Energy 53 (2013) 106e113

Contents lists available at SciVerse ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Cryogenic method for H2 and CH4 recovery from a rich CO2 stream
in pre-combustion carbon capture and storage schemes
K. Atsonios a, b, K.D. Panopoulos b, *, A. Doukelis a, A. Koumanakos a, E. Kakaras a, b
a
Laboratory of Steam Boilers and Thermal Plants, National Technical University of Athens, Heroon Polytechniou 9, Athens 15780, Greece
b
Chemical Process and Energy Resources Institute, Centre for Research and Technology Hellas 6th Klm. Charilaou e Thermi Road, 570 01 Thermi,
Thessaloniki, Greece

a r t i c l e i n f o a b s t r a c t

Article history: The pre-combustion carbon capture technology based on coal gasification or methane reforming fol-
Received 6 September 2012 lowed by (water gas shift reactors assisted with Pd-alloys membranes) WGS-MR is considered very
Received in revised form promising for the production of a rich hydrogen stream that can be combusted in combined cycles.
13 February 2013
However, recovery of the total H2 content is not feasible and a part of it remains in the retentate side. The
Accepted 14 February 2013
Available online 26 March 2013
requirement for upstream high pressure operation of the necessary reforming step has a drawback:
complete reforming of CH4 is not possible; thus small but significant amounts remain in the rich CO2
stream. These CH4 amounts not only affect the efficiency of the process but also are against regulations
Keywords:
Hydrogen production
for the allowed composition of carbon dioxide for storage. Therefore, an efficient purification step before
ATR its compression is of high importance. The current work models a cryogenic method for the separation of
WGS membrane reactor combustibles from a rich-CO2 stream and evaluates its effects on the efficiency of the pre-combustion
CO2 purification carbon capture system. The modeling study is performed with AspenPlusÔ. A study on the effect of
Cryogenic separation operating parameters of the (Purification & Compression Unit) PCU integration on the performance is
ASPEN Plus presented.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction In all carbon capture schemes, it is of paramount importance

that the internal energetic consumption of the CO2 compression
Pre-combustion capture is one of the proposed options for train is minimized [7e12] by purifying the rich-CO2 stream. Huang
carbon dioxide removal for which great interest has been shown et al. [10] and Posch et al. [12] adopted a cryogenic method for CO2
lately [1,2]. Having sequestrated carbon in the form of CO2 before purification for oxy fuel capture schemes, which was proven to be
fuel combustion at high pressures, enables utilization of the energetically and costly inefficient because of high energy con-
resulting fuel gas in various applications, such as the production of sumptions due to cooling loads, while the recovered gas could not
power [3], pure hydrogen [4] or other chemicals [5]. be further utilized and was directed to the stack. It was also shown
In CO2 pre-combustion capture schemes based on (Combined that applying a distillation column rather than flash separators
Cycles using Natural Gas) NGCC or (Integrated with Coal Gasifica- provided better CO2 stream purification accompanied by higher
tion) IGCC the main strategy is to reform or gasify the original fuel energy penalty.
towards a reformate fuel [1]. The standard proposed method for A schematic of such an operation is shown in Fig. 1: the
carbon reduction is then to perform water gas shift, transform CO reformate gas stream is fed to the reactor and CO is shifted with
into CO2 and use amine absorption for its capture [6]. An alternative simultaneous H2 recovery on the permeate side through Pd-alloys
to that is to employ H2 permeable reactors assisting the shift re- membranes. The permeation mechanism of Pd-alloys is through
action: this option has advantages, such as high hydrogen recovery the disassociation of H2 on the membrane surface and trans-
by pushing the shift reaction and H2 purity. portation through the metal structure as atomic hydrogen [13,14].
Pd membranes offer combined advantages of high H2 recovery
rates and high selectivities [15]. The techniques need to be opti-
* Corresponding author. ARKAT building, 357-359 Mesogeion Ave., Halandri, mized for membrane capital cost reduction and avoiding mem-
Athens, Greece. Tel.: þ30 210 6501507; fax: þ30 210 6527539.
brane replacement costs resulting from potential poisoning at
E-mail addresses: [email protected] (K. Atsonios), [email protected],
[email protected] (K.D. Panopoulos), [email protected] (A. Doukelis), low temperatures [16,17]. Furthermore compensates need to be
[email protected] (A. Koumanakos), [email protected] (E. Kakaras). found for the fact that hydrogen recovery is not perfect and

