Energy Conversion and Management 110 (2016) 1–12

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Techno-economic analysis of four concepts for thermal decomposition of

methane: Reduction of CO2 emissions in natural gas combustion
Tiina Keipi a,⇑, Ville Hankalin b, Jaakko Nummelin b, Risto Raiko a
a
Department of Chemistry and Bioengineering, Tampere University of Technology, P.O. Box 541, 33101 Tampere, Finland
b
ÅF-Consult, Bertel Jungin aukio 9, 02600 Espoo, Finland

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

Article history: This paper presents a techno-economic analysis of four concepts that apply the thermal decomposition of
Received 7 September 2015 methane (TDM) with the aim of reducing carbon dioxide emissions in natural gas combustion. Different
Accepted 21 November 2015 technical solutions are applied to convert methane in natural gas to gaseous hydrogen, which is com-
busted to produce electricity with a steam power cycle, and solid carbon, which is assumed to be sold
as carbon black. The cost of electricity production and the potential to reduce CO2 emissions in each con-
Keywords: cept were evaluated and compared to the reference case of direct methane combustion. With a moderate
Carbon capture
emission allowance price ð20 €=tCO2 Þ and product carbon price ð500 €=tcarbon Þ the cost of electricity pro-
Natural gas
Carbon black
duction in the concepts was 12–58% higher than in the reference case. However, the price of product car-
Methane decomposition bon had a significant effect on the feasibility of the concepts. Thus, the methane burner, which showed
Techno-economic analysis the best performance, produced 17% less CO2 emissions per MWhe and had a smaller cost of electricity
Carbon dioxide production than the reference case already with the carbon price of 600—700 €=tcarbon .
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction applied. Li et al. [7] have further presented that increasing pressure
has a negative effect on the TDM reaction. Catalysts can be applied
According to the estimates given by the International Energy to reduce the reaction temperature. This is a process called cat-
Agency, global gas demand will grow from 3:4 trillion m3 in 2012 alytic decomposition of methane (CDM). The properties of catalysts
to 5:4 trillion m3 in 2040 [1]. This is partly because replacing coal applied and the results achieved in CDM have been extensively
with natural gas in power production is seen as attractive due to summarized in reviews [8,9].
the lower carbon dioxide ðCO2 Þ emissions of the latter. However, Muradov [5] almost two decades ago presented a conceptual
the regulation of CO2 emissions is becoming stricter as, for exam- idea to utilize TDM for the on-site production of hydrogen/
ple, the European Union has set a target of a 40% reduction in methane blends from natural gas. According to the presented anal-
greenhouse gas emissions by 2030 compared to the levels in ysis, the TDM process consumes 1.7 times more methane in hydro-
1990 [2]. Consequently, the CO2 emissions has to be reduced also gen production than steam reforming. However, steam reforming
in the use of natural gas [3,4]. produces 5 times more CO2 emissions than TDM, and thus TDM
The thermal decomposition of methane (TDM) is an endother- was seen as the most potential process for fossil fuel-based pro-
mic reaction where methane is thermally converted to solid car- duction of hydrogen without CO2 emissions. Later, Muradov and
bon, gaseous hydrogen, and traces of higher hydrocarbons. Eq. Veziroğlu [10] presented hydrogen production based on TDM as
(1) presents a simplified reaction equation of TDM [5]. a transitional period solution before shifting to a renewable-
based hydrogen economy in the future. The key challenge of
TDM was stated to be the enormous amount of solid carbon pro-
CH4 ! C þ 2H2 DH0r ¼ þ76 kJ=molCH4 ð1Þ
duced. Three potential end uses for large amount of carbon were
The equilibrium curve of TDM reaction presented, for example, presented: material for building and construction, electricity pro-
in [6] shows that the degree of methane conversion increases with duction by direct carbon fuel cells, and soil amendment.
temperature. However, temperatures above 1200 °C are required TDM concept can be considered as an alternative technology for
to achieve a nearly complete conversion when no catalyst is carbon capture and storage (CCS) in natural gas combustion. Com-
pared to the conventional pre-combustion CCS technology that
⇑ Corresponding author. Tel.: +358 50 301 4280. requires several process steps TDM could be a simpler process,
E-mail address: tiina.keipi@tut.fi (T. Keipi). and furthermore solid carbon from TDM can be a valuable product

http://dx.doi.org/10.1016/j.enconman.2015.11.057
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
2 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

instead of gaseous CO2 , which is in most cases an unwanted pro- available technology. All assumptions and input values of the eval-
duct [11,12]. Furthermore, carbon capture in solid form in TDM uations are based on literature data, product information, expert
prevents many of the problems related to the transportation and opinions, and analysis by the authors.
storage of gaseous CO2 such as risk of CO2 leakage or migration
in sediments, which could have serious consequences for the envi- 2. Theory and concepts
ronment and human beings [13–16].
An extensive amount of studies [17–23] that include either or 2.1. Carbon black
both technical and economic analyses related to the pre-
combustion carbon capture from natural gas are found in the liter- Carbon black is a generic term for materials that are mainly
ature, and the most relevant results are briefly summarized. Kan- composed of elemental carbon and have extremely small particle
niche et al. [17] conducted a techno-economic study for different size and high surface area. Currently, carbon black is mainly pro-
CCS technologies that are applied to pulverized coal combustion, duced from heavy oil by the oil-furnace process. The reaction tem-
integrated gasification combined cycle, and natural gas combined perature is 1400–1800 °C and the heat is provided by the partial
cycle (NGCC). Kanniche et al. stated that pre-combustion capture combustion of aromatic oils in the feedstock. A less frequently
by the reforming of natural gas is fairly expensive compared to applied process is the thermal black process (temperature require-
other CCS methods and it should be used rather in the production ment around 1300 °C) in which thermal decomposition of natural
of hydrogen than in energy production. Lozza and Chiesa [22,23] gas is applied to produce so-called thermal black. Thermal black
conducted an economic analysis of pre-combustion capture tech- is used in various rubber and plastic applications [26].
nology where partial oxidation (PO) is combined with a NGCC The production cost of carbon black is dominated by the feed-
and these costs compared to those of post-combustion carbon cap- stock cost, whereas the value of carbon black is highly dependent
ture from NGCC. According to the results, adding the PO process to on its morphology. Public data about the present prices of carbon
NGCC increased the relative investment cost ($/kW) by 40–59%, black is not widely available, and therefore, rough estimations
increased the cost of electricity by 36–43%, and resulted in a cost based on the expertise of the authors are utilized in this study.
of avoided CO2 as 39–46$/tCO2 . with 90% CO2 capture. Manzolini According to a commercial source [27], the price of carbon black
et al. [24] studied the cost of pre-combustion CO2 capture based with special quality may vary from 3300 to 4000 €/tcarbon and with
on pressure swing adsorption (PSA) in a natural gas combined standard quality from 2300 to 3000 €/tcarbon . The price of thermal
cycle. According to the results, the presented pre-combustion pro- black is estimated to vary between 500 and 1400 €/tcarbon . The price
cess based on PSA increased the specific electricity production of the product carbon in this study was chosen to be equal with the
costs by 33–36% compared to NGCC without CO2 capture. With a smallest price of thermal black, 500 €/tcarbon , in the scale given in
conventional pre-combustion capture based on methyldiethanola- [27], The smallest price in the scale varying from 500 to
mine (MDEA) the increase was stated to be 38%. 1400 €/tcarbon was chosen to avoid overestimating the value of the
One of the rare economic analyses of TDM was presented in a product carbon. The amount of produced carbon in the concepts
study by Triphob et al. [25], in which the hydrogen produced by is calculated from the reaction equations of TDM, which are pre-
TDM was utilized in a fuel cell. They presented the costs of electric- sented in Eqs. (2) and (5).
ity production with a solid oxide fuel cell (SOFC) when the hydro-
gen was produced alternatively with methane steam reforming or 2.1.1. Environmental and health effects of carbon black
with TDM. An important benefit of TDM is the absence of COx in The main idea of TDM concept is to decrease the CO2 emissions
the product hydrogen, which extends the lifetime of SOFC. How- to the atmosphere and produce valuable solid carbon instead.
ever, the solid carbon particles from TDM can be harmful to SOFC. However, the product carbon may have negative effects on envi-
As a conclusion, Triphob et al. stated that TDM provides technical ronment and humans. These effects are shortly summarized here
advantages over, and is an economically more feasible method and more detailed information can be found in the references.
than methane steam reforming when producing hydrogen for Nanoparticles have a representative dimension such as diame-
SOFC. This is partly because TDM produces carbon, which was ter of a particle or diameter of a fiber is less than 100 nm. In carbon
assumed to be a valuable by-product, whereas methane steam black, the aggregate dimensions vary from tens to a few hundred
reforming produces undesired CO2 emissions. nanometers. The nanoparticles have a high tendency to agglomer-
There is a very limited amount of research presented in the lit- ate and the carbon black agglomerates typically have a diameter
erature that further discusses the conceptual idea of TDM as a pre- less than 2 mm. Thus, carbon black occur both as nanoparticles
combustion capture technology. Therefore, the techno-economic and as fine particles, which are defined as particles with a diameter
analysis presented in this paper produces valuable information. less than 2.5 lm [28].
TDM can be applied to produce hydrogen for purposes other than The International Agency for Research and Cancer has defined
on direct power production, such as upgrading biofuels or for other carbon black as ‘‘possibly carcinogenic to humans”. The primary
chemical processes, but this option is not further evaluated in this industrial exposure to carbon black occurs in the carbon black pro-
study. The theoretical background of TDM reaction and carbon duction industry. In atmosphere, the effects of carbon black parti-
black as well as the technical descriptions of the four concepts cles are the same as fine particles in general. For humans, fine
are given in Section 2. The assumptions and the basis of the particles cause different lung and heart diseases. The fine particles
techno-economic analysis are presented in Section 3. The results may reduce the visibility of air. However, as carbon black is not sol-
of the technical analysis, including the mass and energy balances, uble in water, it will eventually end up in sediments and soil. Car-
and economic analysis, including the investment and operational bon black is not degradable, and therefore it is persistent in the
costs, are presented and discussed in Section 4. The discussion environment. However, carbon black does not bioaccumulate and
focuses on the cost of electricity production and the cost of avoided the potential of being toxic to aquatic organisms is low [28–31].
CO2 emissions in each concept. Furthermore, sensibility analysis
has been conducted to better understand the effect of product car- 2.2. Reaction equations
bon price, operation time, and degree of methane conversion on
the feasibility of the concepts. Section 5 concludes the results of The overall one-step reaction equations of TDM in the concepts
this study and presents recommendations for further research. are presented in this section. The heat demand of TDM reaction
The technical solutions chosen in this study represent the currently and reaction products in each concept are calculated based on
 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12 3

