Energy 278 (2023) 127868

Contents lists available at ScienceDirect

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

Process analysis and comparative assessment of advanced thermochemical

pathways for e-kerosene production
Konstantinos Atsonios a, Jun Li b, Vassilis J. Inglezakis b, *
a
Centre for Research & Technology Hellas, Chemical Process and Energy Resources Institute, 57001, Thessaloniki, Greece
b
Department of Chemical and Process Engineering, University of Strathclyde, 75 Montrose St., Glasgow, G1 1XJ, UK

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

Handling Editor: Petar Sabev Varbanov Climate change and energy supply are major driving forces for the promotion of sustainable fuels production. In
the aviation sector, due to inherent difficulties to adopt electrification methods for long distance flights, the
Keywords: successful implementation of sustainable aviation fuel (SAF) is crucial for the achievement of greenhouse gas
Synthetic kerosene emissions mitigation strategies. This study presents four different pathways for the valorization of captured CO2
CO2 to jet fuel
into synthetic kerosene using hydrogen and demonstrates the comparative assessment in terms of various
Fischer-tropsch
technical and aspects such as hydrogen consumption, thermal energetic efficiency and produced e-kerosene
Methanol oligomerization
Aspen plus quality. Two pathways are based on Fischer-Tropsch synthesis, a low-temperature CO conversion though reverse
water-gas shift reaction and a high-temperature direct CO2 conversion, while the other two are based on the
valorization and upgrading of light alcohols (methanol and ethanol) derived from CO2 hydrogenation. The
process models were developed in Aspen Plus. Simulation results revealed that the low-temperature CO con­
version pathway is the most efficient to maximize jet fuel yield with the lower energy and exergy losses.
Indicatively for that case, 90.7% of the initial carbon is utilized for kerosene fraction synthesis, the overall
thermal efficiency is 70.9% whereas the plant exergetic efficiency is 72.6%. The basic properties of the produced
e-kerosene for all pathways meet with the required Jet-A1 specifications or are close to them.

1. Introduction kerosene with aromatics (FT-SKA), Hydroprocessed Esters and Fatty

Acids (HEFA), synthesized iso-paraffins (SIP), Alcohol to Jet (ATJ),
Low-carbon energy transition is a critical for a successful climate catalytic hydrothermolysis (CHJ) and synthesized paraffinic kerosene
change mitigation [1]. Global warming and the climate change as a from hydrocarbon-hydroprocessed esters and fatty acids (HC-HEFA-
consequence of it have been recognized as of the most significant con­ SPK), using feedstock from biological (advanced biofuels) or no-
cerns that put the humanity and ecosystems survival on earth into biological (e-kerosene) origin (see Fig. 1). Most of these routes pro­
danger. To avoid that, zero-emission technologies have to be fostered duce paraffinic kerosene comprising blends of acyclic normal and
and rapidly deployed at commercial level in all sectors. Transportation is branched alkanes [4]. Especially, for e-kerosene, renewable or low
the third sector after energy and industry with the largest greenhouse carbon electricity is needed for the production of hydrogen which is one
gas (GHG) emissions [2] and the only one that increased its emissions of the main energy sources for the jet fuel like synthesis, this type of
the last 30 years in EU [3]. Unlike the road and railway transports that sustainable fuels are called electrofuels or e-fuels. Currently, the vast
can eliminate their emissions through electrification, the aviation sector majority of SAF is produced through the HEFA from fats, oils and greases
must rely only on the development of sustainable aviation fuels (SAF) as being Neste and World Energy the world leading companies for HEFA
this is the only way for long distance flights to be carbon neutral since production at commercial-scale. Moreover, Gevo and LanzaJet have
there is no alternative technology at high attitudes than the aircraft developed ATJ pathways for the production of renewable jet fuel from
turbo engine. isobutanol and ethanol, respectively. The main bottleneck for these
The main types of SAFs, approved by ASTM (ASTM D7566 -20) as bio-based technologies is the high feedstock cost and low feedstock
blending components for conventional jet fuel, are the Fischer-Tropsch availability for large scale applications [5]. For that reason, the alter­
synthetic paraffinic kerosene (FT-SPK), Fischer-Tropsch synthetic native way for SAF production using CO2 and renewable H2 as feedstock

* Corresponding author.
E-mail address: vasileios.inglezakis@strath.ac.uk (V.J. Inglezakis).

https://doi.org/10.1016/j.energy.2023.127868
Received 18 November 2022; Received in revised form 7 April 2023; Accepted 16 May 2023
Available online 23 May 2023
0360-5442/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
K. Atsonios et al. Energy 278 (2023) 127868

attracts the interest the last years.

There are several strategies to convert CO2 and H2 into liquid fuels
mainly composed of hydrocarbons adopting either, biochemical, cata­
lytic, electrochemical techniques or a combination of them. Due to the
thermodynamic stability and chemical inertia of CO2, its hydrogenation
usually promotes the synthesis of short-chain compounds such as CO,
methane, methanol, acetic acid and C2–C4 olefins, most of experimental
studies are focused on the optimization of the synthesis process of these
compounds. Hence, these options for synthetic kerosene synthesis are
based on multi-stage pathways where CO2 is firstly converted into an
intermediate compound which in turn becomes the feedstock for the
liquid hydrocarbons synthesis, in one or more process steps (Fig. 2).
Nevertheless, there are studies that aim to produce liquid fuel directly
from CO2 with considerable selectivity to jet fuel fraction [6]. Currently, Fig. 2. Options for synthetic jet fuel synthesis from CO2.
the most common pathway is to reduce CO2 to CO either catalytically via
the reverse water gas shift reaction (rWGS) [7], or electrochemically via the final products with a model compound or a mix of them such as
a co-electrolysis pathway transforming water and CO2 into syngas C12–C14 paraffins [16]. Taking, however, into account the strict stan­
H2/CO [8]. The first pathway consists of established technologies and dards for jet fuel specifications the appropriateness of each technology
has already been deployed at industrial scale [9]. Moreover, methanol for the production of SAF is strongly affected by the target to meet
attracts also the interest of the used feedstock for jet fuel production: standards or the additional actions to be taken. The aim of this research
recently ExxonMobil announces a novel technology based on that route is to study and simulate catalytically-based pathways specifically
[10]. designed to maximize jet fuel production from CO2, the main properties
Several studies that are dedicated to renewable liquid fuels synthesis of which fall under the ASTM specifications for aviation fuels; a con­
using syngas and/or CO2, as feedstock can be found in the literature. ventional low temperature Fischer-Tropsch, a novel, high temperature
Most of these studies are not specifically addressing the synthesis of Fischer-Tropsch without a reverse water gas shift reactor, a novel
kerosene but the designs presented are suited for the production of methanol-based and a novel ethanol-based pathway. The four pathways
several fuels such as diesel and gasoline. Sudiro and Bertucco [11] and are compared in terms of energy efficiency and process effectiveness
Navas-Anguita et al. [12] present a process based on FT for the pro­ while the produced jet fuels can be regarded as potential “drop-in” ac­
duction of gasoline and diesel. Ruokonen et al. [13] present a process cording to the ASTM specifications.
based on methanol-to-olefins and Mobil’s Olefins to Gasoline and
Distillate for the production of diesel (45%), kerosene (27%) and gaso­ 2. Approach and methodology
line (17%). Konig et al. [14] design is based on FT and produces kero­
sene (43.9%), gasoline (31.2%) and diesel (24.9%). Petersen et al. [15] In this paper, four thermocatalytic pathways are developed and the
present a process based on ethanol and produce kerosene (59%), gaso­ respective flowsheets are designed in such way that the jet fuel yield is
line (37%) and diesel (4%). However, when the target is to produce maximized and the rest fuel fractions are eliminated. Three pathways
sustainable aviation fuel, the gasoline and diesel are by-products with are based on CO2 conversion into intermediate compound such CO and
low impact value since the roadmap for zero-emissions in road transport methanol whereas one is based on the conversion of CO2 to medium
is mainly based on vehicles electrification [10]. Another issue that is chain hydrocarbons in one step. Since the focus of this study is to
observed in most of the simulation studies for renewable jet fuel pro­ identify the most effective transformation of CO2 into jet fuel, the way
duction is the restricted information that is given about the final prod­ that hydrogen and pure CO2 stream are produced or supplied in the
ucts specifications. The general approach that is adopted is to represent plant are not taken into account. The origin of CO2 plays decisive role in