0360-5442/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.energy.2013.02.026
 K. Atsonios et al. / Energy 53 (2013) 106e113 107

Purification & Compression Unit

H O removal oxy-burner
cooler
CO Retentate

compressor expander

O
Natural Gas
Autothermal Water Gas Water Gas Shift
Reforming Reactor Shift Reactor Membrane Reactor

N
O
Air Separation Air
Steam Unit
Fuel
Flue Gases Gas Turbine
Air
Combined Cycle

Fig. 1. Process flowsheet diagram of the total power system.

considerable amount remain at the retentate side (from 2 to 10% and purification is performed in a water gas shift e membrane
of the total hydrogen [18]). Besides hydrogen, other combustibles reactor where it is assumed that the (hydrogen recovery factor)
are present, such as residual CH4 that was not reformed as well as HRF is equal to 98% and the operational temperature 400  C (base
CO traces that have not been shifted. Since the retentate stream case). The nitrogen stream that is produced from the ASU is utilized
contains the whole amount of CO2 that is led to storage, it is as a sweep gas, increasing the hydrogen recovery driving force in
essential to increase CO2 purity in it by removing the other spe- the membranes. The H2-fuel mixture is fed to the power plant is-
cies like H2O, H2, CH4, CO, etc. On the one hand, water removal is land which consists of a gas turbine combined with a (heat re-
feasible by cooling the stream under the dew point so that the covery steam generator) HRSG. A more detailed description of the
water is condensed. On the other hand, special attention is paid process can be found in a previous study [20].
to the other species. The conventional suggestion to eliminate The base case for the CO2 purification and compression option
these compounds is to combust them [19], but this has to be is also shown in Fig. 1: The retentate stream after the membranes
accomplished with pure oxygen as the oxidant agent. Technically is expanded and catalytically combusted in an oxy-fired
this would require the use of catalysts to achieve high combus- combustor. Chemical equilibrium predicts that complete com-
tion rates with such a lean fuel and stoichiometric ratio very bustion of the remaining combustible species is feasible with
close to 1. almost no oxygen surplus [20] but practically this is not easily
The present study suggests an alternative choice to handling the achievable. The required oxygen depends on how the upstream
combustibles by recovering them with cryogenic separation tech- units (ATR and WGS performance) are operated: for example
niques. Based on differences in thermodynamic properties as far as assuming for the base case a TATR ¼ 1050  C and S/CATR ¼ 1.5 the
the dew point of each component is concerned, the retentate required amount of oxygen for the post combustor increases the
stream is cooled down and the CO2 is separated in flash separators total oxygen production in the ASU by 17.5%. The hot flue gases
or a distillation column. A parametric investigation for the best deliver heat in a secondary heat recovery steam generator where
operation of the block is performed and a comparison is made with steam cycle feedwater becomes high-pressure superheated
the conventional option for purification with oxy combustion. steam. Consequently, the water content of the flue gases is
removed in a flash separator and the almost pure CO2 is then
2. Plant description compressed and pumped. The operating parameters of the ATR
and WGS-MR play an important role in the amount of heat in the
The outline of the system is presented in Fig. 1. A H2 rich fuel is retentate stream.
produced from natural gas reforming in an (Autothermal Reactor) The present study is focused in the CO2 stream purification and
ATR, and CO is further shifted in a (high temperature water gas compression block based on two separation methods, (a) flash
shift reactor) HT-WGS. The autothermal conditions are met using a separator and (b) distillation column (see Fig. 2). Apart from CO2,
rich oxygen stream (95% purity) which is produced in an (air the retentate stream mainly consists of H2O, H2, CH4, N2 and Ar (see
separation unit) ASU. The maximization of hydrogen production Table 1).