the equations. The required heat demand of TDM reaction is pro- the stoichiometric combustion of input methane. The air factor y is
duced by partial combustion or by an external heat source. The iteratively determined using Eqs. (3) and (4) so that the heat from
technological implementation and the process flow charts of the the partial combustion of input methane equals the heat of
concepts are presented in Section 2.3. conversion.

2.2.1. Concept 1 – partial combustion 2.2.2. Concepts 2–4 – thermal decomposition

In Concept 1, a part of the input methane is combusted to pro- In Concepts 2–4, where no air or oxygen is introduced in the
duce the heat demand of TDM reaction of the remaining methane. conversion reaction, methane reacts according to Eq. (5). This reac-
The overall reaction equation combines the methane combustion tion equation is modified from Eq. (1) by defining the conversion
and TDM as is shown in Eq. (2). degree a as the molar fraction of decomposed methane.

CH4 þ 2yðO2 þ 3:77N2 Þ ! aðC þ 2H2 Þ þ ð1  a  yÞCH4 CH4 ! aðC þ 2H2 Þ þ ð1  aÞCH4 ð5Þ
þ yðCO2 þ 2H2 O þ 7:52N2 Þ ð2Þ The methane conversion degree is denoted by a similarly as in
Concept 1. The TDM reaction occurs according to Eq. (5) in TDM
The degree of methane conversion, i.e., the molar fraction of reactors in Figs. 2 and 4. Furthermore, the same reaction occurs
decomposed methane, is denoted by a. The degree of methane con- in the CDM reactor shown in Fig. 3.
version depends on the reaction conditions, such as temperature For Concepts 2–3 the heat of conversion ðQ Þ, is the sum of sen-
and reaction time. Thus, the achieved degree of methane conver- sible heat and reaction enthalpy ðDHr Þ:
sion is dependent on the selection of reactor parameters such as
dimensions, volumetric gas flow, and temperature. Eq. (2) repre- Q ¼ N_ CH4 cp;CH4 MCH4 DT CH4 þ a  N_ CH4 DHr ; ð6Þ
sents the overall reaction equation that occurs in combustion
where the terms are defined similarly as in Eq. (3). For Concept 4,
and quench chamber shown in Fig. 1.
the heat of conversion ðQ Þ also includes the sensible heat of the
Heat of conversion (Q) is the total energy requirement of the
bed material stream:
conversion process in the concepts. In Concept 1 the heat of con-
version is the sum of the required reaction enthalpy ðDHr Þ and Q ¼ N_ CH4 cp;CH4 MCH4 DT CH4 þ N_ bed cp;bed M bed DT bed þ a  N_ CH4 DHr ; ð7Þ
the sensible heat of air and methane:
where N_ CH4 is the molar flow rate of input methane, cp;CH4 the speci-
Q ¼ N_ CH4 cp;CH4 M CH4 DT CH4 þ a  N_ CH4 DHr þ N_ air cp;air Mair DT air ; ð3Þ fic heat capacity of methane, MCH4 the molar mass of methane, and
DT CH4 the difference between the reaction temperature and the
where N_ CH4 is the molar flow rate of input methane, cp;CH4 the speci-
temperature of the preheated methane stream.
fic heat capacity of methane, MCH4 the molar mass of methane, and
DT CH4 the difference between the reaction temperature and the
2.3. Concept descriptions
temperature of the preheated methane stream, N_ air the molar flow
rate of input air, cp;air the specific heat capacity of air, Mair the molar The four TDM concepts analyzed in this study are presented in
mass of air, and DT air the difference between the reaction tempera- more detailed in this section. In all the concepts methane and other
ture and the temperature of the preheated air stream. The reaction fluids are internally preheated prior to the reactor in order to min-
enthalpy DHr depends on the reaction temperature, and the values imize the heat from additional sources. The internal heat transfer
applied in this study are adapted from [32]. from the product gas and product carbon of TDM reaction to the
The mass flow rate of methane which is combusted to supply input gas streams is conducted by a closed cooling water (CCW).
the energy demand in Concept 1 ðm _ comb:;CH4 Þ is calculated according The additional heating that is required after preheating is produced
to the following equation: by partial oxidation (Concept 1) or electricity (Concept 2 and 4).
Q Solid carriers are utilized for heating in Concepts 3 and 4.
_ comb:;CH4 ¼
m ; ð4Þ
LHV CH4
2.3.1. Methane burner
where LHV CH4 is the lower heating value of methane. The overall air Concept 1 is based on the partial oxidation of methane that pro-
factor denoted by y in Eq. (2) is defined as the ratio of molar air flow vides the heat of conversion. Since adding a catalyst to a burner is
rate ðN_ air Þ to the molar air flow rate that would be needed to achieve extremely challenging, non-catalytic thermal decomposition is

Air Methane Electricity

Cyclone
Preheat Preheat TDM
Methane
Fabric filter
Preheat
Product Carbon
Comb. Cooling
gas Cooling
Cyclone
Product gas Fabric filter
TDM Cooling
Combustion Quench Preheat Preheat
Cooling
Cooling Carbon Cooling
Carbon
Product
carbon Product
carbon
System boundary for mass and energy balance calculations System boundary for mass and energy balance calculations

Fig. 1. Process diagram of Concept 1. Fig. 2. Process diagram of Concept 2.