Fig. 1. Available pathways for SAF production.

2
K. Atsonios et al. Energy 278 (2023) 127868

the GHG footprint of the final products. The CO2 either can come from considering the available energy of the main inlet and outlet streams. A
the atmosphere (Direct Air Capture) or can have biogenic (from bio­ term of exergy efficiency is introduced that is defined as the ratio of total
energy or biorefinery plant) or fossil carbon origin (industrial flue exergy output of the useful streams (i.e. liquid products) to total exergy
gases). In any case, the CO2 stream that will be used for the synthetic input:
kerosene production must meet certain specifications in terms of gas
˙
Ejet˙ fuel + EDiesel+gasoline
impurities (i.e. nitrogen oxides, oxygen gas, sulfur oxides and hydro­ ηEx = eq 6
˙ ˙ ˙
carbons) concentration to avoid the reactors catalysts poisoning. In all EH2,in + ECO2,in + Pin + EQ,in
the examined cases, the CO2 flow rate is set equal to 100 kmol/h
whereas the hydrogen consumption is left as a dependent variable. This where Ejet˙ fuel and EDiesel+gasoline
˙ (in MW) are the chemical exergies of the
CO2 flow rate corresponds to an annual CO2 valorization of 40 kt/y and final liquid fuels calculated according to the following equation [18]:
an electrolyser capacity of 30–50MWe. This size selection has been done (∑ ∑ )
˙ fuel = N • εch = N •
Exliquid xi εo,i + RT xi lnxi eq 7
arbitrarily and although it corresponds to a rather small industrial scale
unit the overall process is definitely scalable and can be applied at larger
scales. The proper size determination is a matter of the techno-economic where N is the molar flow in kmol/s and xi the molar fraction of each
assessment, which is not part of that study. As far as the process analysis component i. The reference conditions used are the standard environ­
in concerned, this is mainly based on the process simulation modeling of mental conditions (To = 298.15 K, po = 1.013 bar) and values of ε0,i for
the examined flowsheets performed with Aspen Plus. The reactors each component are obtained from Ref. [19].
modeling was based either on the assumption of the chemical equilib­ Exergy of power equals power itself and exergy of a heat stream Q
rium or on data from published experimental data. (EQ,i
˙ ) is evaluated with the help of the Carnot factor:
( )
˙ = Qi • 1 − T0 eq 8
2.1. Definition of performance indicators EQ,i
Ti

The following metrics are introduced for the performance evaluation where Ti is the temperature at which Qi is available and i = in or out.
of each case. The overall exergy that is destructed according to the second ther­
The Carbon Utilization (CU) determines the portion of initial carbon modynamic law (Eloss ˙ ) is calculated from the exergy balance of the
that exist in the form of CO2 that is finally found in the synthetic kero­ overall system (see Fig. 3):
sene stream. ( ) ( )
˙ = EH2,in
Eloss ˙ + ECO2,in ˙ − Ejet˙ fuel + EDiesel+gasoline
˙ + Pin + EQ,in ˙ ˙ heat
+ EQ,waste
ṅC,e− jet
CU = eq 1 eq 9
ṅC,CO2in
The term ηwasted expresses the amount of exergy that is wasted in the
where ṅC,e− jet the carbon molar flow (kmol/h) at the produced jet fuel form of unexploited heat to the total exergy input:
stream and ṅC,CO2in the C flow (kmol/h) at the CO2 inlet stream.
˙ heat
EQ,waste
The Energetic Jet Fuel Efficiency (EJFE) measures the thermal effi­ ηwasted = eq 10
˙ + Pin + EQ,in
˙ + ECO2,in ˙
ciency of the process and the ratio of heat input of the produced SAF EH2,in
(MW on a LHV base) to the total used hydrogen heat input plus the
Whereas the term ηloss expresses the ratio of irreversibilities (Eloss) of the
external heat (Qext) if necessary:
overall process to the total exergy input:
ṁe− jet • LHVe− jet
EJFE = eq 2 ˙
Eloss
ṁH2,in • LHVH2 + Qext ηloss = eq 11
˙ ˙ ˙
EH2,in + ECO2,in + Pin + EQ,in
In order to take into account all the external heating and power demands
for the effective and stable operation of the proposed systems the term of 3. Pathways description
overall plant efficiency (ηtot) is defined as:
ṁe− jet • LHVe− jet 3.1. FT based pathways
ηtot = eq 3
ṁH2,in • LHVH2 + Qext + Pext
In this work, as is the common practice in plant level simulation
For the cases where refrigeration loads are required, the respective studies, only the main FT reactions were considered. This approach is
loads are converted into electricity, assuming a constant coefficient of followed in conceptual design and simulation studies to avoid unnec­
performance (COP) of 3.14 [17]. essary complexity which can lead to convergence problems, as the
The e-kerosene yield (Ye-jet) defines the amount of produced syn­ production of other organic groups such as alcohols, aldehydes and acids
thetic aviation fuel mass flow (ṁe− jet ) to the inlet CO2/H2 feed gas mass have negligible impact on the overall process.
flow (ṁH2,in + ṁCO2,in ):
3.1.1. High-temperature Fischer-Tropsch synthesis without rWGS (CO2FT)
ṁe− jet
Ye− jet = eq 4 The available studies on high-temperature Fischer-Tropsch (FT)
ṁH2,in + ṁCO2,in
synthesis without separate reverse water-gas shift (rWGS) reactor are
The e-kerosene fraction (fe-jet) defines the portion of the produced rare and are limited to laboratory experimental research. This process is
synthetic aviation fuel mass flow among the mass flow of other side called non-methanol mediated CO2 hydrogenation [20] and combines
products (i.e. diesel and gasoline):
ṁe− jet
fe− jet = eq 5
ṁproducts,tot