Fig. 2. Schematic of the suggested purification and compression unit with cryogenic fuel recovery with flash separators.
108 K. Atsonios et al. / Energy 53 (2013) 106e113

Table 1
Retentate stream T, P and composition.

Stream T ( C) p (bar) M (kg/s) Molar fraction (%)

CO2 CO H2 CH4 N2 Ar H2O

Retentate 400.0 46.0 67.82 52.92 0.03 3.14 1.92 0.72 0.89 40.40

This gas stream is expanded and then cooled down to around other hand, if the outlet pressure of the rich CO2 stream is low, the
220  C. Some of the heat is recovered for generating superheated energy duty for compression is considerable.
intermediate pressure steam at 315  C. Expanding the gas has a
triple positive effect: firstly, the mixture is separated more easily at 2.2. Scheme 2: separation by distillation column e external &
lower pressures. Secondly, the manufacturing cost of units such as internal cooling
the evaporator is significantly lower if they operate at lower pres-
sures. Thirdly, as the water content in the retentate stream is about The concept of combustibles recovery with distillation mainly
25% w/w, the CO 2 rich mass flow rate that is compressed is less consists of a distillation column, a heat exchanger, a flash separator
than the corresponding stream that is expanded, contributing and an external refrigeration system (Fig. 3). After the heat
positively to the total power balance. Consequently, cooling water exchanger, the inlet stream is partly condensed and is separated at
is employed to bring the stream into water condensation condi- the flash separator. The liquid stream is further cooled by means of
tions at 28  C. an expansion (JouleeThomson effect), while the gas stream is
The next part of the purification and compression units differs cooled through expansion. Both streams enter the distillation col-
for the two proposed schemes: umn to make combustibles separation more effective. The cooling
loads for the column’s condenser are obtained from an external
2.1. Scheme 1: double flash separation e internal cooling refrigerant and are dependent on the desirable rate of carbon
capture efficiency. The corresponding heat at the reboiler can be
This system is auto-refrigerated with no additional cooling obtained by the refrigerant inter-cooling while it is compressed.
system required (Fig. 2). Flash separations are performed at two The distillation column outlet streams can be used to assist cooling
different temperatures, and at the same pressure level. Before each the retentate stream after water removal, thus reducing the energy
flash, there is a heat exchanger that cools the inlet stream. The consumptions.
required cooling loads are taken from the final steams, as can be The refrigeration system comprises a two stage compressor in
seen in Fig. 2. The rich-CO2 liquid steams are throttled adiabatically order to provide cooling loads at two different temperatures:
and their temperature is reduced (JouleeThomson effect). The level at 20  C for the retentate stream cooling and at around 65  C
of throttling has been set so as to permit heat transfer at the two (depending on recovery rate that determines the dew point of the
heat exchangers without temperature crossovers, assuming a outlet gas stream) to fulfill the condenser duties at the distillation
minimum temperature approach DT ¼ 3  C. The final streams come column. The most suitable cooling medium that is employed in the
out of the Purification Unit at the temperature of 18  C and the rich- modeling is R1150 (ethylene e C2H4). The temperature approach of
CO2 stream enters the Compression Unit, where it is compressed in all heat exchangers is assumed 3  C, the pressure levels of the
a three stage inter-cooled compressor up to 80 bar at supercritical evaporation are 2.26 and 23.47 bar and that for condensing 27 bar.
conditions. Next, it is cooled, liquefied and pumped up to 28  C/ Finally, the compressor’s polytropic efficiency is assumed equal to
110 bar and is transported to storage. 0.82. Since the cooling medium cannot reject its heat directly to the
Since the temperature at the 2nd Flash is 54.5  C (near the ambient while it is condensed, a secondary auxiliary cooling cycle is
triple point of CO2), the parameters that determine the system’s required. The cooling medium in this refrigerant cycle is the com-
efficiency are the expander pressure outlet and the temperature of mercial R134a and the COP of the cycle is 3.59.
the 1st flash separation. The thermochemical properties of the PCU block are calculated
The expander outlet pressure determines the CO2 capture rate: according to the PengeRobinson equation of state [12]. The tech-
lower outlet pressures result in better CO2 recovery rates. On the nical data of the system are presented in Table 2.