4 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

from inside and is operated at atmospheric pressure. The product

Fluidized bed reactor Preheat Fabric filter
Product gas carbon is separated from the product gas with a cyclone and a fab-
ric filter. The product gas is cooled down to a temperature suitable
for fabric filters. The product carbon is cooled to a temperature that
CDM Cooling
is below its self-ignition temperature before being exposed to air.
Cooling
Carbon
2.3.3. Catalytic fluidized bed
The catalytic fluidized bed reactor concept is highly similar to
Cooling Product carbon
existing technology applied in the petroleum refining industry
Compression unit [34]. Concept 3 applies the catalytic decomposition of methane
CO2
(CDM), and therefore the process can be conducted at typical flu-
Preheat idized bed reaction temperatures (750–950 °C).
Electricity
Bed material O2 The fluidizing agent in Concept 3 is methane. The catalyst is
ASU Cooling
Regeneration mixed with bed material (sand) which also acts as a heat carrier
Methane reactor Air N2 to provide the heat needed in the TDM reaction. Adding the cata-
System boundary for mass and energy balance calculations
lyst to the fluidized bed enables a large reactive surface area
between the catalyst and methane. The drawback of this concept
Fig. 3. Process diagram of Concept 3. is that the product carbon contains catalyst particles as an impu-
rity. Furthermore, regeneration of the catalyst is required. The cat-
alyst regeneration is conducted by circulating bed material to the
applied in this concept. The degree of methane conversion that can regeneration reactor. Additionally, a part of the product carbon is
be achieved is highly dependent on the burner geometry. Increas- combusted is the regeneration reactor to heat up the bed material.
ing the reaction time by lengthening the reaction zone increases In order to enable a more efficient capture of the resulting CO2
the reaction time and further enables higher methane conversion stream, oxy-combustion is applied in the regeneration reactor.
degrees. Oxygen is produced in an air separation unit (ASU). A part of the
The methane burner process may have two possible practical flue gas it is circulated back to the regeneration reactor in order
arrangements. In the first option, the partial oxidation and the to decrease the oxy-combustion temperature to a suitable level.
TDM reaction occur at the same time in a burner. In the second The rest of the CO2 is captured and compressed to a suitable pres-
option, part of the methane is combusted separately in a burner sure level (110 bar) for storage [35].
and the flue gas is introduced to an insulated quench chamber The process diagram of Concept 3 is illustrated in Fig. 3. The
where the TDM reaction occurs. The latter configuration was cho- reactor type in Concept 3 is a cylinder shaped fluidized bed reactor
sen as the basis of Concept 1. with a cyclone for the separation of carbon from the product gas
The total methane flow that is combusted was chosen to be stream. The reactor is refractory-lined from inside. Methane is pre-
divided into four parallel burners in order to achieve convenient heated with hot product gas before it enters the reaction chamber.
burner dimensions. The dimensions of the reaction chambers and The product gas is further cooled to a temperature that is suitable
burners were chosen based on the gas velocity and the residence for fabric filtering.
time in the reactor. The TDM reaction occurs at atmospheric pres-
sure. A process flow diagram of Concept 1 is shown in Fig. 1. For 2.3.4. Regenerative heat exchanger
simplicity the combustion and the quench are shown as two sepa- The regenerative heat exchanger reactor (RHER) operates as a
rate processes in the figure. counter-flow heat exchanger between methane (flowing upwards)
The product carbon is separated from the product gas with a and bed material (flowing downwards). Therefore, the preheating
cyclone and a filter. Besides the heat transfer from the product of the input flow occurs inside the reactor. The reactor type is a
gas and carbon to the input streams, the products have to be fur- refractory-lined cylinder and the bed material is chosen based on
ther cooled down to achieve a suitable temperature for conven- its capacity to transfer heat to and from the gases. The reactor tube
tional fabric filters. is divided into three zones: the reaction zone in the middle, where
the TDM reaction occurs; the upper zone, where the incoming bed
2.3.2. Power-to-carbon material receives heat from the generated product gas; and lower
In the power-to-carbon concept, electricity is used to provide zone, where the incoming methane stream receives heat from the
the heat of conversion. The TDM reactor in Concept 2 can be pow- bed material stream. Thus, the bed material acts both as a pre-
ered with electrical heating elements or alternatively with a heater for the methane and as a cooler for the product gas. An
plasma torch. The heat of conversion can be provided by utilizing, external heat source is introduced in the reaction zone to provide
for example, heating elements based on materials such as silicon- the heat of conversion. Heat to the reactor can be produced in sev-
carbides (SiC) or molybdenum-disilicates ðMoSi2 Þ. These materials eral ways, e.g., natural gas combustion or electrical heating ele-
can typically handle temperature ranges up to 1300–1600 °C. ments. In this study, electric heating was chosen and the heating
Applying a plasma torch enables the utilization of a high tem- elements were placed near the inner surface of the reactor. The
perature range (1200–2200 °C) resulting in high degrees of heat transfer in the reactor is further analyzed in Section 3.2.
methane conversion and high purity of product gas. The product The process diagram of Concept 4 is shown in Fig. 4. Part of the
carbon properties and quality can be modified by adjusting the product carbon accumulates on the surface of the bed material. The
temperature of the plasma torch. According to the computational bed material is constantly flowing downwards, and thus the pro-
model presented by [33] the temperature should be restricted duct carbon can be removed from the reactor. In a commercial
below 2200 °C to prevent the production of a significant amount scale reactor, after the carbon extraction the bed material is circu-
of harmful C2 H2 and C2 H radicals. lated back to the reactor by lifting the bed material up either
Concept 2 is illustrated in Fig. 2. Electrical heating elements mechanically or pneumatically. The particle size and the material
were chosen to provide the heat of conversion in Concept 2. The of the bed material can be chosen so that the heat transfer in the
input methane is preheated by hot product gas before being fed reactor and carbon removal are optimized. Part of the solid carbon
to the reaction chamber. The reaction chamber is refractory-lined is extracted by filtering the product gas.
 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12 5

product carbon price on the feasibility of the concepts was studied

Fabric filter
with a sensibility analysis.
Bed material Product gas The reference case in this study is the direct combustion of
Cooling natural gas, which is for the sake of clarity replaced with methane
in all calculations, to produce electricity with a steam power
Regenerative
Electricity
heat cycle. The reference plant is assumed to be an operating plant
TDM exchanger Cooling which still has operation time left and whose investment costs
reactor have been paid back already. The fuel input in the reference case
Bed is 150 MW, which is equal with the concepts. The efficiency of the
material Cooling electricity production in the reference case is 30%, the same as in
silo
Carbon all concepts, and no CCS is applied. Thus, the electricity produc-
tion cost includes the cost of emission allowances (a moderate
Methane Product carbon
value of 20 €=tCO2 assumed). It was decided that the product gas
would be utilized to produce only electricity, and not, for exam-
System boundary for mass and energy balance calculations
ple, additional heat. As the efficiency of the electricity production
Fig. 4. Process diagram of Concept 4. is rather low, the cost of electricity production in the reference
case is 163 €=MWhe which is considerably higher than the 10-
year average Nord Pool system price ð40 €=MWhe Þ [36]. However,
3. Techno-economic analysis despite the high absolute costs of electricity production, it is pos-
sible to compare the concepts to the reference case in relative
The calculation basis and the assumptions made in the analysis terms.
are presented in this section. The analysis presented in this paper is It is important to note that the results achieved in this study are
the preliminary evaluation of the feasibility of these concepts. not general as those highly depend on the assumptions made in
Therefore, some input data is still uncertain as it is taken from the analysis. However, the results can be used to compare the four
the literature or is based on assumptions made by the authors. different TDM concepts with each other.
The methodology used in the techno-economic analysis is illus-
trated in Fig. 5.
3.2. Technical assumptions