2.2. Exergy analysis

The methodological tool of exergy analysis is employed in order to

have a more qualitative assessment of the examined schemes by Fig. 3. Exergy balance.

3
K. Atsonios et al. Energy 278 (2023) 127868

FT and rWGS reactions in a single reacrtor by use of Fe catalysts. The As seen in Table 1, all the reaction processes take place under high
reactions considered are [21]: pressure and temperature. All reactors apart from ATR produce excess
heat because of the exothermic nature of the reactions that are carried
CO2 + H2 ⟺ CO + H2 O (12)
out. The excess heat is utilized either for the effective preheating other
streams or for low pressure (LP) steam generation. The same strategy is
nCO + (2n + 1)H2 ⟺ Cn H2n+2 + nH2 O (13)
also followed in the remaining cases.
nCO + 2nH2 ⟺ Cn H2n + nH2 O (14)
3.1.2. Low-temperature Fischer-Tropsch synthesis (LTFT)
To reach satisfactory CO2 conversions and C5–C15 yields it is neces­ In the examined pathway, CO2 is firstly transformed into CO in the
sary to use high temperature, pressure and excess H2 in combination to a rWGS reactor:
catalyst that suppresses CH4 production [22]. In the present paper
experimental data of a Fe-based catalyst containing K as a promoter CO2 + H2 ⟺ CO + H2 O (18)
were used [23]. The transformation of CO2 to CO is thermodynamically favored by
The process flow diagram is shown in Fig. 4. The FT reactor product high temperature because it is reversible and endothermic and is pres­
is cooled and sent to adsorption units for the removal of CO2 and H2 and sure independent. The rWGS is always accompanied by undesired CO2
then is decanted to separate water, light gases and the hydrocarbons. methanation, also called the Sabatier reaction, which is exothermic,
The light gases are recompressed and sent to an oligomerization reactor favored by lower temperature and high pressure [26]:
where the light olefins (C2–C9) are dimerized. The products are sent to a
flash drum where the gases and olefins are separated. The gases are sent CO2 + 4H2 ⟺ CH4 + 2H2 O (19)
to a H2 adsorption unit, a flash drum to separate the remaining olefins Hence to maximize the CO production low pressures and high tem­
and then to an autothermal reactor (ATR). Reforming of light gases, peratures are required. The process flow diagram is shown in Fig. 5. The
mainly CH4, produces syngas and increases overall conversion [24]. The rWGS reactor products are sent to flash where water is separated and
main reactions are [25]: then to an adsorber for the removal of CO2. The syngas compressed and
CH4 + 3/2 O2 ⟺ CO + H2 O (15) then is transformed into hydrocarbons by low-temperature Fischer-
Tropsch (FT) synthesis (LTFT). When Co is used as catalyst in LTFT, it is
CH4 + H2 O ⟺ CO + 3H2 (16) commonly assumed that only paraffins are produced and the main re­
action is accompanied by the water-gas shift reaction (WGS) [27–29]:
CO + H2 O ⟺ CO2 + H2 (17)
nCO + (2n + 1)H2 ⟺ Cn H2n+2 + nH2 O (20)
Besides CH4 all light hydrocarbons are converted to syngas. The
excess H2O is separated in a flash drum and the syngas is mixed with CO + H2 O ⟺ CO2 + H2 (21)
fresh CO2 and H2. The hydrocarbons separated at the decanter and the The FT reactor product is cooled and decanted to separate water,
olefins produced in the oligomerization reactor are sent to a hydro­
light gases and the hydrocarbons. The light gases are recompressed and
treater where olefins are transformed to paraffins and heavy paraffins sent to a H2 adsorption unit and then to an autothermal reforming
(C24+) are cracked to smaller paraffins. The products are sent to
reactor (ATR). The excess H2O is separated in a flash drum and the
adsorption units for the removal of CO2 and H2 and then to the first syngas is mixed with fresh CO2 and H2. The hydrocarbons are sent to a
distillation column where kerosene is taken at the bottom and the lighter hydrocracker where the heavier molecules are broken to lighter mole­
hydrocarbons at the top. The top product is sent to a second distillation cules [28]:
column where gasoline is taken at the bottom and light tail gas at the
top. The latter is sent to the ATR for reforming. C2n H4n+2 + H2 ⟺ 2Cn H2n+2 (C24 and C2 8) (22)

Fig. 4. Process flowsheet of the CO2 to Jet fuel pathway via CO2FT (direct CO2 into FT liquids).

4
K. Atsonios et al. Energy 278 (2023) 127868

Table 1 at the bottom.