Fig. 3. Schematic of the suggested purification and compression unit with cryogenic fuel recovery with distillation column.
 K. Atsonios et al. / Energy 53 (2013) 106e113 109

Table 2 losses correspond to irreversibilities from various processes like

Process model specifications. heat exchanger, valves, expanders, separators etc.
Compressors polytropic efficiency, hp (%) 82.0 According to Fig. 4a, the exergetic efficiency of the Unit can be
Expander isentropic efficiency, his (%) 80.0 defined as:
Mechanical efficiency, hm (%) 99.6
Pressure drop in the column (bar) 0.50 Ex;rec þ EQ ;useful þ PPCU
Heat exchangers minimum temperature approach ( C) 3.0 hex;PCU ¼ (1)
Column stages, N 5 Ex;in

where, PPCU > 0 when the Unit operation yields power and PPCU < 0
3. Results and discussion when the Unit consumes power. The ability of combustibles’ re-
covery and re-usage can also be quantified by the following factor:
3.1. Process parameters investigation
LHVrec ½kW
hrec ¼ (2)
LHVin ½kW
In order to come up with concluding remarks on the efficiency
of the proposed systems, a comparison of the exergetic efficiencies
which stands for the fraction of combustibles heat input in terms of
of the schemes is performed. The exergy balance of the PCU block is
LHV entering the PCU that is recovered. The corresponding values of
shown in Fig. 4a. The exergy input comes from the retentate stream,
these parameters for each PCU case are presented in Table 3. Pro-
while any power or heat duties are considered as part of the exergy
vided that the initial PCU scheme is technically feasible, CO2 is
outputs.
completely captured, compressed and stored while electrical power
The exergy dissipation for the cases under investigation is
is generated.
depicted by Grassmann diagrams which are displayed in Fig. 4bed.
The most important streams of the process are shown in Table 4.
Arrows that are at upper side of the main exergy arrow correspond
The second method of separation is more efficient as far as hrec is
to exergy that is not lost (power production or heat recovery at the
concerned, resulting in higher purity CO2-rich stream. What is
steam cycle) whereas the remaining arrows refer to exergy losses
more, in this case, the CO2 content in the recovered gas stream is
(irreversibilities). As far as the process of the oxy-combustion option
lower.
is concerned, heat recovery and power generation are exergy that is
It should be mentioned that high recovery rates are not neces-
not lost. The retentate stream at the exit of the burner is at high
sarily linked to high efficiency of the total plant. The way that the
temperature and a large fraction of the exergy input is utilized for
recovered combustibles are exploited plays a significant role on the
high enthalpy superheated steam generation. The ASU consumption
choice of the most suitable technique. To this end, the thermody-
for extra oxygen production is not taken into account in the exergy
namic comparison of the purification options under investigation is
balance and its effect on total efficiency is investigated below. On the
completed by the process integration and the investigation of their
other hand, both separation options can recover almost half of the
effect on total plant operation.
total exergy by recovering the combustibles content. The corre-
sponding gains from heat recovery are lower in the combustibles
separation cases, as the temperature inlet at the heat exchanger is 3.2. Effect of purification methods on total plant
400  C instead of 708  C in the oxy-combustion case. Consequently,
the steam quality differs from case to case, as well as the corre- Unlike other cryogenic CCS applications, due to the fact that the
sponding exergy utilization. The final CO2 stream in the oxy- recovered gas stream has a considerable amount of chemical en-
combustion case has larger exergy fraction than the separation ergy, it can be fed back to the system for increased energy effi-
options, due to increased CO2 content from CH4 combustion. Other ciency. The proposed alternatives for its utilization are either to

a b

d
Fig. 4. a) Exergy balance of the PCU and Grassmann diagram of PCU with b) expander and post combustor, c) flash separation and d) distillation column.
110 K. Atsonios et al. / Energy 53 (2013) 106e113

Table 3 Table 5
Comparison of the three CO2 purification and compression options. Base case results for the three purification methods (CCR ¼ 90%).

