exponential factor 2. Sequestration and/or utilization
kb reversible reaction coefficient
kf forward reaction coefficient The harvest of a typical carbon capture processes is a highly
M stoichiometric number concentrated and high-pressurized CO2 stream ready for seques-
tration procedures. Gaseous CO2 is typically compressed to a
r rate of reaction
pressure more than 110 bar in order to avoid two-phase flow
R gas law constant (kJ/kmol K)
regimes and increase the density of the CO2 which makes it easier
T absolute temperature (K) and less costly to transport. Depending on the storage option,
several factors may affect CO2 transportation and storage cost
including the amount of CO2, the transport distance, the storage
petroleum. Fischer–Tropsch gas to liquid (FT-GTL) is a well- medium (gas field, oil field, aquifer or ocean bed) and the depth of
developed technology that converts NG to a variety of liquid fuels the storage. Among several CO2 storage options, three major
[9]. The FT-GTL technology may provide a sensible solution to options can be named; geological, oceanic and chemical storage.
avoid the larger expense of gas transport by transforming this gas Detailed analyses of the cost and benefits, advantages and
into liquid fuel which is easier to confine, control and transport. disadvantages of these options can be found in literatures
Global FT-GTL demand for NG is projected to increase from [3,14–16].
4 billion m3 in 2000 to 170 billion m3 in 2030 [10]. One possible There is a growing recognition that CO2 sequestration which
outcome of the GTL process is correlated with light hydrocarbon treats CO2 as a waste stream to be sequestered may not be the
products which can be regarded as building blocks for more silver bullet for the reduction of GHG emissions. Alternatively, CO2
complex hydrocarbons such as gasoline and diesel. can be viewed and utilized as a valuable feedstock or commodity.
Methanol (methyl alcohol, CH3OH or MeOH) is the simplest Historically, CO2 has been directly used in many useful chemicals
liquid hydrocarbon that can be regarded as a fuel, a hydrogen such as urea, dry ice and polycarbonates (Fig. 2), but this small-
carrier or a feedstock for producing more complex chemical scale utilization is incomparable with the large-scale CO2
compounds. Methanol production can be achieved by using fossil production from fossil fuel combustions. Some of these products
fuel sources (NG, coal) or renewable sources (biomass, waste are commodity chemicals, fuels and building blocks for other
wood, atmospheric CO2). More than 80% of the MeOH produced products (Fig. 2). A productive utilization of CO2 as a feedstock for
worldwide is obtained from NG [11]. However, the process of liquid fuels may provide an attractive alternative over sequestra-
producing MeOH from NG only is often associated with extremely tion. This also would help to satisfy part of the increasing demand
high GHG emission. In common practice, each GJ of produced for conventional petroleum. Therefore, recycling CO2 as a fuel
MeOH will emit between 17 and 72 kg of CO2 excluding the CO2 feedstock may provide a reasonable opportunity to consume CO2
equivalent of CH4 loses and the energy consumed in the NG at a scale proportionate with its production by conversion to liquid
extraction, purification and transport. Part of the NG input is hydrocarbon fuels [7,17]. Multiple processes can be combined to
contained in the energy content of the produced MeOH utilize CO2 with syngas, a mixture of hydrogen (H2) and carbon
(18.82 kgcarbon/GJMeOH) and not resulting in CO2 emissions while monoxide (CO), including methanol synthesis reaction. The
the other part is used as process fuel causing CO2eq emissions production of methanol from syngas is a mature process and
[12]. Two possible pathways of reducing CO2 emissions and there have been numerous studies into the synthesis and
producing ‘‘renewable methanol’’ are either by the syngas commercialization of methanol [11,18–20].
14 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22

Fig. 1. Schematic of various methanol production methods.

Methanol is one of the most important raw chemical materials. energy carriers [11]. In both cases, hydrogen or oxygen sources
About 85% of the methanol produced is used in the chemical must be provided through water electrolysis or air separation unit
industry as a building block or solvent for synthesis and the (ASU). Water electrolysis or ASU both have high demand for
remainder is used in the fuel and energy sector [21]. Methanol is electrical energy that is often associated with additional GHG
mainly used in the production of formaldehyde (35%), methyl emissions unless it comes from renewable sources (Fig. 1).
tertiary-butyl ether (MTBE) (25%) and acetic acid (9%), respectively
[22]. Methanol can also be blended with different grades of 3. CO2 stream in the feed of methanol synthesis
gasoline for existing automobiles and hybrid flexible vehicles. One
big advantage is that methanol can be produced in the same Methanol synthesis is a mature process developed from a
quality from fossil raw materials and from renewable primary steady-state kinetic model for hydrogenation of CO2 and reverse

Fig. 2. Possible CO2 utilization trends in various products and procedures.