Flash separators and distillation columns process specifications. It should be underlined that in both FT-based pathways, it is assumed
CO2FT that hydrocarbons (CxHy) is the only organic compounds group that are
found in the FT reactor outlet. Since some experimental studies report
Flash T = 30 ◦ C, p = 25 bar Distil Ntot = 10, Nfeed = 5, BR = 1.16, RR =
1 1 0.5 the formation of alcohols and acids, there are certain methodologies like
Flash T = 40 ◦ C, Δp = 0 bar Distil Ntot = 10, Nfeed = 5, BR = 4.08, RR = the use of adsorbents [30]or tripper-sidestream decanter [31] that can
2 2 3 effectively remove them from the FT crude stream.
Flash T = 40 ◦ C, p = 5 bar
3
LTFT
3.2. Light alcohols-based pathways
Flash T = 30 ◦ C, p = 5 bar Distil Ntot = 10, Nfeed = 5, BR = 3, RR =
1 1 0.735
Flash T = 30 ◦ C, Δp = 0 bar Distil Ntot = 20, Nfeed = 10, BR = 40.99, RR 3.2.1. Methanol-based pathway
2 2 =2 In this pathway, CO2 is firstly transformed into methanol which is
Flash T = 30 ◦ C, p = 5 bar subsequently converted into medium/long chain hydrocarbons through
3
MeOH based
a series of catalytic processes. The methanol synthesis through CO2
Flash Q = 0, Δp = 0.15 bar Distil Ntot = 35, Nfeed = 30, BR = 0.54, RR hydrogenation is carried out according to the reactions:
1 1 = 1.22
Flash Q = 0, Δp = 0 bar Distil Ntot = 40, Nfeed = 12, BR = 2.94, RR CO + 2H2 ⟺ CH3 OH (24)
2 2 = 0.89
Flash T = 28 ◦ C, Δp = 0 bar Distil Ntot = 30, Nfeed = 10, BR = 4.11, RR CO2 + H2 ⟺ CO + H2 O (25)
3 3 = 0.36
Flash Q = 0, Δp = 0 bar Distil Ntot = 10, Nfeed = 3, BR = 28.3, RR = CO2 + 3H2 ⟺ CH3 OH + H2 O (26)
4 4 1.5
Flash T = 30 ◦ C, Δp = 4 bar Distil Ntot = 10, Nfeed = 11, BR = 0.56, RR The process flow diagram of a typical methanol synthesis unit from
5 5 = 1.14 CO2 is shown in Fig. 6. The unit consists of the methanol synthesis, gas
Flash Q = 0, Δp = 0 bar
6
separation and product purification. The inlet gas is heated up to a
Flash T = 28 ◦ C, Δp = 0 bar certain temperature level. The required heat for crude methanol heating
7 and separation is obtained from the gas outlet cooling and the excess
EtOH based heat from the methanol synthesis reactor (exothermic process). In order
Flash T = − 18 ◦ C, Δp = 0 Distil Ntot = 20, Nfeed = 8, BR = 0.60, RR =
to achieve high purity levels in the methanol product (>99.2%), the
1 bar 1 0.31
Flash Q = 0, Δp = 0 bar Distil Ntot = 35, Nfeed = 14, BR = 1.37, RR liquid stream after the first flash separator is throttled down to atmo­
2 2 = 3.14 spheric pressure. A small portion (0.5%) of the recycling gas is extracted
Distil Ntot = 3, Nfeed = 1, BR = 0.45, RR = 0 as purge gas in order to avoid by-products accumulation such as hy­
3 drocarbons, inert gas etc.
Distil Ntot = 30, Nfeed = 14, BR = 6.62, RR
The rationale behind that is based on the separate handling of each
4 = 6.33
Distil Ntot = 70, Nfeed = 41, BR = 4.55, RR light olefin as it is produced at the methanol-to-olefins (MTO) reactor
5 = 1.48 aiming to maximize the yield of the desired hydrocarbons within the
Distil Ntot = 5, Nfeed = 3, BR = 5.79, RR = range of C12–C14. The main units of that pathway is a) the MTO unit, b)
6 0.67
the olefins oligomerization unit c) the oligomers hydrotreatment and d)
Ntot: number of stages, Nfeed: stage number of feed stream, BR: boilup ratio, RR: the ATR unit. The produced methanol enters the MTO reactor and the
reflux ratio. produced light olefins/paraffins are recovered according to the UOP/
Hydro MTO pathway [32]. The light gases are recovered and sent to ATR
for reforming. The olefins (i.e. ethylene, propylene and butylene) are
C4n H8n+2 + 3H2 ⟺ 4Cn H2n+2 (C32 − C64 ) (23)
sent for oligomerization.
Following hydrocracking the hydrocarbons are sent to a H2 adsorp­ The recovered ethylene undergoes dimerization for the production of
tion unit and then to the first distillation column where light hydro­ 1-butene. The reaction parameters for this process were obtained from
carbons are taken at the top and sent to the ATR and heavy Ref. [33]. The recovered 1-butene mixes with the product stream from
hydrocarbons at the bottom. The heavy hydrocarbons are sent to a ethylene dimerization and undergo oligomerization (mainly dimeriza­
second distillation column where kerosene is taken at the top and diesel tion and trimerization). The process parameters and the associated

Fig. 5. Process flowsheet of the CO2 to Jet fuel pathway via LTFT (through rWGS).

5
K. Atsonios et al. Energy 278 (2023) 127868

Fig. 6. Process flowsheet of the CO2 to Jet fuel pathway via methanol synthesis.

reactions were obtained from Ref. [34]. The recovered propene un­ The produced DME is recovered after a subsequent distillation col­
dergoes oligomerization for the production of various oligomers (di­ umn and a flash separator whereas the unconverted methanol is sepa­
mers, trimers, tetramers, etc.). The process parameters and the rated from water in a second column. Part of the unreacted gas is split
associated reactions were obtained from Ref. [35]. All the produced and mixed with DME that is needed for the Ethanol synthesis according
oligomers mix and send to the hydrotreatment unit for hydrogenation to the following set of reactions.
and their conversion into alkanes. The same reactor that has been
CH3 O CH3 + CO ⟺ CH3 COOCH3 (30)
employed in the MTO/MOGD pathway was used, enriched with the
respective hydrogenation reactions of the olefins that are not considered
CH3 COOCH3 + 2H2 ⟺ CH3 OH + CH3 CH2 OH (31)
in the former scenario (i.e. same conversion rate = 90%). The Auto­
thermal Reformer (ATR) reactor is pressurized, oxygen blown, in order The first is the DME carbonylation where methyl acetate (MA) is
to avoid some of the compression duty of the recycling gas and to exploit formed in the presence of H-Mordenite (H-MOR) zeolite, whereas the
the oxygen that is produced as the electrolyser together with the second one is the produced ester hydrogenation over the Cu/ZnO
hydrogen. catalyst. The reactions take place in a dual bed reactor sequentially, at
15 bar and 220 ◦ C. Moreover, CO2 and ethyl acetate (EA) are also by-
3.2.2. Ethanol-based pathway products [37]. After the products separation, DME, MA and EA reenter
Fig. 7 present the process flow diagram of the fourth pathway. to the reactor, whereas the CH3OH and CO2 are sent at the DME syn­
The main idea of this pathway is to transform CO2 into ethanol and thesis unit. To accomplish that, a series of two distillation columns and a
the latter to be the basis for the medium/long chain hydrocarbons based flash separation is employed as seen in Fig. 7. This way of producing
on an alcohol-to-jet (ATJ) scheme. The fresh H2/CO2 stream after synthetic ethanol from CO2 has been introduced earlier in Ref. [38] and
compression is mixed with the recycling streams (internal gas loop, is adopted it again in the present study as the direct catalytic conversion
external gas loop coming from the Ethanol Synthesis unit and methanol) of CO2 into ethanol does not yield high conversion and selectivity rates
and undergoes dimethyl ether (DME) synthesis: together, according to the recent relevant studies [39].
The third section of the process consists of the ethanol upgrade into
CO2 + H2 ⟺ CO + H2 O (27)
medium/long chain hydrocarbons based on four consecutive catalytic
reactions: ethanol condensation, n-butanol dehydration, light olefins
CO2 + 3H2 ⟺ CH3 OH + H2 O (28)
oligomerization, and oligomers hydrogenation [40]. In the first step,
2CH3 OH ⟺ CH3 O CH3 + H2 O (29) ethanol is converted into n-butanol according to the following Guerbet
reaction:
In order to produce DME in one step, bifunctional catalysts should be
applied, where the first reaction is catalyzed by an acidic catalyst such 2C2 H5 OH → C4 H9 OH + H2 O (32)
HZSM-5 and the rest two by a methanol synthesis catalyst such as Cu/ High ethanol conversion rates and selectivity in n-butanol can be
ZnO/Al2O3 [36]. achieved if a catalyst such as RuCl2 is applied [41]. Apart from butanol,