Post Flash Distillation Separation method Flash Distillation Oxy

combustor separation column separator column combustor
Power consumption þ87.65 þ2.46 92.13 Specific H2 production 0.328 0.328 0.291
(kJPCU =kgCO2 ) (kgH2 =kgNG )
hex,PCU, (%) 27.5 31.9 42.45 Specific O2 production 0.89 0.89 0.99
hrec (%) e 62.6 99.2 (kgO2 =kgNG )
Max. CO2 capture 100.0 98.0 99.0 GT power production 39.64 39.99 38.06
efficiency (CCE) (%) (%kWGT/LHV)
ST power production 15.53 15.72 17.75
(%kWST/LHV)
Specific PCU power þ0.01 0.49 þ0.45
reform or to burn it. Although returning them to the ATR would (%kWPCU/LHV)
decrease fuel consumption, the inert compounds that are contained Specific ASU power 5.37 5.37 5.94
in the stream (mainly N2 and Ar) would accumulate in the reactor. (% kWPCU/LHV)
Net efficiency (%) 50.74 50.77 51.36
Given that a method to remove them is not available, this option is
Specific CO2 emissions 42.4 42.4 42.1
abandoned. The alternative choice suggests the injection to the GT (gCO2 =kWhel )
combustor. In this case, the carbon capture rate of the system is less SPECCA (kJLHV =gCO2 ) 2.95 2.94 2.67
than the capture rate at the PCU due to the produced CO2 from the
combusted CH4 in the GT. Given that the system under investiga-
tion is aimed to the maximization of the final H2 rich fuel and not 10% in these cases, implying a smaller ASU. However, the more
the maximization of electrical power production, a different complex purification system compensates this feature. The high
approach on recovered fuel should be adopted. heat recovery rate in the oxy-combustion case results in the
The basis of comparison of each method from a performance increased power production in the ST (c. 2% increase).
perspective can be the parameter named SPECCA (Specific Energy
Consumption for CO2 Avoided). It expresses the additional fuel 3.2.1. Effect of (carbon capture rate) CCR
energy in MJ that is required to avoid 1 kg of CO2: It should be mentioned that (carbon capture efficiency)
CCE does not coincide with the (carbon capture rate) CCR of the
total plant because it is independent of the final usage of the
 
1 1 recovered gases. In this study, this stream is fed to the GT to be
3600$ 
h hREF   combusted. As a result, the CCR is also dependent on the recovered
SPECCA ¼ ; kJLHV =gCO2 (3)
EREF  E CH4 and CO.
Fig. 5 provides useful information on the effect of Carbon Cap-
where E is the CO2 emission rate, in kgCO2/kWhel, and h the net ture Rate on plant performance. For low CCE, the flash separation
electrical efficiency of the plants. The (reference plant) REF refers to method has a positive effect on energy balance as the PCU produces
the corresponding NGCC plant without Carbon Capture. The net instead of consuming electrical power (Fig. 5a). This is owed to the
efficiency of the reference plant is hREF ¼ 58.3% with specific CO2 expansion of the CO2-rich stream. However, the combustibles are
emission rate EREF ¼ 354:3 grCO2 =kWh. not recovered as effectively as in the case of the distillation column
Among the most important parameters for total plant specifi- method and therefore the latter is exergetically more efficient. As
cation are the operating parameters of the ATR and the WGS-MR. the CCE approaches 100%, energy duties for both methods tend to
Table 5 summarizes the characteristics of the base case model be equal and the divergence of the corresponding exergy efficiency
for the three purification methods under investigation. It is clear increases. The effect of CCR on the overall plant efficiency and the
that oxy combustion of the retained combustibles (namely CH4 that corresponding SPECCA (Fig. 5b) is the same for both combustibles
is not reformed, CO that is not shifted and H2 not recovered at the separation options (especially for CCR < 95%). According to the
membranes) is the most efficient method for the CO2 rich gas process specifications adopted (more specifically the methane rate
treatment in terms of energy efficiency. However, the specific that is assumed not having been reformed) the maximum CCR for
quantity of the produced H2 that enters the GT combustor is the distillation column and flash separator options is 96% and 97%
increased by 11% in the cases of cryogenic separation. In other respectively, for maximum CH4 recovery. For lower CH4 recovery
words, for systems dedicated to H2 production, cryogenic separa- rates, a large part of methane would remain at the CO2 stream and
tion methods are considered to be more efficient in terms of H2 the total efficiency would drop considerably. If a higher CCR is
production yield. Additionally, the oxygen demand is reduced by preferable, the process specifications at the ATR which are related