 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22 15

water–gas shift (RWGS) reactions on a commercial Cu/ZnO/Al2O3 plant and blended with syngas stream that is coming from a
catalyst. The ideal gaseous composition for methanol reactor feed detached SMR plant (Fig. 3). The mixing ratio is adjusted to satisfy
is expressed based on stoichiometric number M, where M defined stoichiometric number M  2 which determines the molar flow of
in Eq. (1) is equal to the difference of H2 and CO2 molar fractions CH4 and H2O at the feed of SMR as well as the optimal steam to
divided by the sum of CO and CO2 molar fractions and ideally methane (S/M) ratio. These two streams (syngas and CO2) are then
should be equal to 2 [11,20]. After SMR unit, if syngas components thermally homogenized before entering into the methanol
are blended with CO2 stream at specific molar compositions that synthesis plant (Fig. 4).
satisfy (M  2) and passed through a Cu/ZnO/Al2O3 catalyst
reactor, methanol can be produced. 4. Process flowsheet description

MH2  M CO2 At the PCC outlet and before the last stage of intercooling and
M¼ (1)
M CO þ M CO2 pumping for sequestration (Fig. 3), a CO2 stream of 0.985 mole
fraction at 74.9 bar and 116 8C is diverted into the methanol
The fundamental scenario for this paper is built on a 660 MWel synthesis process (stream 1 Fig. 4). This stream is compressed to
coal-fired power plant (PP) burning pulverized black coal. In base- 78 bar and the temperature rises up to 120 8C before entering a co-
case operation conditions, the flue gas of this PP contains current heat exchanger (HX2) with stream 15 that is coming from a
590.1 tonne/h of CO2 with 13.0 v% concentration. Because of the detached SMR plant. At the SMR plant, the main water (stream 4) is
electrical market demand fluctuations, the PP load varies in the compressed to 25 bar and preheated within the heat exchanger
range of 50–100% of the design capacity. The PP has a retrofitted HX1 by the high temperature stream that is coming out of the
PCC unit with a design capacity of taking 50% of the PP flue gas at reformer. The efficiency of extracting and purifying NG depends on
full capacity. This arrangement guarantees that the flow to PCC will the field pressure and gas quality. When impurities are low, and
not fluctuate. Clearly, if the PP works on partial load down to 50% of field pressure is high, the energy demand can be negligibly small
the nominal load, the PCC can still have the same uptake which will [24]. In the present study, the CO2 and other impurities
be equal to the entire flue gas mass flow rate out of the PP stack. concentration in NG assumed to be negligible and the NG stream
This configuration helps to stabilize the operation conditions at the consists of purified methane only. Purified methane (stream 7) is
PCC and any plant to be retrofitted thereafter. The PCC uses a also preheated in the heat exchanger HX1 to be balanced in
30 wt% monoethanolamine (MEA) solvent and able to capture 90% pressure and temperature with the inlet steam. Both streams
of CO2 emissions (265.9 tonneCO2 /h). The gas exiting from the top (steam and methane) are mixed and heated within a preheater to
of the stripper (out of PCC) mainly consists of CO2 (78.7%) and the adjust the temperature up to 1100 8C, the appropriate operational
rest is H2O and other inert gases. In usual practice, to prevent temperature in the reformer. The high-temperature high-pressure
corrosion at the pipelines and compressor blades, the gas must be stream (stream 9) enters the reformer, a rigorous multiphase
dehydrated on several stages prior to final compression and equilibrium reactor based on Gibbs free energy minimization. The
transport. The relatively pure CO2 (98.5 wt%) is then compressed reformer RGibbs reactor is set free to consider all components as
(usually above 100 bar) before being exported for sequestration products and calculate phase and chemical equilibriums at 1100 8C
[23]. In this paper, before the last stage of compression procession, temperatures and 25 bar pressure. The reformer outlet (stream 10)
this stream is diverted towards an existing methanol synthesis enters the heat exchanger (HX1) where it loses most of its enthalpy

Fig. 3. Schematic of CO2 stream compression and dehydration stages towards sequestration fate and showing the diversion point of this stream towards methanol synthesis
process as proposed in this study.
16 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22

Fig. 4. The Aspen Plus SMR process model. The produced syngas is cooled, dehydrated, compressed and blended with concentrated CO2 stream.

for heating up water and methane feed streams as mentioned In methanol synthesis plant (Fig. 5), the inlet (stream 17) is re-
earlier. The outlet (stream 11) is further cooled down for mixed with a recycle stream (stream 29) in a second mixer (MIX2).
dehydration purpose in water knock-out drum (DRUM1) where The outcome (stream 18) is then heated in a counter-current heat
95% of water content and impurities are drained away. Stream 14 is exchanger (HX3) to 210 8C and injected into the RPlug reactor
compressed in CP2 (at 80% polytropic efficiency) to 78 bar before (RPLUG) where methanol synthesis reactions take place. The gases
being thermally homogenized with incoming CO2 stream (stream leaving the reactor (stream 20) are divided in a splitter (SPLT1) into
2). Both streams at vapour phase are mixed in MIX1 and the outlet two streams: the first (75% of initial current) is used in heating up
(stream 17) with stoichiometric number M = 2.00167 is sent to the fresh feed (stream 21), while the second (stream 24) is used in
methanol synthesis plant (Fig. 5). heating up the feed of the distillation column (stream 34) in

Fig. 5. The Aspen Plus model of the methanol synthesis process.