3.1. Calculation basis

The main assumptions of the concepts are summarized in
Table 1. All the concepts have a fuel input of 150 MW and for the
The technical solutions chosen in this study represent the cur-
sake of clarity, all the calculations are conducted for methane, as
rently available technology. In all the concepts, the product gas is
natural gas contains between 70% and 98% methane, depending
combusted after TDM in order to produce electricity with a steam
on its origin [37,38]. The product gas is assumed to be a mixture
power cycle. It might also be possible to adjoin the TDM pro-
of hydrogen and unreacted methane that contains solid carbon
cesses to an existing power or heat boiler. The possible end use
particles. In Concept 1, the combustion products (N2 ; H2 O, and
of the product gas in a gas turbine is not considered in this paper
CO2 ) are also present in the gas mixture due to the partial oxida-
since the amount of hydrogen that the current gas turbines can
tion that is used to provide the heat of conversion. It is assumed
handle is limited and also the solid carbon in the product gas
that in this case the combustion produces CO2 , and thus no CO is
might cause special requirements for the gas turbine [11]. Since
present.
the focus in the study is to reduce CO2 emissions, the purity in
The reactor temperatures were selected based on the literature
terms of hydrogen content in the resulting product gas was not
values summarized in [10]. The degree of methane conversion and
of interest.
reaction time in each concept were selected based on experimental
In this study, the produced carbon is assumed to be sold in the
research and calculations conducted at Tampere University of
carbon black market with a price of 500 €=tcarbon . The effect of
Technology. The degree of methane conversion as a function of
temperature and residence time has been experimentally studied
and the results have been used to determine the reactor dimen-
sions in each concept. Heat losses for the concepts were assumed
TDM Mass and
Reactor according to typical levels in similar thermal conversion chambers
reaction energy
design with an emphasis on reaction temperature and reactor surface
equations balances
area. The heat losses in all the concepts were assumed to be 1.0–
2.0% of the fuel input.
Produced Avoided Methane and air are assumed to enter the system at a temper-
Product Process ature of 25 °C. Due to the poor heat transfer coefficient factors in
net CO2
carbon design
electricity emissions gas-to-gas heat transfer, the chosen temperature differences
between heating streams are kept at a moderate level (logarithmic
Income temperature differences of 400–750 °C) in order to decrease the
Cost of
from the cost of heat exchangers. Therefore, the temperature of methane
similar
Cost of net product and air after preheating was chosen to be in the range of 600–
equipment
electricity carbon 700 °C. The lower temperature of 600 °C in Concept 3 is based on
production Annual CAPEX & the lower temperature of the product gas. The in situ preheating
Cost of cost OPEX in Concept 4 was calculated with a chosen terminal temperature
avoided CO2 difference (TTD) between bed material and methane stream
emissions (500 °C). Since the temperature of the bed material was 1200 °C
when it leaves the reaction zone, the temperature of the preheated
Fig. 5. Diagram of the methodology applied in the analysis. methane was 700 °C. Thus, it is assumed that heat is effectively
6 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

Table 1
Input values and assumptions made in analysis.

Assumption Concept 1 Concept 2 Concept 3 Concept 4

Fuel input 150 MW 150 MW 150 MW 150 MW
Annual operation time 4500 h 4500 h 4500 h 4500 h
Degree of conversion 0.40 0.35 0.40 0.30
Reaction temp. 1350 °C 1350 °C 900 °C 1200 °C
Residence time 3.2 s 3.1 s 4.0 s 3.4 s
Heat of conversion 93 MJ/kmol 93 MJ/kmol 91 MJ/kmol 92 MJ/kmol
Heat losses 2.0% 2.0% 1.5% 1.0%
Temp. (air in) 25 °C N/A N/A N/A
Temp. (methane in) 25 °C 25 °C 25 °C 25 °C
Temp. of methane after preheating 700 °C 700 °C 600 °C 700 °C
Temp. of air after preheating 700 °C N/A N/A N/A
Temp. of product gas in filtering 250 °C 250 °C 250 °C 250 °C
Carbon separation eff. in cyclone 60% 60% 60% N/A
Carbon separation eff. in reactor N/A N/A N/A 50%
Carbon separation eff. of filter 100% 100% 100% 100%
TTD on carbon cooling 5 °C 5 °C 5 °C 5 °C
Temp. of CW in 25 °C 25 °C 25 °C 25 °C
Temp. of CW out 35 °C 35 °C 35 °C 35 °C
Pressure drop in CW cycle 4 bar 4 bar 4 bar 4 bar
CW pump eff. 78% 78% 78% 78%
Air factor (overall reaction) 0.20 N/A N/A N/A
Catalyst No No Yes No
Catalyst charge in reactor N/A N/A 0.1 kg/(m3CH4 h) N/A
Catalyst lifetime N/A N/A 50 h N/A
Regeneration temp. N/A N/A 1200 °C N/A
Power of oxygen production N/A N/A 160 kWh/tO2 N/A
CO2 compression work N/A N/A 0.4 GJ/tCO2 N/A
Heat capacity of bed material N/A N/A N/A 0.42 kJ/(kg K)
Heat duty of preheaters N/A N/A N/A Equal to each other
Mass flow of bed material N/A N/A N/A 20 kg/s

transferred from the bed material to methane, justifying the rela- Bed material
tively high temperature of methane after the preheating zone.
Product gas
Due to the small particle size of the product carbon, the Tgas =1200 °C
cyclones in Concepts 1–3 are assumed to have a carbon separation Product gas
efficiency of 60%. However, the following filters are assumed to Tbed
Upper T
gas,out
extract all the remaining carbon from the product gas. As a whole, zone
it is assumed that all the product carbon is collected and can be Bed material
further utilized. The product gas temperature after cooling is
Tbed=25 °C
assumed to be 250 °C, which is a typical temperature that conven-
tional fabric filters can handle.
Reaction Heat input
The cooling of the product carbon is assumed to have TTD of zone
5 °C compared to cooling water (CW), which means that the
temperature of the product carbon is 30 °C when it leaves the Tbed= 1200 °C
system. The closed cooling water system in each concept is Bed material
dimensioned based on the following initial values: inlet temper- Lower TCH4 =700 °C
ature, mass flow, pressure drop, and pump efficiency for cooling zone T
water. bed,out
Methane
Nickel is a widely studied and generally found as suitable cat- TCH4=25 °C
alyst for methane decomposition [6,8,9]. Therefore, a nickel
based mixture (80% Ni with a support material based on alu- Bed material Methane
minum), was chosen as a catalyst material in Concept 3. The
required catalyst charge of 0:1 kgcat =ðnm3CH4 hÞ was chosen based Fig. 6. Heat transfer in Concept 4. The numeric values in figure are based on expert
opinions and are not experimentally verified.
on study [39], and the lifetime of 50 h for a Ni-based catalyst
based on the study in [9]. The minimum fluidization velocity
was one parameter determining the reactor dimensions. The In Concept 4, the bed material was assumed to have an input
bed sand was heated by oxy-combustion in the regeneration temperature of 25 °C and mass flow of 20 kg=s. The bed material
chamber to a temperature of 1200 °C to maintain the reaction was assumed to be steel and the material properties were chosen
temperature of 900 °C in the reactor. Also the auxiliary electric- accordingly. The heat transfer in Concept 4 is presented in more
ity consumption of the ASU was taken into account with the detail in Fig. 6.
best practice value for cryogenic air separation process The amount of heat transferred from the bed material to
ð160 kWh=toxygen Þ [40]. CO2 was captured from the ASU and its methane in the lower zone and from product gas to bed material
compression to 110 bar was assumed to require a work amount in the upper zone are assumed to be equal, as is presented in the
of 0:4 GJ=tCO2 [41]. following equation:
 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12 7