6
K. Atsonios et al. Energy 278 (2023) 127868

Fig. 7. Process flowsheet of the CO2 to Jet fuel pathway via DME/ethanol synthesis.

hexanol and 2-ethyl hexanol are also produced as by-products. In the

second reaction step, the produced alcohols undergo dehydration and Table 2
Utilities specifications.
the respective alkenes are produced which in turn are converted into
oligomers in the third reactor. The last catalytic process is the hydro­ Type of utility Temperature (oC) Use
treatment of all the produced alkenes for the production of paraffinic Very High T source 1000 Hot
hydrocarbons that will be the basic compounds of the final fuel products. Fired Heat 400 Hot
As depicted in Fig. 7, each catalytic step is followed by the necessary HP/IP/LP Steam 250/175/125 Both Hot & Cold
Air ventilation 30 Both Hot & Cold
separation step for the removal of impurities or the recovery of the Cooling Water 20 Cold
desired products. Refrigerant 1/2/3 − 25/-40/-65 Cold

4. Process model methodology

simulation. In the present paper experimental data of a Fe-based
The flowsheets development and the process simulations were per­ (10Fe0⋅8 K) catalyst were used with an overall CO2 conversion of
formed in Aspen Plus. The following subsections present the method­ 0.417 [23]. For this catalyst and the specific operational conditions
ology for the main reactors modeling in each case. Regarding the rest (300 ◦ C, 25 bar, H2/CO2 = 3) the fractional selectivity of light hydro­
components, a strategy of using the same specifications: the compressors carbons and CO were estimated based on the experimental data: CH4 =
isentropic efficiency is set at 85%, the pumps efficiency at 70% and the 0.103, C2H4 = 0.072, C2H6 = 0.0207, C3H6 = 0.072, C3H8 = 0.0207,
H2 recovery efficiency at PSA unit is 99%, retrieving 100% pure C4H8 = 0.072, C4H10 = 0.0207 and CO = 0.06. The distribution of the
hydrogen. The process specifications of the flash separators and distil­ C4+ paraffins in the FT reactor is calculated through the
lation columns are summarized in Table 1. Anderson-Schulz-Flory (ASF) distribution which define the stoichiom­
Moreover, the heat demands or the excess heat removal are fulfilled etry of the overall reaction or the selectivity of the products. The chain
by external utilities, the specification of which are uniform for all the growth probability factor (α) for CO2FT is between 0.57 and 0.79
examined cases. Table 2 presents their basic characteristics. The re­ depending on the carbon number and the value of 0.72 was used in this
frigerants are considered as a separate utility but the required power for paper [42]. The distribution of C4+ olefins was based on this of the
their generation is also taken into account in the energy analysis. respective paraffins and the following olefin/paraffin ratio [43]:
18.9/6.8 for C5–C11, 14.8/3.8 for C12–C20 and 1.8/0.5 for C24 and C28.
Stoichiometric amount of H2 was fed to the hydrotreater. Combustion in
4.1. High-temperature Fischer-Tropsch synthesis (CO2FT) an ATR is sub-stoichiometric with an overall oxygen to hydrocarbon
ratio of 0.55–0.6 [25]. Reforming and WGS reactions take place in the
Table 3 presents the models parameters that are used for the reactors

7
K. Atsonios et al. Energy 278 (2023) 127868

Table 3 catalytic zone of the reactor by using steam with H2O/C ratios ranging
Reactors models specifications for the CO2FT pathway. from 0.2 to 3.5 [44]. In this study, the estimated optimum O2/C and
Reactor name Reactor Process Associated Fractional H2O/C molar ratios were 0.7 and 4, respectively.
model specifications reactions conversion