 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22 17

another counter-flow heat exchanger (HX4). The outcome of these hydrocarbons, representing a considerable heating value of the
two heat exchangers (HX3 and HX4) brought to a similar pressure gas. This heating value is utilized in the heat exchanger (HX1) to
value of 73.6 bar and mixed in a mixer (MIX3). The outcome of preheat the feedstock.
MIX3 (stream 26) is cooled down to 45 8C and entered a phase
separation process in (DRUM2). The gaseous phase (stream 28) is
compressed again up to 78 bar in (CP3) and re-mixed with the feed 2CO , CO2 þ C ðDH300 K ¼ 172:6 kJ=molÞ (4)
stream (stream 17) as mentioned earlier. The liquid crude
methanol (stream 30) composed of methanol, water and residual Having a pre-determined CO2 stream (stream 3) that is coming
dissolved gases is drained and expanded in two pressure relief from PCC plant, the molar ratio of steam and methane are
stages in (VLV1 and VLV2) to reduce the pressure down to 1.2 bar. sensitized to obtain the ideal compositions for Eq. (1). Fig. 6 shows
The outlet (stream 32) is then entered a purge flash (DRUM3) to how the stoichiometric number M responds to the variation of the
remove non-reactant gases at 43 8C temperature and 1.2 bar steam and methane to CO2 molar ratio at the feed streams. It was
pressure (stream 33). The drained liquid (stream 34) is preheated found that the lowest molar flow of water and methane that can
to 90 8C in HX4 as mentioned earlier and fed to the distillation accommodate the pre-determined CO2 stream and achieve M =
column (RADFRC). The water comes out from the bottom of the 2.00167 is 21,000 kmol/h for both at S/M ratio of 1. Higher molar
distillation column (stream 36) at 82 8C, while the methanol flows of steam and methane can also achieve similar M ratios, but
(99.7 wt%) comes out of the top of the distillation column (stream that will be at the expense of higher energy consumption in the
37) in gaseous phase and then cooled down in COOL3 to liquefy at preheater, the reformer and at the successive compression and
42 8C (stream 38). In the above-mentioned sequence, three major intercooling stages. Increasing the S/M ratio will significantly
blocks are crucial for the overall methanol production efficiency. reduce CO and increase CO2 formation at higher temperatures
These major blocks are the reformer, the reactor and the which undermines the pre-determined CO2 fraction that is coming
distillation column. These are discussed next. from the PCC.

4.1. The reformer 4.2. The reactor

Three main technologies that have been developed for syngas Current industrial reactors for methanol production, primarily
production are steam-methane reforming (SMR), partial oxidation differ in the reactor design. The two main types are categorized to
(POX) and auto-thermal reforming (ATR). For NG-based syngas be either quench adiabatic (e.g. ICI) or quasi-isothermal (e.g. Lurgi).
production, SMRs are often used because these units are abundant The former accounts for 60% while the latter is used in 30% of
and relatively cheaper [25,26]. The methane feedstock is catalyti- worldwide methanol production [21]. Unlike quench adiabatic
cally cracked over a Nickel catalyst with the addition of steam and reactors, Lurgi (adopted in this model) is a tubular reactor
in the absence of Oxygen. The SMR reaction proceeds in two steps, consisting of shell and tube design similar to Fischer–Tropsch
the reforming reaction which is strongly endothermic (206.3 kJ/ configuration. The tubes contain a proprietary Lurgi methanol
mol) and leads to an increase in volume (Eq. (2)) and the WGS catalyst and are surrounded by boiling water for reaction heat
which is slightly exothermic (41.2 kJ/mol) and proceeds without removal. These tubes operate in low pressure regime at 50–100 bar
change in volume (Eq. (3)). The excess heat is often provided from and temperatures around 230–265 8C [22]. The temperature of
an external source [21]. Lurgi reactor can be simply controlled by varying the pressure of
the boiling water. The outlet steam can be utilized in other process
such as compensating part of the distillation column reboiler duty.
CH4 þ H2 O , CO þ 3H2 ðDH300 K ¼ 206:3 kJ=molÞ (2) In this model, each tube is 10 m length and 16 mm diameter
packed with Cu/ZnO/Al2O3 commercial catalysts that have a bed
CO þ H2 O , CO2 þ H2 ðDH300 K ¼ 41:2 kJ=molÞ (3) voidage of 0.4 and the particle density of 1775 kg/m3 [30]. A
detailed kinetic model for methanol production using this catalyst
The degree of conversion of methane increases with tempera- is firstly proposed by Bussche et al. [31]. Mignard and Pritchard
ture elevation, increasing partial pressure of steam, and decreas- [32] have readjusted the parameters to reflect a better precision for
ing absolute pressure. The production of syngas components (H2
and CO) is highly influenced by S/M ratio at any particular
temperature. Depending on the feed compositions, the reforming
process ultimately produces syngas with a H2/CO ratio in the
range of about 2.2–4.8 [27]. At a constant reaction temperature,
higher S/M ratio will results in higher H2 production and lower
methane mole fraction. However, higher concentrations of the
H2O in the feed mixture decrease the likelihood of CO formation
and increase the likelihood of CO2 formation especially at the
higher range of reaction temperatures (>627 8C) [28]. At higher
temperatures where reforming and WGS reactions prevail, the
dependence of CO2 formation on the amount of steam is quite
strong which results in higher CO2 concentration obtained at the
higher S/M ratio. Conversely, low S/M ratios and low temperature
reactions could generate free solid carbon as a product which
causes major problems in the reformer reactions and subsequent
unit operations [29]. It is crucial to eliminate any production of
solid carbon in the reformer reactions (i.e. S/M ratio less than
3 may generate solid carbons (Eq. (4)) at 727 8C or lower) [28]. At
the end of reforming unit, the products can still contain
a considerable amount of steam, methane and other light Fig. 6. The effect of SMR feeds to CO2 ratios on the stoichiometric number M.
18 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22