_ CH4 cp;CH4 DT CH4 ;lower zone ¼ m

Q preheat ¼ m _ bed cp;bed DT bed;lower zone 4. Results and discussion
_ bed cp;bed DT bed;upper zone ¼ m
¼m _ gas cp;gas DT gas;upper zone ð8Þ
4.1. Mass and energy balance results
Heat transfer in the reactor is based on expert opinions and is
still to be experimentally verified. The mass and energy balance calculations were conducted with
the aforementioned calculation assumptions. The system bound-
aries for the mass and energy balance calculations are shown in
3.3. Evaluation of investment costs
the process diagrams in Section 2.3. The results of the mass balance
calculations are summarized in Table 2. All the concepts have a
The investment costs for the process equipment were based on
methane mass flow of 3 kg/s, which is equal to the fuel power
nominal price approximations in existing similar devices found in
input of 150 MW based on the lower heating value of methane.
the power industry, such as natural gas burners, thermal conver-
Concept 1 has also an additional incoming stream of air (8 kg/s),
sion reactors, CFB reactors, air-to-air heat exchangers, flue gas
which considerably increases the total mass flow in this concept.
cleaning components, and flue gas ducts. For simplicity the cost
The outcoming mass flows of product carbon correlate well with
of the preheaters is approximated according to the price data from
the degree of methane conversion in each concept. In Concept 3,
non-pressurized air-to-air shell and tube heat exchangers. The
0.4 kg/s of the primarily produced carbon (0.9 kg/s) is combusted
price of the steel bed material in Concept 4 was assumed to be
to provide the heat of conversion (see Table 2).
600 €/t. The erosion rate of the bed material is assumed to be slow,
Results from the energy balance calculations (Table 3) show the
and therefore the cost of the bed material is allocated to the invest-
division between chemical (fuel) energy and sensible heat in pro-
ment costs.
duct streams. Around 11–20% of the outcoming energy is bound
Cost distribution for total capital expenditure (CAPEX) was
to the solid carbon. The rest can be accounted for the product
assumed to be similar to a typical power plant EPC-delivery, which
gas as a chemical energy or sensible heat part of which is removed
includes equipment costs, process costs, automation and
by cooling.
electrification costs, civil costs, and project costs (e.g. project
In Concept 3 a part of the carbon stream is combusted to pro-
management, engineering, and start-up). Costs for the initial
vide the heat of conversion, and the energy content of the product
supply material loads (catalyst in Concept 3 and bed material in
carbon is thus smaller than in the other concepts. Applying the cat-
Concept 4) were included in the project costs. Contingency
alyst in Concept 3 decreases the temperature that is required to
for the cost estimate was assumed to be 30% of the equipment
achieve the TDM reaction, and consequently the heat of conver-
costs.
sion. The reaction enthalpy is 91 MJ/kmol in Concept 3 compared
to the 93 MJ/kmol required in Concepts 1 and 2 where the temper-
3.4. Evaluation of operating costs ature is 1350 °C. However, the differences in the sensible enthalpy

The operational expenditure (OPEX) consisted of operating and Table 2

maintenance (O&M) costs (2% of CAPEX), fuel costs (45 €=MWhe ), Results from the mass balance calculations.
and other input material costs. The other materials were nickel Concept 1 Concept 2 Concept 3 Concept 4
based catalyst (10 €=kg) and the sand of the fluidized bed (4 €=t) (kg/s) (kg/s) (kg/s) (kg/s)
in Concept 3. The costs of supply materials were based on the price Mass balance, input
level of nickel, aluminum, and sand. The costs of CO2 emissions Methane 3.0 3.0 3.0 3.0
were calculated based on the emissions in each concept with a Air 8.1 0 5.0 0
moderate cost of emission allowances (20 €=tCO2 ). The product car- 11.1 3.0 8.0 3.0

bon was assumed to have a price of 500 €=tcarbon , which has been Mass balance, output
presented as a moderate value for low quality carbon black named Product gas 10.2 2.2 2.1 2.3
Product carbon 0.9 0.8 0.5 0.7
as thermal black. Selling the product carbon was the only source of
CO2 (regeneration) 0 0 1.6 0
income in the concepts. N2 (ASU) 0 0 3.8 0
In this study CAPEX was evenly divided in the chosen invest-
Reaction products 11.1 3.0 8.0 3.0
ment period of 10 years. The annual cost of electricity production
was calculated as follows:

Annual cost of electricity production

Table 3
¼ OPEXannual þ CAPEXannual  INCOMEannual ð9Þ Results from the energy balance calculations.

Concept 1 Concept 2 Concept 3 Concept 4

The annual cost of electricity production was divided by the
(MW) (MW) (MW) (MW)
annual electricity production to find out the specific costs
Energy balance, input
ð€=MWhe Þ. With certain input parameters, the annual cost
Methane, fuel energy 126 150 150 150
becomes negative, which indicates a positive income with the Methane, combustion 23.6 0 0 0
assumptions in question. The owner’s costs (e.g. taxes, permits Electricity 0 16.3 1.3 16.0
and insurances), site specific costs (e.g. land use), and costs related Total 150 166 151 166
to the existing power plant (e.g. modifications in the power cycle)
Energy balance, output
were not included in the analysis. Carbon, fuel energy 29.5 25.8 16.3 22.1
For simplicity, the interest rate and inflation were assumed to Carbon, sensible heat 0 0 0 0
be zero. In all concepts, CAPEX is small in comparison with OPEX, Cooling 9.5 6.0 2.6 7.3
and therefore the interest rate and inflation have a minor influence N2, sensible heat 0 0 0 0
CO2, sensible heat 0 0 2.4 0
on the annual cost of electricity production. However, CAPEX Product gas, sensible heat 4.4 2.1 2.5 2.5
and OPEX in each concept are detailed in Table 5, and thus the Product gas, fuel energy 107 132 127 134
effect of the investment rate and inflation can be calculated Total 150 166 151 166
afterwards.
8 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

of methane has a greater impact. The sensible enthalpy of methane net electricity production in each concept was calculated by sub-
is 89.2 MJ/kmol at 1350 °C, but only 51.4 MJ/kmol at 900 °C [32]. tracting the internal electricity consumption of the process from
Consequently, due to the implementation of the catalyst, the heat the total electricity production. The specific cost of net electricity
of conversion in Concept 3 is 22% less than that in Concept 1 and 2. production is calculated by dividing the annual cost of electricity
The resulting product gas molar compositions and lower heat- production by the annual net electricity production.
ing values (LHV) after carbon extraction are shown in Table 4. Fuel is the most significant cost in all the concepts. CAPEX is a
According to the results, the volumetric hydrogen content in the minor cost compared to OPEX in all concepts except Concept 3.
product gas was in the range of 24–57%. The highest content of Concept 3 is the most expensive both in terms of CAPEX and OPEX,
hydrogen occurred in Concept 3 and the lowest in Concept 1, which correlates well with the complexity of this process. How-
which is due to the presence of combustion products in Concept ever, due to the low internal electricity consumption of the pro-
1. Partial combustion in Concept 1 resulted also in lower hydrogen cess, the net electricity production in Concept 3 is relatively high,
and methane fractions due to the presence of the combustion resulting in moderate cost of electricity production. In Concept 1
products in the product gas. Consequently, the calculated LHV the low internal electricity consumption and high amount of pro-
was also lower in Concept 1. duct carbon results in the lowest cost of electricity production
among all the concepts. The cost of electricity production in the
4.2. Investment and operational costs reference case is 163 €/MWhe which includes both the fuel cost
and the cost of CO2 emission allowances. Thus, with the current
The annual cash flow estimate for each concept including assumptions, the cost of electricity production is 12–58% higher
CAPEX, OPEX, and operational incomes are shown in Table 5. The in the concepts than in the reference case. The effect of the product
carbon price to the feasibility of the concepts is further analyzed in
Table 4 Section 4.4.
Product gas molar composition and LHV in each concept after extraction of solid
carbon.
4.3. CO2 emissions
Molar Concept 1 Concept 2 Concept 3 Concept 4
composition (%) (%) (%) (%) The CO2 emissions of the concepts originate from the combus-
H2 (g) 28.0 51.9 57.1 46.2 tion of the product gas and in Concept 3 additionally from the
CH4 (g) 15.2 48.1 42.9 53.8 regeneration reactor. In the reference case, which is the direct
H2O (g) 10.8 0 0 0
CO2 (g) 5.4 0 0 0
combustion of methane, the CO2 emissions originate from the
N2 (g) 41.1 0 0 0 methane combustion. The CO2 emissions from the concepts were
LHV (MJ/kg) 10.4 59.8 60.6 57.7
compared with those from the reference case, as summarized in
Table 6. The reduction of the CO2 emissions in the concepts is

Table 5
Annual cash flow analysis for each concept. The arrows next to the values in the bottom row indicate whether the cost of electricity production is slightly higher (an upward
pointing arrow) or notably higher (two arrows pointing upward) compared to the reference.