Fischer-Tropsch RSTOIC 300 ◦ C/25 bar CO2 + H2 ⟺ 0.417 (CO2) 4.2. Low-temperature Fischer-Tropsch synthesis (LTFT)
H2/CO2 = 3 CO + H2O nCO
+ (2n+1)H2 ⟺ Table 4 presents the models parameters that are used for the reactors
CnH2n+2 + nH2O
nCO + 2nH2 ⟺
simulation. The H2/CO molar ratio at the FT reactor inlet was kept equal
CnH2n + nH2O to 2 in order to maximize the selectivity for hydrocarbons within the
Oligomerization RSTOIC 350 ◦ C/40 bar 2C2H4 → C4H8 1 kerosene chain length [45]. The overall CO conversion was set to 0.85.
2C3H6 → C6H12 1 The distribution of the C4+ hydrocarbons in the FT reactor is calculated
2C4H8 → C8H16 1
through the Anderson-Schulz-Flory (ASF) distribution which define the
2C5H10 → 1
C10H20 stoichiometry of the overall reaction or the selectivity of the products.
2C6H12 → 0.85 The chain growth probability factor (α) depends on the process condi­
C12H24 tions and in this paper the value of 0.92 was used [27]. The ASF dis­
2C7H14 → 0.85 tribution fails to represent the data for light hydrocarbons and the
C14H28
following fractional selectivity for light hydrocarbons and the produced
2C8H16 → 0.8
C16H32 CO2 though the WGS reaction were used [28]: CH4 = 0.05, C2H4 =
2C9H18 → 0.8 0.0005, C2H6 = 0.01, C3H6 = 0.02, C3H8 = 0.01, C4H8 = 0.02, C4H10 =
C18H36 0.01 and CO2 = 0.01. Stoichiometric amount of H2 was fed to the hy­
Hydrotreater RSTOIC 300 ◦ C/30 bar C3H6 + H2 → 0.9
drocracker. For the ATR the estimated optimum H2O and O2/C molar
C3H8
C4H8 + H2 → 0.9 ratios were 0.6 and 0.2, respectively.
C4H10
C5H10 + H2 → 0.9 4.3. Methanol-based pathway
C5H12
C6H12 + H2 → 0.9
C6H14 Table 5 presents the models parameters that are used for the reactors
C7H14 + H2 → 0.9 simulation. The adopted methodology for the methanol synthesis is
C7H16 based on the assumption of the chemical equilibrium. The methanol to
C8H16 + H2 → 0.9
olefins products yields were obtained from Avidan [46]. The oligo­
C8H18
C9H18 + H2 → 0.9
merization reaction process specifications for ethylene was adopted
C9H20 from Ref. [33], propylene from Ref. [35] and butane from Ref. [34].
C10H20 + H2 → 0.9 Finally, for the hydrotreatment reactor process specifications it is
C10H22 assumed that all the olefins are converted into their respective n- and
C11H22 + H2 → 0.9
i-paraffins with a conversion rate of 90%.
C11H24
C12H24 + H2 → 0.9
C12H26 Table 4
C13H26 + H2 → 0.9 Reactors models specifications for the LTFT pathway.
C13H28
C15H30 + H2 → 0.9 Reactor name Reactor Process Associated Fractional
C15H32 model specifications reactions conversion
C16H32 + H2 → 0.9 Reverse Water REQUIL 900 ◦ C/5 bar CO2+H2 →CO + –
C16H34 Gas Shift H2O
C17H34 + H2 → 0.9 CO2+ 4H2
C17H36 →CH4+2H2O
C18H36 + H2 → 0.9 Fischer- RSTOIC 240 ◦ C/35 bar nCO + (2n+1)H2 0.85 (CO)
C18H38 Tropsch H2/CO = 2 →CnH2n+2 +
C19H38 + H2 → 0.9 nH2O
C19H40 CO + H2O →
C20H40 + H2 → 0.9 CO2 + H2
C20H42 Hydrocracker RSTOIC 240 ◦ C/35 bar C24H30 + H2 → 1
C24H50+ H2 → 1 2C12H16
2C12H26 C28H58 + H2 → 1
C24H48 + 2H2 → 1 2C14H30
2C12H26 C32H66 + 3H2 → 1
C28H58+ H2 → 1 4C8H18
2C14H30 C36H74 + 3H2 → 1
C28H56 +2H2 → 1 4C9H20
2C14H30 C40H82 + 3H2 → 1
C48H98 + 3H2 → 1 4C10H22
4C12H26 C48H98 + 3H2 → 1
C48H96 + 4H2 → 1 4C12H26
4C12H26 C56H114 + 3H2 → 1
Autothermal RGIBBS 950 ◦ C/5 bar – – 4C14H30
reactor O2/C = 0.7 C64H130 + 3H2 → 1
H2O/C = 4 4C16H34
The distillation columns are modeled as RADFRAC. The Peng-Robinson property Autothermal RGIBBS 950 ◦ C/5 bar – –
reactor O2/C = 0.6
method with Boston-Mathias modification is selected with STEAMNBS as free
H2O/C = 0.2
water method.
The distillation columns are modeled as RADFRAC. The Peng-Robinson property
method with Boston-Mathias modification is selected with STEAMNBS as free
water method.

8
K. Atsonios et al. Energy 278 (2023) 127868

Table 5
Reactors models specifications for the MeOH-based pathway.
Reactor name Reactor model Process specifications Associated reactions Fractional conversion

Methanol synthesis REQUIL 250 C/65 bar H2/CO2 = 3.0

◦
CO + H2O →H2+CO2 –
CO+ 2H2 →CH3OH
CO2+3H2→CH3OH + H2O
Methanol to Olefins RSTOICa 450 ◦ C/2 bar 2CH3OH → DME + H2O 1
DME →C2H4+H2O 0.637
3DME→ 2C2H4+3H2O 1
2C2H4 → C4H8 0.077
C2H4 + C3H6→C5H10 0.008 (C2H4)
2C3H6 → C6H12 0.008
C6H12→C6H6+3H2 0.500
Ethylene dimerization RSTOIC 25 ◦ C/50 bar 2C2H4 →nC4H8 0.890
3C2H4 →C6H12 0.035
2C2H4 →iC4H8 0.075
Propylene oligomerization RSTOIC 270 ◦ C/40 bar 2C3H6 →C4H8 0.034
3C3H6 →C9H18 0.231
4C3H6 →C12H24 0.284
5C3H6 →C15H30 0.157
6C3H6 →C18H36 0.052
Butene oligomerization RSTOIC 350 ◦ C/10 bar 2C4H8 →C8H16 0.394
3C4H8 →C12H24 0.344
4C4H8 →C16H32 0.041
ATR RGIBBS 980 ◦ C/150 bar
O2/C = 1.5
H2O/C = 3.0

All the distillation columns are modeled as RADFRAC. The Peng-Robinson property method is selected for all subsections apart from the methanol synthesis, where
NRTL-RK is used.
a
reactions occur in series.

4.4. Ethanol-based pathway

Table 7
Main flow rates and KPIs from mass balance calculations.
The methodology for the modeling of the ethanol production via
DME is presented in detail elsewhere [47]. The only difference is that for CO2FT LTFT MeOH EtOH

the DME synthesis reaction model, a more simplified approach based on CO2 flow in kg/h 4401 4401 4401 4401
equilibrium is adopted (REQUIL). Table 6 presents the models param­ H2 flow total kg/h 771 669 906 622
Oxygen demand kg/h 1176 374 1514 0
eters that are used for the reactors simulation.
Steam demand for ATR kg/h 3783 105 1669 0
The hydrotreatment model and the distillation columns modeling are Liquid products kg/h 1420 1420 1357 1355
based on the same approach to former case. The property method that is Jet Fuel flow kg/h 1073 1288 1165 885
used for that model is ENRTL-RK. Wastewater flow kg/h 8711 4129 6284 3666
CU % 75.7% 90.7% 82.7% 62.4%
Ye-jet % 20.8% 25.4% 22.0% 17.6%
5. Results and discussion
fe-jet % 75.5% 90.7% 85.8% 65.3%