both methanol production and reverse water–gas shift (RWGS) represent the experimental results of Klier et al. [36]. They also
reactions [33,34]. The gas phase reaction stoichiometry and kinetic extended the range of applicability of the model to pressures up to
expressions for this catalyst are following Langmuir–Hinshel- 75 bar [32]. The derivation of the corresponding set of kinetic
wood–Hougon–Watson (LHHW) kinetics. For rate-controlled equations is performed by Bussche et al. under the hypothesis of
reactions, Aspen Plus provides a built-in LHHW kinetics for pseudo-steady-state of the concentration of the different surface
calculating the rate of the reaction. The general LHHW kinetic intermediates and the following expressions are obtained for the rate
expression consists of a kinetic factor, a driving force expression, of methanol synthesis and the reverse water gas shift reaction [31].
and an adsorption term (Eq. (5)). All reactions are occurring in the
vapour phase and the reaction rate is based on catalyst weight. The k1 PCO2 PH2  k2 PH2 O P CH3 OH PH22
concentration basis for the driving force is partial pressure r CH3 OH ¼  3 ðkmol=kgcat sÞ (12)
powered by the concentration exponents for forward and 1 þ k3 PH2 O PH12 þ k4 PH0:5
2
þ k5 PH2 O
backward reactions (terms 1 and 2), respectively.
k6 P CO2  k7 P H2 O P CO PH12
r RWGS ¼ ðkmol=kgcat sÞ (13)
1 þ k3 P H2 O PH12 þ k4 PH0:5
2
þ k5 P H2 O
ðKinetic factorÞðDriving forceÞ Van-Dal and Bouallou converted the parameter values for the
r¼
ðAdsorption expressionÞ steady-state kinetic model into a compatible kinetic model for
  Aspen Plus [30]. The equations of the thermodynamic equilibrium
E fk f ½W½X  kb ½Y½Zg were incorporated into the kinetic constants and the units of the
r ¼ k exp (5)
RT ðAdsorption expressionÞ equations were modified to match Aspen Plus requirements
In reversible reactions where kb does not equal zero, the general [30]. Furthermore, the same catalyst is also functional for CO and
term of the reaction can be simplified by merging the kinetic factor CO2 methanation and WGS reaction as reported by Choi and
into the driving force coefficients kf and kb setting k=1 and E=0 in Stenger [37] and has been demonstrated by Robinson [38]. The
the kinetic tab and updating kf and kb values in the driving force kinetic mechanism proposed by Xu and Froment [39] use the
term 1 for the reactants and 2 for the products, respectively. components CO, CO2, H2, H2O, and CH4 similar to Eqs. (2) and (3) in
addition to the following reaction:

fk f ½W½X  kb ½Y½Zg CH4 þ 2H2 O , CO2 þ 4H2 ðDH300 K ¼ 41:21 kJ=molÞ (14)
r¼ (6)
ðAdsorption expressionÞ
As the feed stream for the reactor might have a small fraction of
The equation for each temperature dependent expression kf and CH4 and H2O that come from imperfect reforming reactions, these
kb as the following: components may react again in this reactor and influence the
formation of various products. The reaction rate for the water gas
shift, CO methanation and CO2 methanation are shown below:
Bi
Inðki Þ ¼ Ai þ þ C i lnðTÞ þ Di T (7)
T

or k ðPCO P H2 O PH12  k5 PCO2 Þ
r WGS ¼ 2
ðkmol=kgcat sÞ
ð1 þ k1 PCO þ k2 PH2 þ k3 PCH4 þ k4 P H2 O PH12 Þ
  (15)
Bi C i
ki ¼ expðAi Þexp T expðDi TÞ (8)
T

k ðP CH4 P H2 O PH2:5
2
 k6 P CO PH0:5
2
Þ
Where the values of C and D are usually insignificant and can be r COCH4 ¼ 2
ðkmol=kgcat sÞ
ignored. For the adsorption expression, the adsorption expression ð1 þ k1 P CO þ k2 P H2 þ k3 P CH4 þ k4 PH2 O PH12 Þ
exponent is specified at the top and the expression exponent for (16)
each coefficient is entered in a separate term as they appear in the

equation. Then the adsorption coefficients similar to Eq. (8) are also k ðP CH4 PH2 2 O PH3:5
2
 k7 P CO2 PH0:5
2
Þ
entered depends on the number of terms. Three overall reactions r CO2 CH4 ¼ 2
ðkmol=kgcat sÞ
ð1 þ k1 P CO þ k2 P H2 þ k3 P CH4 þ k4 PH2 O PH12 Þ
are projected in the methanol synthesis namely hydrogenation of
carbon monoxide, hydrogenation of carbon dioxide and the reverse (17)
water–gas shift reaction. In this kinetic model, two independent The WGS is already incorporated in Bussche et al. model
reactions (hydrogenation of carbon dioxide and reverse water–gas (Eq.(13)) and can be ignored at this stage because it has no
shift reaction) out of the three following dependent reactions are influence on the overall reaction outcomes. Robinson [38] applied
considered: the parameter values for the steady-state kinetic model into a
compatible kinetic model for Aspen Plus and incorporated the
CO þ 2H2 , CH3 OH ðDH300 K ¼ 90:77 kJ=molÞ (9) thermodynamic equilibrium into the kinetic constants matching
Aspen Plus requirements. Although the production of by-products
CO2 þ 3H2 , CH3 OH þ H2 O ðDH300 K ¼ 49:16 kJ=molÞ (10) was not considered in this model, 0.4% of by-products can be
expected mainly in the form of methyl-formate [30]. Our model is
validated against Rezaie et al. [40] model which was also validated
CO2 þ H2 , CO þ H2 O ðDH300 K ¼ 41:21 kJ=molÞ (11)
in comparison with historical process data over a period of
The model assumes that the CO2 is the main source of carbon for 1200 operating days at Shiraz petrochemical methanol plant. The
the synthesis of methanol and that there is negligible inhibition of flow compositions and the reactor specifications are presented in
the reaction represented by Eq. (10) as demonstrated by Sahibzada Table 1. Fig. 7 shows our model performs satisfactorily against
et al. [35]. In addition, the model considers the inhibitory effect of Rezaie et al. model. The MeOH mole fraction trendline at exit
water formed by the RWGS reaction. The activation energies of stream satisfactorily approaches Rezaie et al. model at longer
reactions were readjusted by Mignard and Pritchard [32] to better tubes.
 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22 19