Table 4
Outlet streams from PCU for the three purification methods.

Separation method Flash separator Distillation column Post combustor

Stream Rich CO2 Recovered combustibles Rich CO2 Recovered combustibles Rich CO2
T ( C) 18.0 29.2 18.0 16.7 18.0
p (bar) 110.0 25.0 110.0 25.0 110.0
M (kg/s) 49.96 2.21 49.66 2.52 46.70
% Mole fraction CO2 96.28 18.65 99.77 7.74 96.59
CO 0.01 0.23 0.00 0.26 0.00
H2 0.62 47.95 0.00 43.50 0.00
CH4 1.80 16.20 0.04 26.25 0.00
N2 0.40 8.68 0.00 10.03 1.38
Ar 0.75 8.29 0.01 12.22 1.89
H2O 0.14 0.00 0.18 0.00 0.13
 K. Atsonios et al. / Energy 53 (2013) 106e113 111

Fig. 5. (a). Impact of CCE on PCU performance for the two purification methods. (b). Effect of CCR on the power plant for the two purification methods.

Fig. 6. (a). Impact of S/CATR on PCU performance for the two purification methods (CCE ¼ 90%). (b). Effect of S/CATR on the power plant for the two purification methods.

Fig. 7. (a). Impact of HRF on PCU performance for the two purification methods (CCE ¼ 90%). (b). Effect of HRF on the power plant for the two purification methods.
112 K. Atsonios et al. / Energy 53 (2013) 106e113

Fig. 8. (a). Impact of TATR on PCU performance for the two purification methods (CCE ¼ 90%). (b). Effect of TATR on the power plant for the two purification methods (CCR ¼ 90%).