Concept 1 (k€) Concept 2 (k€) Concept 3 (k€) Concept 4 (k€)

CAPEX
Equipment and project costs 11 110 7 226 24 916 9 029
Contingency (30%) 1 536 999 3 438 1 110
Total 12 646 8 225 28 354 10 139
OPEX
Fuel costs 30 375 30 375 30 375 30 375
CO2 emission costs 1 601 1 734 1 601 1 868
Other supply material costs 0 0 4 341 0
O&M costs (2%) 253 165 567 203
Total 32 229 32 274 36 884 32 445
Annual income
Product carbon sales 7 282 6 371 4 036 5 461
Annual cost of electricity production (k€) 26 211 26 726 35 683 27 998
Cost of net electricity production (€/MWhe) 183 255 216 257

Table 6
Net efficiency of electricity production and absolute and specific CO2 emissions from the concepts. The arrows next to the values in the bottom row indicate whether the concept
produce less (a downward pointing arrow) or more (an upward pointing arrow) CO2 emissions per produced electricity compared to the reference case.

Ref. case Concept 1 Concept 2 Concept 3 Concept 4

Fuel input MW 150 150 150 150 150
Operational hours h 4500 4500 4500 4500 4500
Annual net electricity production GWh 203 144 105 165 109
Electricity production net efficiency % 30.0 21.4 15.5 24.5 16.1
CO2 stored as carbon ktCO2 /a 54 47 30 40
CO2 capture in gaseous form ktCO2 /a 0 0 24 0
CO2 stored, total ktCO2 /a 54 47 54 40
CO2 emissions ktCO2 /a 134 80 87 80 93

Specific CO2emissions tCO2 /MWhe 0.66 0.55 0.83 0.48 0.85

 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12 9

mainly due to the production of solid carbon. Thus, in Table 6 the Fig. 7 summarizes the costs of electricity production in each
product carbon is presented as ‘CO2 stored as carbon’ which repre- concept when the price of product carbon was varied. According
sents the amount of CO2 emissions avoided when the product car- to the results, in Concept 1 the cost of electricity production was
bon is not combusted. As the fuel input was fixed in all concepts, the lowest among the concepts. With the aforementioned assump-
the absolute amount of CO2 emissions in the reference case repre- tions related to the operational parameters, the price of product
sents the total annual amount of CO2 from the concepts, which is carbon should be at least 700 €=tcarbon so that the cost of electricity
either stored or released to the atmosphere. Hence, the CO2 emis- production in Concept 1 would be equal or less than in the refer-
sions from each concept are the total amount of stored CO2 sub- ence case. The price of thermal black can be up to 1400 €=tcarbon ,
tracted from the CO2 emissions in the reference case. The as stated in Section 2.1. Assuming the price of product carbon of
absolute CO2 emissions in each case are divided by the correspond-
ing electricity production to achieve the specific CO2 emissions

Cost of electricity production (€/MWhe)

350
ðtCO2 =MWhe Þ. Concept 1
As the annual net electricity production in the concepts is less Concept 2
300
than in the reference case, the concepts actually produce more Concept 3
250 Concept 4
CO2 per MWh than the reference case, but part of the emissions
ref.
are captured. Therefore, the amount of captured CO2 is more than
200
is actually avoided by applying the concepts, as has been presented
in [17]. The actually avoided CO2 emissions are the difference 150
between the quantity of CO2 emissions from the reference case
and the quantity of CO2 emitted by a concept. This is the difference 100
between the specific CO2 emissions in the reference case and in a
concept in Table 6. 50
The amount of electricity produced in the concepts is less than
0
in the reference case. If the electricity production is wanted to 0 500 1000 1500 2000 2500 3000
remain as a constant, the fuel consumption has to increase. This -50
may lead to the increase of the CO2 emissions in the fuel transport- Price of product carbon (€/tcarbon)
ing. However, this has not been considered in this study. Further-
Fig. 7. Cost of electricity production in concepts with the variable prices of product
more, utilizing the product carbon may cause additional CO2 carbon.
emissions, but those have been left out from this study.
It is shown in Table 6 that the specific CO2 emissions of only
Concepts 1 and 3 are lower than in the reference case. Therefore,
Cost of electricity production (€/MWhe)

with the assumptions in question only Concepts 1 and 3 provide 250 4000 h/a
an actual CO2 reduction compared to the reference case. When Concept 3 6000 h/a
Concepts 1 and 3 are applied, the avoided CO2 emissions are 8000 h/a
200 ref.
respectively 0:11 tCO2 =MWhe and 0:18 tCO2 =MWhe compared to Ref
the reference case.
For Concepts 1 and 3 it is possible to calculate the cost of 150
avoided CO2 by dividing the difference between the electricity pro-
duction costs in a concept and in the reference case by the amount 100
of avoided CO2 in a concept. With the current assumptions, the cost
of avoided CO2 emissions was 182 €=tCO2 in Concept 1 and 294 Concept 1
€=tCO2 in Concept 3. The effect of product carbon price on the cost 50
of avoided CO2 emissions is further analyzed in Section 4.5.
In all the concepts, the efficiency of net electricity production is 0
lower compared to the reference case due to the reduced electricity 500 1000 1500 2000
production in the concepts. The relatively high initial electricity Price of product carbon (€/tcarbon)
consumption in Concepts 2 and 4 further decreases the efficiency
of electricity production. Due to both the high net efficiency of
350 a=0.4
Cost of electricity production (€/MWhe)

electricity production among the concepts and low specific CO2

a=0.6
emissions in Concepts 1 and 3, these two concepts will be further 300 Concept 3 a=0.8
evaluated.
250 ref.

200
4.4. Analysis of the cost of electricity production
150
The assumption that the internal electricity consumption of the 100
concepts is covered by the electricity production in concepts is Concept 1
unfavorable, as the cost of electricity production in the concepts 50
is significantly higher than the average market price of electricity.
0
By contrast, an assumption that the electricity consumed in the 0 500 1000 1500
concepts would be bought from the market at the market price -50
would increase the net electricity production and decrease the cost
-100 Price of product carbon (€/tcarbon)
of electricity production ð€=MWhe Þ in Concepts 2–4. The decrease
in the cost of electricity production would be most significant in Fig. 8. Electricity production costs in Concept 1 (gray) and Concept 3 (black) as a
Concepts 2 and 4 which have the highest internal electricity con- function of product carbon price when the operation time (upper chart) and the
sumption among the concepts. degree of methane conversion (lower chart) are varied.
10 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