Table 1 feed gas, the type and lifetime of the catalyst. Most of the unreacted
Feed compositions and reactor specifications in Rezaie et al. model [40].
gases are purged out in DRUM3 (stream 33). The crude methanol
Feed comp. Mole Reactor specifications that enters the distillation column mainly contains methanol and
frac. (%) water and may contain a small fraction of low-boiling and high-
CO2 9.4 Molar flowrate per 2.412 kmol/h boiling components. The low-boiling components include remain-
one tube ing dissolved gases, dimethyl ether, methyl formate, and acetone.
CO 34.6 Inlet temperature 230 8C The high-boiling components include higher alcohols, long-chain
H2 65.9 Pressure 76.98 bar
hydrocarbons, higher ketones, and esters of lower alcohols with
H2O 0.04 Tube length 7.022 m
MeOH 0.5 Number of tubes 2962 formic, acetic, and propionic acids. Higher waxy hydrocarbon
CH4 10.26 Tube diameter 10 mm compounds are also formed in small amounts. They have low
N2 9.3 Catalyst density 1770 kg/m3 volatility and thus remain in the distillation bottoms, where they
can easily be removed because of their low solubility in water and
low density [21]. The column is simulated with the rigorous model
RadFrac in Equilibrium mode of Aspen Plus. The distillation
column consists of 44 rectification and 13 stripping stages. The
condenser type is partial condenser with vapour distillate only
and water is used as an energy recovery fluid. The reboiler type is
kettle with standard convergence that is recommended for most
two-phase columns. To ensure high quality methanol (>99.5 wt%)
and to prevent water presence in the methanol stream, the reflux
ratio is set to 0.9 and the distillate to feed ratio is 0.75. Increasing
distillate to feed ratio not only will not have significant increase of
methanol production, but also will let water molecules to
evaporate with methanol which will increase the condenser duty
and reduce methanol purity in the product stream (stream 37).
Similarly, decreasing reflux ratio further may reduce the reboiler
duty, but will waste high quantities of methanol in the water
stream (stream 36).

5. Results and discussions

Fig. 7. Methanol mole fraction at the exit of Lurgi reactor as a function of tubes
The mass and energy balance for CO2 utilization in methanol
length for this study and Rezaie et al., respectively.
synthesis (this study) is compared with an equivalent standalone
NG-based methanol plant (with no CO2 utilization) that produces
4.3. Distillation column similar quantity of MeOH. In this comparison, three scenarios for
CO2 emission may arise (Fig. 8). In business as usual scenario (BAU
Crude methanol leaving DRUM2 contains water, unreacted scenario), the stand-alone NG-based methanol synthesis plant
gases and other impurities (stream 30). The amount and operates discretely to the power plant. In the second scenario, the
composition of these impurities depend on the reaction conditions, stand-alone NG-based methanol synthesis plant works as usual

Fig. 8. Three different scenarios for CO2 emissions; A – business as usual (no PCC), B – with CCS (PCC product is sequestered), C – with CCU (PCC product is utilized in the
methanol synthesis plant).
20 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22