to methane conversion should be modified. On the other hand, for autothermal conditions in the reactor, more oxidant agent is
the oxy combustion case, since the CCR is correlated with the required in the case that the temperature is high. Furthermore,
fraction of NG that is by-passed and not reformed, efficiency drops materials stability limits the maximum operating temperature to
linearly as CCR increases. around 1100  C. As is shown in Fig. 8a, at low ATR temperatures,
The parametric investigations that follow have a CCR ¼ 90% for cryogenic systems consume more energy for CO2 purification due
all cases. to increased presence of the methane at the retentate stream. Re-
covery and exergy efficiencies for separation with distillation col-
3.2.2. Effect of steam to carbon ratio of the ATR (S/CATR) umn are greater than those of the corresponding flash separation
The amount of steam that is injected in the ATR plays a signif- method. This feature has a positive effect on the total system when
icant role on methane reforming. The higher hydrocarbons in the the combustion of the recovered fuel satisfies the required CO2
(natural gas) NG are easily reformed at the pre-reforming stage at capture rate. In this case, for TATR ¼ 950  C, part of the methane is
lower temperatures and hence, a reforming efficiency equal to 100% selected not to be recovered in order to meet the goal of CCR ¼ 90%.
is assumed. High S/CATR ratios benefit CH4 reforming and lower the Consequently, the net efficiency drops considerably (Fig. 8b). At
energy consumption for its recovery at the PCU. high temperatures, the oxy-combustion option is more efficient
On the other hand, electrical power derived from steam turbines than the cryogenic options by ca. 1%.
is decreased when S/CATR increases, as more of the produced steam
is extracted instead of being expanded. According to Fig. 6b, the
4. Conclusions
optimum value for plant efficiency is 1.5 for all cases. For high S/CATR
ratios, the flash separation method is slightly better as the exergy
This study investigates the cryogenic method as an alternative
efficiencies are almost the same and the power gain from this
choice for rich-CO2 purification after membrane separation, instead
method is quite high. What is more, for low S/CATR ratios, where
of simply combusting the retained combustibles. Two proposed
steam extraction is less, an efficient recovery of the unreformed
cryogenic systems are investigated: flash separation with internal
CH4 is rather preferable and approaches the oxy-combustion case.
cooling and separation with distillation column. In the first case,
electrical power is produced while the separation efficiency is quite
3.2.3. Effect of (hydrogen recovery factor) HRF of the WGS-MR
high, as 62.6% of the combustibles heat input is recovered. On the
This parameter refers to the ability of the reactor to make
other hand, separation by a distillation column may result in the
hydrogen available at the opposite side (permeate side), where the
complete separation of combustibles providing high purity in the
H2-rich fuel is produced. Recent developments for Pd-alloys
final CO2 stream (>99%). However, more energy duty is required
membranes combined with WGS catalysts showed that hydrogen
due to the external cooling system.
recovery is achievable at rates greater than 90% [21] and in some
The sensitivity analysis showed that the cryogenic methods can
cases even close to 100% [18]. The hydrogen recovery factor
overbalance any “weak” operating mode of the hydrogen block, such
strongly affects membranes cost, as the higher the hydrogen re-
us low methane conversion rates at the ATR and low hydrogen re-
covery the larger membrane area required [22].
covery rates at the membranes. Nevertheless, as far as the total
Fig. 7b shows that cryogenic methods may be beneficial for
system efficiency is concerned, the oxy-combustion option is pref-
membranes with low HRF (>90%), as the variation of this parameter
erable as it can combine both high capture rates and performance.
does not affect the total efficiency, unlike the oxy combustion op-
Future work that correlates membrane area and investment cost
tion. Membranes with high HRF favor the application of an oxy-
of the whole plant will finally determine under which conditions a
combustor instead of combustibles recovery.
cryogenic recovery system is required.
3.2.4. Effect of the ATR operating temperature (TATR)
The ATR operating temperature is of high importance, as it plays Acknowledgments
a significant role on syngas quality: according to the chemical
equilibrium, higher temperatures beneficiate increased methane The authors would like to gratefully acknowledge the support of
conversion rates. On the other hand, in order to secure the the European Commission (CACHET II, FP7 Project No. 241342).
 K. Atsonios et al. / Energy 53 (2013) 106e113 113