5.1. Material balance

one achieves the best utilization of the H2/CO2 feed gases. The MeOH
The basic characteristics of the main streams of the four pathways based case has good performance indicators in terms of product yields
are presented in the Supplementary Information. Table 7 presents the and carbon utilization with slightly lower values that those of LTFT.
basic inlet/outlet streams flows and the respective indices as defined in The H2/CO2 ratio for the ethanol-based case is the lowest, however
Section 2.1. It is worth mention that all the examined cases have no this pathway results in the lowest kerosene yield. This is attributed to the
direct CO2 emissions as the initial amount of CO2 is entirely transformed relatively high selectivity in low carbon olefins after the oligomerization
into synthetic fuel. The FT-based pathways show the best performance reaction process that cannot be used for synthetic kerosene formulation.
in terms of final products yield with LTFT pathway being superior as The LTFT pathway has the lowest specific demands in oxygen for
illustrated at the highest Ye-jet and fe-jet values as well. Moreover, the last reforming owed to the reduced needs for light gas reforming compared

Table 6
Reactors models specifications for the ATJ section.
Reactor name Reactor model Process specifications Associated reactions Fractional conversion

Guerbet reactor RSTOIC 150 ◦ C/1.5 bar 2Ethanol → n-butanol + H2O 0.271
3Ethanol → hexanol + 2H2O 0.026
3Ethanol→2ethyl-hexanol+ 2H2O 0.009
Alcohols dehydration RSTOIC 285 ◦ C/1 bar Ethanol → C2H4+ H2O 1
Propanol → C3H6+ H2O 1
n-butanol → 1-C4H8+ H2O 0.8
2ethyl-hexanol→ C6H12+ H2O 1
Hexanol→ C6H12+ H2O 1
Hexene oligomerization RSTOIC 350 ◦ C/10 bar 2C6H12→C12H24
Butene oligomerization RSTOIC 350 ◦ C/10 bar 2 1-C4H8 →C8H16 0.2
3 1-C4H8 →C12H24 0.7
4 1-C4H8 →C16H32 0.1

9
K. Atsonios et al. Energy 278 (2023) 127868

to the other two cases. Supposing that the green hydrogen flow comes greater than then final product streams (both jet fuel and diesel). This is
from a renewable driven electrolysis unit, the amount of oxygen that is attributed to the inevitable selectivity in light olefins at the oligomeri­
co-produced is enough to cover the oxygen demands at the three cases zation catalyst.
where an ATR unit is considered. Special attention should be paid on the In the case of the EtOH route (Fig. 11), it is obvious that there is a
proper water management, as the wastewater is the largest stream of all considerable large amount of gas recycling between the DME and the
at four cases. This effluent must be purified and reused either at the Ethanol synthesis which is more than 5 times greater than the heat input
green hydrogen production unit or at the steam generation for ATR of the hydrogen feed. The Sankey diagram indicates that a more efficient
operation. This represents around 64% (MeOH case) to 81% (CO2FT way to handle the CO/H2 stream that is needed for DME carbonylation
case) of the total process water demands. should be adopt in future improved versions of that pathway. As for
upgrading part of ethanol into advanced paraffinic fuels, the energetic
efficiency is 95.6% implying that the heat content of the inlet steams
5.2. Energy balance (ethanol and hydrogen) has a minor degradation during the 3-step
ethanol conversion into jet fuel and diesel .
Table 8 summarizes all the aggregated main aspects of the heat and Table 9 summarizes the hot and cold utilities for the examined cases.
energy balance of the four pathways. The electrolyser consumption that CO2FT case has the lower hot utilities (only 0.11 MW HP steam) and a
is roughly calculated based on the assumption of a specific energy considerable amount of LP steam (10.6 MW) can be exploited externally
consumption of 53.79 kWh/kgH2 [48] for a typical PEM electrolyser is or sold in the framework of industrial symbiosis. Similarly, at the LTFT
included in the table results for comparison purpose but it is not taken case, 8.2 MW of LP steam is produced but 1.2 MW of heat with tem­
into account at the overall calculations. Following the previous analysis perature >900 ◦ C is required for the rWGS reactor operation. The MeOH
on mass balance, LTFT demonstrates the best performance of all in terms case has practically no need for external heat as the amount of HP/IP/LP
of energy efficiency. More than 70% of fresh hydrogen energy content is steam can be fulfilled from the IP/HP steam that is generated internally.
finally converted into synthetic aviation fuel proving the effectiveness of The EtOH case has considerable amount of energy inflows and outflows
this pathway. at different forms (steam and refrigerants with multiple properties). All
An illustrative view of how the energy is distributed along each the cases apart from the last one have positive balance at inlet/outlet
process is granted by the Sankey diagrams. The flows that represents the utilities, having thus an additional beneficial environmental impact by
heat content of material streams are expressed on Higher Heating Value providing steam with zero carbon footprint.
basis. In addition, the sensible heat of these streams has been taken into Although a life cycle analysis is not considered in that study, it is
consideration in order to close the heat balance. easily extracted that the environmental impact of the produced synthetic
In Fig. 8, the important role of ATR is illustrated. The gas (reformate kerosene and its potential GHG emissions reduction after replacement of
gas) that is produced from all the low-quality, low-importance gas from fossil derived aviation fuel is mainly depended on the way that hydrogen
the oligomerization and hydrotreatment has slightly higher heat content is produced and the origin of the CO2 stream. Moreover, the effective
of the fresh hydrogen that enters the FT synthesis reactor. Oligomeri­ management of the utilities at the three first cases further decreases the
zation and hydrotreatment are set to operate in high efficiency as the carbon footprint at the produced synthetic fuels.
waste heat from these reactors is low. The waste heat from the ATR can
be potentially exploited for the steam generation (this part has not been 5.3. Exergy analysis
taken into detail in the CO2FT analysis).
Fig. 9 shows the Sankey diagram for the LTFT case. Although rWGS is The results from the exergy analysis are presented in Table 10.
an endothermic, energy demanding process, the required heat for its CO2FT pathway is characterized from the high exergy that exits the
stable operation is 5.7% of the fresh hydrogen heat content. The ratio of process in the form of (waste) heat. The exergetic efficiency (ηex) of LTFT
the recycling streams to the fresh gas stream is the lowest of all the pathway is the highest, verifying what has been previously mentioned in
examined cases, this has beneficial impact on the reactors size and the energy analysis section. The high final products yield, the low exergy
consequently to the total equipment cost. Thanks to the effective heat inlet because of the low power demand and relatively low heat input are
integration network, the waste heat from the overall process is 29% of the main factors for that. On the other hand, there is room for
the hydrogen heat input. improvement in the performance of the alcohol-based pathways as it is
Fig. 10 depicts the impact of the ATR use for the light gas/purge gas illustrated from their high exergy loss. Especially for the ethanol-based
utilization. A recycling stream of 19.8 MW with CO and H2 is created pathway, new ways for producing DME more effectively or direct
and reused for methanol synthesis. As a result, this stream has a com­ ethanol from CO2 should be investigated. For the methanol-based
parable heat content with that of hydrogen. The methanol to oligomers pathway, the olefins oligomerization was based on certain studies that
process chain has a very good energy conversion efficiency, as almost handle each light olefin separately. If their oligomerization can be
10% of the heat content is lost as unexploited (waste) heat. Nevertheless, accomplished as much as effectively even with the presence of the rest
the paraffins that are final formed after the hydrotreatment process and compounds after the MTO reactor, the refrigeration loads and the
their carbon number is < 9 and are not suitable for aviation fuel is respective electricity consumption can be avoided at the MTO section for
the light olefins separation.
Table 8
Heat and Energy balance main results. 5.4. Energy analysis of the LTFT pathway
CO2FT LTFT MeOH EtOH