1400 €=tcarbon , applying Concepts 2–4 results in cost of electricity economically more feasible to produce less carbon and more elec-
production that is almost equal with the reference case. Therefore, tricity. However, the price of the product carbon has to be at least
all the concepts could provide a feasible option to reduce CO2 emis- 600 €=tcarbon so that the cost of electricity production would be less
sions in natural gas combustion. than in the reference case.
Sensitivity analysis was conducted to find out the effect of sev-
eral parameters (the product carbon price, annual operational 4.5. Analysis of the cost of avoided CO2
time, and degree of methane conversion) on the feasibility of the
concepts. Concepts 1 and 3 have shown the best performance The effect of the product carbon price, annual operational time,
among all the concepts, and therefore those were chosen for the and degree of methane conversion on the cost of avoided CO2
sensibility analysis. Fig. 8 shows the effect of the annual operation emissions in Concepts 1 and 3 was analyzed and the results are
time and the degree of methane conversion on the cost of electric- shown in Fig. 9. Increasing the annual operation time from
ity production in Concept 1 and 3. When the degree of methane 4000 h/a to 8000 h/a decreased the cost of avoided CO2 emissions
conversion is varied, it is assumed that CAPEX remains the same, by 59—66 €=tCO2 . Thus, the annual operation time has a remarkable
meaning that the reactor or auxiliary components do not require effect on the cost of avoided CO2 emissions in the concepts.
any modifications. In Concept 3, the effect of conversion degree on the cost of
According to the analysis, in both concepts increasing the oper- avoided CO2 was remarkable. Increasing the degree of methane
ational time from 4000 h/a to 8000 h/a decreased the cost of elec- conversion from 0.4 to 0.8 decreased the cost of avoided CO2 in
tricity production by 6—12 €=MWhe when the investment period is Concept 3 by 159—242 €=tCO2 with product carbon prices of
10 years. Hence, the effect of the annual operation time on feasibil- 500—2000 €=tcarbon , respectively. In Concept 1, a moderate price
ity is almost insignificant. By contrast, the influence of the degree of product carbon (500 €=tcarbon ) and high conversion degree (0.8)
of methane conversion on the cost of electricity production is resulted in a moderate cost of avoided CO2 ; 51 €=tCO2 .
much more significant. Increasing the conversion degree increases Especially in Concept 1, increasing the product carbon price
the income from the product carbon and results in decreased rapidly decreases the cost of avoided CO2 . At product carbon prices
annual cost of electricity production according to Eq. (9). The pro- of 650—700 €=tcarbon the cost of avoided CO2 emissions becomes
duct carbon prices 500 and 600 €=tcarbon are found as limiting val- negative, indicating that the costs were smaller than in the refer-
ues in Concepts 1 and 3, respectively, above which it is ence case. Typically, the cost of CCS (including the capture, trans-
economically more feasible to produce more carbon and less port, and geological storage of CO2 ) in current technology for a
electricity. By contrast, below these limiting values it is new natural gas combined cycle power plant is in the range of
40—90 $=tCO2 avoided. [42] This cost level was achieved with pro-
duct carbon price of 500 €=tcarbon in Concept 1 and 800 €=tcarbon in
400 4000 h/a Concept 3 when the degree of methane conversion was in both
300 Concept 3 6000 h/a cases above 0.6.
Increasing the price of CO2 emission allowances does not
Cost of avoided CO2 (€/tCO2)

8000 h/a
200 directly improve the feasibility of the concepts due to the produced
100
CO2 in the concepts. If a complete conversion were achieved, that is
the degree of methane conversion of 1, Concepts 2 and 4 would not
0 emit any CO2 . In Concept 3, there would be CO2 emissions also
500 1000 1500 2000 with complete conversion, but they would be fully captured as gas-
-100
eous CO2 and stored.
-200

-300 5. Conclusion

-400
Concept 1 This study presents and analyses four concepts that utilize ther-
-500 mal decomposition of methane to reduce CO2 emissions in natural
Price of product carbon (€/tcarbon) gas combustion. Based on the results of the techno-economic anal-
ysis with the current assumptions, applying TDM reduces the CO2
400 a=0.4
emissions in electricity production (tCO2 =MWhe ) in two concepts.
300 Concept 3 The CO2 emissions decrease when a part of the carbon is taken
a=0.6
out from the feedstock in a solid form. Furthermore, the cost of
Cost of avoided CO2 (€/tCO2)

200 a=0.8
avoided CO2 decreases rapidly when more solid carbon is taken
100
out from the feedstock.
0
Concept 1, a methane burner, had a high performance in terms
-100500 1000 1500 2000
of the cost of net electricity production and the CO2 emissions per
-200 MWh of produced electricity. With the price of product carbon of
-300 600—700 €=tcarbon the price of electricity production in Concept 1
-400 was lower than in the reference case. With the assumptions in
-500 question, applying Concept 1 decreased the specific CO2 emissions
Concept 1
-600
ðtCO2 =MWhe Þ compared to the reference. At the same time, the cost
of avoided CO2 emissions became negative indicating smaller costs
-700
in Concept 1 compared to the reference. The high internal electric-
-800
ity consumption is the main drawback in Concept 2. Replacing
Price of product carbon (€/tcarbon)
electricity with plasma in Concept 2 could improve the quality of
Fig. 9. Cost of avoided CO2 emissions in Concept 1 (gray) and Concept 3 (black) as a
product carbon, and thus result in better feasibility. In Concept 3,
function of product carbon price when the operation time (upper chart) and the the complexity of the process results in high CAPEX. However,
degree of methane conversion (lower chart) are varied. the annual net electricity production is high, which results in low
 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12 11