Abbreviations [7] Romeo LM, Lara Y, González A. Reducing energy penalties in carbon capture
with Organic Rankine Cycles. Appl Therm Eng 2011;31:2928e35.
[8] Zanganeh KE, Shafeen A, Salvador C, Beigzadeh A, Abbassi M. CO2 processing
and multi-pollutant control for oxy-fuel combustion systems using an
ASU air separation unit advanced CO2 capture and compression unit (CO2CCU. Energ Procedia 2011;4:
ATR autothermal reformer reactor 1018e25.
[9] Kakaras E, Koumanakos A, Doukelis A, Giannakopoulos D, Vorrias I. Oxyfuel
CCE carbon capture efficiency boiler design in a lignite-fired power plant. Fuel 2007;86:2144e50.
CCR carbon capture rate [10] Huang Y, Wang M, Stephenson P, Rezvania S, McIlveen-Wright D, Minchener A,
CCS carbon capture and storage et al. Hybrid coal-fired power plants with CO2 capture: a technical and eco-
nomic evaluation based on computational simulations. Fuel 2012;101:244e53.
COP coefficient of performance [11] Romeo LM, Bolea I, Lara Y, Escosa JM. Optimization of intercooling compres-
GT gas turbine sion in CO2 capture systems. Appl Therm Eng 2009;29:1744e51.
HRSG heat recovery steam generator [12] Posch S, Haider M. Optimization of CO2 compression and purification units
(CO2CPU) for CCS power plants. Fuel 2012;101:254e63.
HRF hydrogen recovery factor [13] Scholes CA, Smith KH, Kentish SE, Stevens GW. CO2 capture from pre-
HT-WGS high temperature water gas shift reactor combustion processes e strategies for membrane gas separation. Int J
LHV lower heating value Greenh Gas Con 2010;4:739e55.
[14] Marano JJ, Ciferino JP. Integration of gas separation membranes with IGCC e
NG natural gas identifying the right membrane for the right job. Energy Procedia 2009;1:
PCU purification & compression unit 361e8.
ST steam turbine(s) [15] Ryi S-K, Lee C-B, Lee S-W, Park J-S. Pd-based composite membrane and its
high-pressure module for pre-combustion CO2 capture. Energy, in press.
S/CATR steam-to-carbon ratio in ATR
[16] Peters TA, Kaleta T, Stange M, Bredesen R. Hydrogen transport through a
WGS-MR water gas shift membrane reactor selection of thin Pd-alloy membranes: membrane stability, H2S inhibition, and
flux recovery in hydrogen and simulated WGS mixtures. Catal Today
2012;193:8e19.
References [17] Peters TA, Kaleta T, Stange M, Bredesen R. Development of ternary PdeAgeTM
alloy membranes with improved sulphur tolerance. J Membr Sci 2013;429:
[1] Olajire A. CO2 capture and separation technologies for end-of-pipe applica- 448e58.
tions e a review. Energy 2010;35:2610e28. [18] Mendes D, Chibante V, Zheng JM, Tosti S, Borgognoni F, Mendes A, et al.
[2] Kanniche M, Gros-Bonnivard R, Jaud P, Valle-Marcos J, Amann JM, Bouallou H. Enhancing the production of hydrogen via water gas shift reaction using Pd-
Pre-combustion, post-combustion and oxy-combustion in thermal power based membrane reactors. J Membr Sci 2010;35:12596e608.
plant for CO2 capture. Appl Therm Eng 2010;30:53e62. [19] Dijkstra JW, Jansen D. Novel concepts for CO2 capture. Energy 2004;29:
[3] Lund H, Mathiesen BV. The role of carbon capture and storage in a future 1249e57.
sustainable energy system. Energy 2012;44(1):469e76. [20] Atsonios K, Panopoulos K, Doukelis A, Kakaras E. Exergy analysis of a
[4] Liszka M, Malik T, Manfrida G. Energy and exergy analysis of hydrogen- hydrogen fired combined cycle with natural gas reforming and membrane
oriented coal gasification with CO2 capture. Energy 2012;45:142e50. assisted shift reactors for CO2 capture. Energ Convers Manage 2012;60:
[5] Clausen LR, Elmegaard B, Houbak N. Technoeconomic analysis of a low CO2 196e203.
emission dimethyl ether (DME) plant based on gasification of torrefied [21] Dijkstra JW, Pieterse JAZ, Li H, Boon J, van Delft YC, Raju G, et al. Development
biomass. Energy 2010;35:4831e42. of membrane reactor technology for power production with pre-combustion
[6] Skorek-Osikowska A, Janusz-Szymanska K, Kotowicz J. Modeling and analysis CO2 capture. Energy Procedia 2001;4:715e22.
of selected carbon dioxide capture methods in IGCC systems. Energy 2012;45: [22] Criscuoli A, Basile A, Drioli E, Loiacono O. An economic feasibility study for
92e100. water gas shift membrane reactor. J Membr Sci 2001;181:21e7.