Table 2 are required. This is an equivalent of 11.46 GJth/tonneMeOH thermal

Mass balance of all components entering and exiting the system.
utility and 0.64 GJel/tonneMeOH electrical utility. For total energy
Component In (tonne/h) Out (tonne/h) consumption to produce 1 tonne of methanol in this study,
CO2 265.9 16.3 considering the LHV of CH4 and MeOH is equal to 50 GJ/t and
CO 0 0.3 20.1 GJ/t, respectively, the overall energy required to generate
H2 0 0.1 1 tonne of methanol is equal to 33.45 GJth/tonneMeOH and 0.64 GJel/
H2O 378.7 154.1 tonneMeOH. This is relatively lower than water electrolysis method
MeOH 0 766
of 40.6 GJel/tonneMeOH and 3 GJth/tonneMeOH [30].
CH4 336.9 45.1
AR 0 0 The overall efficiency and CO2 emissions are highly dependent
N2 1.7 1.7 on fuel type and the design of the methanol plant. In a comparative
O2 0.3 0 stand-alone NG-based methanol production plant, numerous
Total 983.6 983.6 studies reported thermal efficiencies range 62–70% [11]. Based
on LHV, the world largest methanol supplier (Methanex Corpora-
tion) conducted an assessment of the energy efficiencies for
various methanol plants and reported an efficiency of 62.4% for
but the power plant is retrofitted with PCC where the product CO2 conventional SMR and 67.1% for ATR [41]. More recent literature
goes for sequestration. In the third scenario, the captured CO2 from have reported technical efficiencies in the range of 68–72% [21]. In
PCC is diverted and utilized in the methanol synthesis process. this study, an efficiency of 68% for a comparable stand-alone NG
plant with a conventional SMR is considered. The utilities process
5.1. Analysis of mass and energy balances fuel for the unit operations in NG-based methanol plant is usually
sourced from NG only [12]. Based on this efficiency, the energy
The capacity basis for this analysis is the ability to digest the input for an equivalent 766 tonne/h methanol at a comparable
entire CO2 stream diverted from the PCC and fulfil the stoichio- stand-alone NG plant would be 22,642 GJ/h (Table 4). Considering
metric number M for methanol synthesis process simultaneously. the LHV, the equivalent methane flow rate to generate 22,642 GJ/h
Table 2 summarizes the mass balance of all components entering is equal to 453 metric tonne/h at STP. This value is corresponding
and exiting the system. In total, 766 tonne/h of methanol is to a ratio of 1.69 tonneMeOH/tonneCH4 , while this study compara-
produced and an amount of 265.9 tonne/h of CO2 is integrated in tively showed a ratio of 2.27 tonneMeOH/tonneCH4 . Table 4 sum-
methanol synthesis process. For this purpose a total of 112.4 m3/h marizes the energy input and output and the overall conversion
of standard liquid methane (336.9 tonneCH4 /h at STP) and efficiencies for this study and a comparable stand-alone NG plant
224.6 tonne/h of water is consumed. Assuming an average of with a conventional SMR. Table 4 also indecates that CO2
8000 operational hours per annum, this will generate a total of utilization decreased the mass flow rate of methane feed from
6128 KTPA of liquid methanol. 453 to 336.9 tonne/h saving 25.6%.
Table 3 shows the energy consumed in the foremost blocks of
the system categorized by their tasks. A total of 2458.3 MWth of 5.2. CO2 emission
thermal duty has been exchanged between cold and hot streams in
three heat exchangers. Apart from heat exchanger duties, Table 3 It is assumed that utility process fuels all come from NG and the
shows that the reformer duty is the highest thermal duty (46.8%) CO2 emission factor is equal to 5.12e05 kg/kJ for both cases
followed by cooling duties (34.3%). The extremely exothermic [42]. By combustion, each mole of CH4 has the potential to be
reactor (REACTOR) is able to evaporate 10 tonne/h of boiling water converted to an equivalent mole of CO2 (Eq. (14)). The molar flow of
and transform it to LP quality steam at 220 8C and 9.8 bar. This CH4 is considered to calculate the CO2 content for both cases. In a
stream (stream 41) can be utilized to compensate 62% of the stand-alone NG plant, methane mass flow rate of 453 tonne/h has
distillation column reboiler duty (RADFRCreb) and improve the an equivalent of 28,237 kmol/h which determines (if combusted)
overall thermal duty by 2.8%. In total, 2438.5 MWth of thermal duty the potential of 1242.7 tonneCO2 /h emissions in the flow. By saving
and 135.5 MWel of electrical duty for pumping and compression 116.1 metric tonne/h of CH4 (25.6%) in this study compared with
the equivalent stand-alone NG-based methanol plant, this will also
be reflected on the CO2 content of comparable CH4 feed. The CO2
Table 3
Energy consumer unit operations categorized by their tasks.
Table 4
Comparative energy balance for this study and an equivalent stand-alone NG-based
Task Units Thermal Electrical methanol plant.
duty duty (MWel)
(MWth)

Cooling COOL1 þ COOL2 þ COOL3 835.2 This study Equivalent NG-based

Heating PREHEAT 208.3 methanol plant using
Compressing CP1 þ CP2 þ CP3 135.2 conventional SMR
Pumping PUMP1 0.3
Mass Net Mass Net
Flush drums DRUM1 þ DRUM2 þ DRUM3 0
flowrate energy flowrate energy
Heat exchange HX1 þ HX2 þ HX3 þ HX4 2458.3a
(tonne/h) (GJ/h) (tonne/h) (GJ/h)
The reformer REFORMER 1140.8
a
The reactor REACTOR 71.5 Methane feed at STP 336.9 16,845 453 22,642
Dist condenser RADFRC_cond 210.2 Thermal duties 8779 7246
Dist rebolier RADFRC_reb 115.5b Electrical duties 488 –c
Methanol productb 766 15,397 766 15,397
Net duty 2438.5 135.5
a Overall efficiency 59% 68%
Hot and cold streams exchange duties with no external duties and CO2
a
emissions. The LHV of methane at STP is equal to 50.035 GJ/t [44].
b b
Integrated with the reactor duty (71.5 MWth is compensated). CO2 emissions The LHV of methanol is equal to 20.094 GJ/t [44].
c
calculated for the extra duty only. The input fuel comes from NG only, no external electrical duty is assumed [12].
 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22 21

Table 5
CO2 fate in different scenarios.