CO2 emissions per MWh of produced electricity. The cost of CO2 research strategies. UK: Woodhead Publishing; 2013. p. 149–78. [chapter
Geological storage of carbon dioxide (CO2) – Geoscience, technologies,
capture and compression are taken into account, but the trans-
environmental aspects and legal frameworks].
portation and storage of CO2 will cause additional costs. In Concept [14] Blackford J, Widdicombe S, Lowe D. Environmental risks and performance
4, the low degree of methane conversion and high internal electric- assessment of carbon dioxide (CO2) leakage in marine
ity consumption results in the highest cost of net electricity pro- ecosystems. UK: Woodhead Publishing; 2010. p. 344–73 [chapter
Developments and innovation in carbon dioxide (CO2) capture and storage
duction and CO2 emissions per MWh of produced electricity technology, Volume 2: Carbon dioxide (CO2) storage and utilization].
among all the concepts. The technical solution to remove product [15] Ardelan M, Steinnes E, Lierhagen S, Linde S. Effects of experimental CO2
carbon from the reactor is an advantage in Concept 4. One poten- leakage on solubility and transport of seven trace metals in seawater and
sediment. Sci Total Environ 2009;407:6255–66. http://dx.doi.org/10.1016/j.
tial option to improve the feasibility of Concept 4 could be an scitotenv.2009.09.004.
autothermal heat production. [16] Lewicki J, Birkholzer J, Tsang C-F. Natural and industrial analogues for leakage
The price of product carbon was found to have a highly signifi- of CO2 from storage reservoirs: identification of features, events, and processes
and lessons learned. Environ Geol 2007;52:457–67. http://dx.doi.org/10.1007/
cant effect on the economy of the concepts. With the current s00254-006-0479-7.
assumptions, at the lowest a carbon price of 600—700 €=tcarbon [17] Kanniche M, Gros-Bonnivard R, Jaud P, Valle-Marcos J, Amann J-M, Bouallou C.
was required to obtain an equal cost of electricity production with Pre-combustion, post-combustion and oxy-combustion in thermal power
plant for CO2 capture. Appl Therm Eng 2010;30:53–62. http://dx.doi.org/
the reference case. In contrast to conventional CO2 capture tech- 10.1016/j.applthermaleng.2009.05.005 [Selected Papers from the 11th
nologies, in which unwanted CO2 is the end product, the value of Conference on Process Integration, Modelling and Optimisation for Energy
the product carbon from TDM could make the process profitable. Saving and Pollution Reduction].
[18] Petrakopoulou F, Tsatsaronis G. Production of hydrogen-rich fuels for pre-
Increasing the price of product carbon to the level of
combustion carbon capture in power plants: a thermodynamic assessment. Int
700—1500 €=tcarbon rapidly decreased both the cost of electricity J Hydrogen Energy 2012;37:7554–64. http://dx.doi.org/10.1016/j.
production and the cost of avoided CO2 emissions. TDM could be ijhydene.2012.01.147 [7th Petite Workshop on the Defect Chemical Nature
an especially suitable technology for peak load power plants. In of Energy Materials, 14–17 March 2011, Storaas, Kongsberg, Norway].
[19] Ertesvåg I, Kvamsdal H, Bolland O. Exergy analysis of a gas–turbine combined-
these plants, it is not cost-efficient to apply conventional CCS tech- cycle power plant with precombustion CO2 capture. Energy 2005;30:5–39.
nologies that require large investments in the infrastructure. More http://dx.doi.org/10.1016/j.energy.2004.05.029.
research and optimization is required in order to achieve high con- [20] Corradetti A, Desideri U. Analysis of gas–steam combined cycles with natural
gas reforming and CO2 capture. J Eng Gas Turb Power 2005;127:545–52.
version rates and product carbon with suitable quality in TDM pro- http://dx.doi.org/10.1115/1.1850941.
cess. The price of the product carbon has a crucial effect on the [21] Bolland O, Undrum H. A novel methodology for comparing CO2 capture
feasibility, and therefore more research on the utilization of ther- options for natural gas–fired combined cycle plants. Adv Environ Res
2003;7:901–11. http://dx.doi.org/10.1016/S1093-0191(02)00085-0.
mal black is required. [22] Lozza G, Chiesa P. Natural gas decarbonization to reduce CO2 emissions from
combined cycles – Part I: Partial oxidation. J Eng Gas Turb Power
Acknowledgments 2000;124:82–8. http://dx.doi.org/10.1115/1.1395581.
[23] Lozza G, Chiesa P. Natural gas decarbonization to reduce CO2 emissions from
combined cycles – Part II: Steam–methane reforming. J Eng Gas Turb Power
This work was carried out in the Carbon Capture and Storage 2000;124:89–95. http://dx.doi.org/10.1115/1.1395582.
Program (CCSP) research program coordinated by CLIC Innovation [24] Manzolini G, Macchi E, Gazzani M. CO2 capture in natural gas combined cycle
with SEWGS. Part B: Economic assessment. Int J Greenhouse Gas Control
Ltd. with funding from the Finnish Funding Agency for Technology 2013;12:502–9. http://dx.doi.org/10.1016/j.ijggc.2012.06.021.
and Innovation, Tekes. Furthermore, the financial support received [25] Triphob N, Wongsakulphasatch S, Kiatkittipong W, Charinpanitkul T,
from ÅF-Consult Ltd., Fortum Plc, Gasum Ltd., Helen Ltd., Neste Plc, Praserthdam P, Assabumrungrat S. Integrated methane decomposition and
solid oxide fuel cell for efficient electrical power generation and carbon
and Gasum Gas Fund is gratefully acknowledged. John Shepherd is
capture. Chem Eng Res Des 2012;90:2223–34. http://dx.doi.org/10.1016/j.
acknowledged for proof reading this article. cherd.2012.05.014.
[26] Kirk-Othmer, editor. Encyclopedia of Chemical Technology. New Jersey,
References USA: John Wiley & Sons Inc; 2007 [chapter Carbon Black].
[27] Sid Richardson Carbon and Energy Co. Carbon black pricing. <http://www.
sidrich.com/products-and-pricing/pricing/carbon-black-pricing/>; 2014
[1] International Energy Agency (IEA). World Energy Outlook 2014, 2014. <http://
[accessed 03-07-15].
www.worldenergyoutlook.org/publications/weo-2014/>.
[28] Hosokawa M, Nogi K, Naito M, Yokoyama T, editors. Nanoparticle technology
[2] European Commission. 2030 Energy strategy. <http://ec.europa.eu/energy/
handbook. Elsevier; 2008.
en/topics/energy-strategy/2030-energy-strategy>; 2015 [accessed 11.05.15].
[29] International Agency for Research on Cancer. IARC monographs on the
[3] International Energy Agency (IEA). Technology Roadmap – Carbon Capture and
evaluation of carcinogenic risks to humans, Volume 93: carbon black,
Storage, 2013. <http://www.iea.org/publications/free-publications/
titanium dioxide, and talc. Tech. rep., International Agency for Research on
publication/technology-roadmap-carbon-capture-and-storage-2013.html>.
Cancer; 2010.
[4] Rokach J. Should we pursue carbon capture for natural gas? Electr J
[30] International Carbon Black Association. Carbon black user’s guide: safety,
2012;25:64–7. http://dx.doi.org/10.1016/j.tej.2012.06.001.
health, & environmental information. 2004.
[5] Muradov N. CO2-free production of hydrogen by catalytic pyrolysis of
[31] Environment Canada. Screening assessment for the challenge: carbon black;
hydrocarbon fuel. Energy Fuels 1998;12:41–8. http://dx.doi.org/10.1021/
2013. <https://www.ec.gc.ca/ese-ees/default.asp?lang=En&n=2CF>.
ef9701145.
[32] National Institute of Standards and Technology. NIST-JANAF, Thermochemical
[6] Amin A, Croiset E, Epling W. Review of methane catalytic cracking for
tables. <http://kinetics.nist.gov/janaf/> [accessed 19-12-14].
hydrogen production. Int J Hydrogen Energy 2011;36:2904–35. http://dx.doi.
[33] Kim K, Hong S, Lee S, Ju W. Continuous synthesis of nanostructured
org/10.1016/j.ijhydene.2010.11.035.
sheetlike carbons by thermal plasma decomposition of methane. IEEE
[7] Li Y, Li D, Wang G. Methane decomposition to COx-free hydrogen and nano-
Trans Plasma Sci 2007;35:434–43. http://dx.doi.org/10.1109/TPS.2007.
carbon material on group 8–10 base metal catalysts: a review. Catal Today
892556.
2011;162:1–48. http://dx.doi.org/10.1016/j.cattod.2010.12.042.
[34] Winter F, Schratzer B. Applications of fluidized bed technology in processes
[8] Abbas H, Wan Daud W. Hydrogen production by methane decomposition: a
other than combustion and gasification. UK: Woodhead Publishing; 2013.
review. Int J Hydrogen Energy 2010;35:1160–90. http://dx.doi.org/10.1016/ j.
http://dx.doi.org/10.1533/9780857098801.5.1005. p. 1005–33. [chapter:
ijhydene.2009.11.036.
Fluidized bed technologies for near-zero emission combustion and
[9] Ashik U, Wan Daud W, Abbas H. Production of greenhouse gas free hydrogen
gasification].
by thermocatalytic decomposition of methane – a review. Renew Sustain
[35] International Energy Agency (IEA). Improvement in power generation with
Energy Rev 2015;44:221–56. http://dx.doi.org/10.1016/j.rser.2014.12.025.
post combustion capture of CO2. Tech. rep., Report PH4/33, International
[10] Muradov N, Veziroğlu T. From hydrocarbon to hydrogen–carbon to hydrogen
Energy Agency; 2004.
economy. Int J Hydrogen Energy 2005;30:225–37. http://dx.doi.org/10.1016/ j.
[36] Stavseth E. The Nordic/Baltic power market. <http://www.nasdaqomx.com/
ijhydene.2004.03.033.
digitalAssets/86/86050_npspotjune112013.pdf>; June 11th 2013 [accessed
[11] Rackley S. Carbon capture and storage. USA: Elsevier; 2010.
31-08-15].
[12] Herzog H, Smekens K. IPCC special report on carbon dioxide capture and
[37] International Energy Agency (IEA), Natural gas. <http://www.iea.org/
storage. Cambridge University Press; 2005. p. 105–78. [chapter Capture of
aboutus/faqs/gas/>; 2014 [accessed 29-12-14].
CO2].
[38] Finnish Gas Association. Maakaasukäsikirja (Handbook of Natural Gas); 2014.
[13] Blackford J, Hattam C, Widdicombe S. CO2 leakage from geological storage
[in Finnish].
facilities: environmental, societal, and economic impacts, monitoring and
12 T. Keipi et al. / Energy Conversion and Management 110 (2016) 1–12

[39] Pinilla J, Suelves I, Lázaro M, Moliner R, Palacios J. Parametric study of the [41] International Energy Agency (IEA). IEA Greenhouse gas R&D programme,
decomposition of methane using a NiCu/Al2O3 catalyst in a fluidized bed improvement in power generation with post-combustion capture of CO2. Tech.
reactor. Int J Hydrogen Energy 2010;35:9801–9. http://dx.doi.org/10.1016/j. rep., Report PH4/33, International Energy Agency; 2004.
ijhydene.2009.10.008. [42] Herzog H, Smekens K. IPCC special report on carbon dioxide capture and
[40] Tranier J-P, Dubettier R, Perrin N. Air separation unit for oxy-coal combustion storage. Cambridge University Press; 2005. p. 339–62. [chapter: Cost and
systems. In: IEAGHG 1st international oxyfuel combustion conference, economic potential].
Cottbus; 2009.