Plant 1a Plant 2b Total CO2 Total CO2

emission abatement
Flue gas Vented CO2 to CO2 content Utilities Vented
sequestration of NG feed

NG-based MeOH plant – – – 1242.7 451.4c NA 1694.1 0

PP 591 – – – – – 591 0
PP þ CCS 295.5 29.6 265.9d – – – 325.1 265.9
PP þ CCU (This study) 295.5 29.6 0 924.2 517.8e 16.3 1783.4 501.7
a
Plant 1 denotes to power generation process and the associated CO2 emission (tonne/h).
b
Plant 2 denotes to MeOH synthesis process and the associated CO2 emission (tonne/h).
c
Based on an equivalent stand-alone NG-based methanol plant designed to generate similar MeOH quantity that produced in this study (766 tonneMeOH/h).
d
Based on 90% CO2 capture rate.
e
Based on Aspen model which uses steam for heating and water for cooling and electricity for pumping and compressing (process fuel is NG only).

content of CH4 feed is reduced from 1242.7 to 924.2 tonneCO2 /h stand-alone NG-based methanol synthesis plant account for
saving 318.5 tonneCO2 /h (Table 5). 2285.1 tonneCO2 /h. This integration can reduce the CO2 emission
The CO2 emissions from utility production and operation based to 1783.4 tonneCO2 /h (21.9% lower) and avoid an equivalent of
on NG combustion is equal to 62.3 kgCO2 /GJ including CO2 501.7 tonneCO2 /h (Table 5). Although the energy demand for this
emissions from process fuel production [12]. This value is in line integration is comparatively higher, the calculated ratio of
with CO2 emission intensity factor that is reported by Methanex 33.45 GJth/tonneMeOH and 0.64 GJel/tonneMeOH is relatively lower
Corporation for their overall methanol production after 2009 than other hydrogen production methods such as water electroly-
[43]. Multiplying 62.3 kgCO2 /GJ by the consumed thermal energy sis. Overall, the analysed process, integrating PCC with SMR and
(7246 GJ/h) will calculate the utilities CO2 emissions for the MeOH synthesis, demonstrates enhanced performance over
comparable stand-alone NG-based methanol plant. Table 5 standalone NG-based MeOH synthesis but also contributes to
compares the CO2 emissions of this study with two other scenarios the utilization of CO2 from power plants (i.e. reducing the
and shows the fate of CO2 with/without process integration geo-sequestration burden).
proposed in this paper. In business as usual scenario (BAU
scenario), the stand-alone NG-based methanol synthesis plant
References
operates separately from the power plant where the combined CO2
emission is equal to 2285.1 tonne/h (scenario A Fig. 8). In the [1] D. Day, R.J. Evans, J.W. Lee, D. Reicosky, Energy Int. J. 30 (14) (2005) 2558–2579.
second scenario, the stand-alone NG-based methanol synthesis [2] O. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A.
plant works as usual but the power plant is retrofitted with PCC Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer,
C.V. Stechow, T. Zwickel, J.C. Minx (Eds.), Climate Change 2014, Mitigation of
and can save 265.9 tonneCO2 /h (11.6% of combined CO2 emissions) Climate Change, Cambridge University Press, Cambridge, United Kingdom and
ready for sequestration process (scenario B Fig. 8). In the last New York, USA, 2014.
scenario, utilizing the CO2 captured from PCC in the methanol plant [3] H. Vidas, B. Hugman, A. Chikkatur, B. Venkatesh, Analysis of the Costs and Benefits
of CO2 Sequestration on the U.S. Outer Continental Shelf. Technical report, 2012.
operating by NG can decrease the CO2 emissions by [4] K.Z. House, C.F. Harvey, M.J. Aziz, D.P. Schrag, Energy Environ. Sci. 2 (2) (2009)
501.7 tonneCO2 /h (21.9% of combined CO2 emissions) from their 193–205.
combined initial CO2 emissions to be recycled in the methanol [5] G.T. Rochelle, et al. Science 325 (5948) (2009) 1652–1654.
[6] S.M. Benson, D.R. Cole, Elements 4 (5) (2008) 325–331.
synthesis process (scenario C Fig. 8). [7] J. Kim, C.A. Henao, T.A. Johnson, D.E. Dedrick, J.E. Miller, E.B. Stechel, C.T. Mar-
avelias, Energy Environ. Sci. 4 (9) (2011) 3122–3132.
6. Conclusions [8] British Petroleum, BP statistical review of world energy, 2014.
[9] D.A. Wood, C. Nwaoha, B.F. Towler, J. Nat. Gas Sci. Eng. 9 (2012) 196–208.
[10] S. Cornot-Gandolphe, O. Appert, R. Dickel, M.F. Chabrelie, A. Rojey, 22nd World
This paper analyses the concept of carbon capture and Gas Conference, 2003, 1–5.
utilization (CCU) by recycling the high-purity CO2 stream coming [11] P. Biedermann, T. Grube, H. Bernd, Methanol as an Energy Carrier, Zentralbi-
from the PCC plant into a detached methanol synthesis plant. An bliothek, Forschungszentrum, 2006.
[12] M.P. Hekkert, F.H.J.F. Hendriks, A.P.C. Faaij, M.L. Neelis, Energy Policy 33 (5) (2005)
existing methanol synthesis plant was proposed to use a 579–594.
conventional SMR and fuelled by NG. Water and methane feeds [13] P. Galindo Cifre, O. Badr, Energy Convers. Manage. 48 (2) (2007) 519–527.
for this plant were specifically optimized to digest the concentrat- [14] H.J. Herzog, Environ. Sci. Technol. 35 (7) (2001) 148A–153A.
[15] Unites States Environmental Protection Agency. Geologic CO2 Sequestration
ed CO2 stream (265.9 tonne/h) with the syngas product from SMR. Technology and Cost Analysis, 2010.
The mixture enters the methanol synthesis reactor where [16] C. Hanisch, Environ. Sci. Technol. 32 (1) (1998) 20A–24A.
methanol is formed. A flowsheet model was developed and [17] J.E. Miller, R.B. Diver, N.P. Siegel, E.N. Coker, A. Ambrosini, D.E. Dedrick, M.D.
Allendorf, A.H. McDaniel, G.L. Kellogg, R.E. Hogan, et al. Metals & Materials
validated for this entire process integration, before it was used to Society (TMS) Annual Meeting & Exhibition: Session I, 2010.
analysis the performance (mass, energy and CO2 emissions) of the [18] G. Zahedi, A. Elkamel, A. Lohi, Energy Fuels 21 (5) (2007) 2977–2983.
proposed integration. A total of 766 tonneMeOH/h is produced, [19] S. Abrol, C.M. Hilton, Comput. Chem. Eng. 40 (2012) 117–131.
[20] P.J. Dahl, T.S. Christensen, S.M. King, S. Winter-Madsen, Nitrogen þ Syngas,
equivalent to 2.27 tonneMeOH/tonneCH4 . This ratio is higher than
International Conference and Exhibition, 2014.
1.69 tonneMeOH/tonneCH4 calculated for the equivalent stand-alone [21] O. Jörg, V. Gronemann, F. Pontzen, E. Fiedler, G. Grossmann, D. Burkhard Kerse-
NG-based methanol synthesis plant with a typical 68% efficiency. bohm, G. Weiss, C. Witte, Methanol, in: Ullmann’s Encyclopedia of Industrial
Chemistry, 2012.
This high ratio is caused by CO2 stream integration which resulted
[22] P.L. Spath, D.C. Dayton, Preliminary screening – technical and economic assess-
in methane uptake reduction by 25.6% which saved up to ment of synthesis gas to fuels and chemicals with emphasis on the potential for
116.1 metric tonne/h of CH4 at STP conditions. The reduction of biomass-derived syngas, Technical report, DTIC Document, 2003.
NG consumption resulted in less CO2 emission in addition to [23] R. Khalilpour, A. Abbas, Int. J. Greenh. Gas Control 5 (2) (2011) 189–199.
[24] A.V. Johansson, A.A. Brandberg, A. Roth, The Life of Fuels: Motor Fuels from Source
the recycled CO2 stream. In BAU case, the combined CO2 emission to End Use, Ecotraffic AB, Stockholm, 1992.
of a 660 MWel coal-fired power plant (PP) and a comparable [25] R.F. Aguilera, Energy 35 (8) (2010) 3332–3339.
22 D. Milani et al. / Journal of CO2 Utilization 10 (2015) 12–22

[26] D. Möst, H. Perlwitz, Energy 34 (10) (2009) 1510–1522. [37] Y. Choi, H.G. Stenger, J. Power Sources 124 (2) (2003) 432–439.
[27] K. Liu, C. Song, V. Subramani, Hydrogen and Syngas Production and Purification [38] P.J. Robinson, Dynamic Modeling and Plantwide Control of a Hybrid Power and
Technologies, Wiley, New York, 2010. Chemical Plant: An Integrated Gasification Combined Cycle Coupled with a
[28] E.C. Vagia, A.A. Lemonidou, Int. J. Hydrogen Energy 32 (2) (2007) 212–223. Methanol Plant, 2011 (Ph.D. thesis).
[29] D.L. Trimm, Catal. Today 37 (3) (1997) 233–238. [39] J. Xu, G.F. Froment, AIChE J. 35 (1) (1989) 88–96.
[30] É.S. Van-Dal, C. Bouallou, J. Clean. Prod. 57 (2013) 38–45. [40] N. Rezaie, A. Jahanmiri, B. Moghtaderi, M.R. Rahimpour, Chem. Eng. Process.:
[31] K.M. Vanden Bussche, G.F. Froment, J. Catal. 161 (1) (1996) 1–10. Process Intensif. 44 (8) (2005) 911–921.
[32] D. Mignard, C. Pritchard, Chem. Eng. Res. Des. 86 (5) (2008) 473–487. [41] M. Allard, CO2 Emissions from Methanol Production. Technical Report, American
[33] ÉS. Van-Dal, C. Bouallou, Chem. Eng. 29 (2012). Methanol Institute, Washington, DC, 1999.
[34] T.H. Pedersen, R.H. Schultz, S.K. Kær, P.F.J. Hoffmann, Technical and economic [42] Climate Change Science Research Department of Industry, Innovation and Ter-
assessment of methanol production from biogas, LAP LAMBERT Academic tiary Education, Referance Guide, 2013.
Publishing, 2014. [43] Methanex Corporation. Responsible care and social responsibility report,
[35] M. Sahibzada, I.S. Metcalfe, D. Chadwick, J. Catal. 174 (2) (1998) 111–118. 2012.
[36] K. Klier, V. Chatikavanij, R.G. Herman, G.W. Simmons, J. Catal. 74 (2) (1982) 343– [44] R.G. Boundy, S.W. Diegel, L.L. Wright, S.C. Davis, Biomass Energy Data Book,
360. 2011.