Thermal input H2 (LHV base) MWth 25.67 22.29 30.20 20.70

As shown above, the LTFT pathway presents the best performance in
Electrolyser consumption MWe 41.12 35.70 48.37 33.17 terms of all indicators. In this paragraph, an energy analysis is presented
Compression Consumptions MWe 8.54 2.18 4.30 10.17 by using the Aspen Energy Analyser. The results show that the current
Refrigeration demands MWth 0.00 0 0.21 7.822 utility duty use is 29.88 MW and the target utility duty is 28.04 MW.
Power for refrigeration MWe 0 0 0.65 24.56
These results demonstrate the efficient use of energy as energy savings
Total power demands MWe 8.54 2.18 4.95 34.73
External heat demands MWth 0.24 1.74 0 10.00 potential is very low at 6.19%. The composite curves for ΔTmin = 10 ◦ C is
Liquid fuels heat input (LHV) MWth 17.77 17.40 15.99 16.53 shown in Fig. 12. The large amount of cold utility duty needed is due to
Jet Fuel heat input (LHV) MWth 13.14 15.80 13.75 10.84 the residual duty of the ATR (at 135 ◦ C) and rWGS (at 150 ◦ C) streams
EJFE % 51.2% 70.9% 45.5% 52.4% after the pre-heating of inlet streams of these two reactors and the steam
Total efficiency % 51.6% 66.4% 45.5% 25.3%
generation required for ATR.

10
K. Atsonios et al. Energy 278 (2023) 127868

Fig. 8. Sankey diagram for CO2FT route (HHV basis).

Fig. 9. Sankey diagram for LTFT route (HHV basis).

Fig. 10. Sankey diagram for methanol-based route (HHV basis).

11
K. Atsonios et al. Energy 278 (2023) 127868

Fig. 11. Sankey diagram for ethanol-based route (HHV basis).

5.5. Jet fuel characteristics

Table 9
Energy utilities (in kW) per each case.
Some of the most critical properties of the final synthetic fuels that
CO2FT LTFT MeOH EtOH are calculated from Aspen Plus and are compared with the respective
Very High T 1243 values from Jet-A1 specifications [49]. Table 11 summarizes these
Fired Heat 549 values. It is clear that the four pathways produce kerosene with basic
HP Steam 108 497 353 1007 properties close to the required Jet-A1 specifications. The deviations
IP Steam 23.8 1686
LP Steam 129 1075 9209
from the specifications are observed for density (in the range of
Air vent (heating) 686 2.7–4.4%) and for 100% distillation for CO2FT (2.8%). Although the
Total Hot Utilities 237 1740 2138 12,451 deviations are quite small, these can be eliminated if a better distillation
Cooling Water 4945 2639 12,300 18,220 of the kerosene fraction is configured by reducing a bit the lighter
Air vent (cooling) 577
fractions that contribute to the decrease of density and the heavier
LP Steam Generation 10,764 8164 1333
IP Steam Generation 3564 318 fractions that favour the elevation of Distillation 100% temperature.
HP Steam Generation 1305 239 Also, the addition of other compunds, e.g. aromatics, would result in
Refrigerant 1 67 better results for deisnty.
Refrigerant 2 926 5456 It should be underlined that even though it is not feasible to consider
Refrigerant 3 166
Total Cold Utilities 15,709 11,380 18,261 25,633
all the chemical compounds that exist in the final kerosene stream, the
basic hydrocarbons that are considered in all cases can lead to a very
good prediction at least of the basic jet fuel properties. This also is
confirmed by other similar simulation studies that reported the respec­
Table 10
tive jet fuel properties.
Exergy balance results.
Finally, freezing point is of the most important property parameters
Case CO2FT LTFT МeΟН EtOH for the evaluation of the e-kerosene product. However, Aspen does not
Pin (MW) 8.54 2.18 4.95 34.73 support the calculation of the freezing point of mixtures and it is
EH2,in
˙ (MW) 25.06 21.77 29.49 20.22 generally complicated to estimate the value using a prediction model.
ECO2,in
˙ (MW) 0.54 0.54 0.54 0.54 Indicatively, if a methodology based on the freezing point of each
EQ,in
˙ (MW) 3.81 1.09 0.00 2.12 component found in the kerosene final stream and the respective volume
Ein,tot
˙ (MW) 37.96 25.59 34.98 57.62
fraction is applied for the four examined cases [50] the calculated
Ejet˙ fuel (MW) 14.15 17.95 15.05 11.62
freezing point values varies from − 39.8 to − 6.7 ◦ C, which is above the
Ediesel&gasoline
˙ (MW) 4.28 0.63 2.27 6.10
maximum value for Jet A1 (− 47 ◦ C) and requires further investigation.
EQ,out
˙ (MW) 10.89 3.90 2.91 4.24
Eout,tot
˙ (MW) 29.50 22.48 20.23 21.96
Eloss
˙ (MW) 8.65 3.11 14.76 35.66 5.6. Comparison with other similar studies
ηwasted (%) 28.7% 15.2% 8.3% 7.4%
ηloss (%) 22.8% 12.2% 42.2% 61.9%
ηex (%) 48.5% 72.6% 49.5% 30.8% The key performance indicators of the four pathways are compared
with the respective results of indicative studies from the literature and
summarized in Table 12. Apart from the main KPIs that are defined in
Section 2.1, one more is added in order to express the conversion effi­
ciency of all the produced fuels:

12
K. Atsonios et al. Energy 278 (2023) 127868

Fig. 12. Composite curve of the heat integrated LTFT flowsheet.

conventional LTFT process) and two new (light alcohols based) path­
Table 11
ways for CO2 catalytic conversion into synthetic jet fuel. They were
Produced synthetic jet fuel properties.
developed in such way that the jet fuel fraction is maximized compared
Jet A-1 CO2FT LTFT MeOH EtOH to other liquid fuels fraction (gasoline and diesel) and the main prop­
LHV MJ/kg >42.80 44.12 44.15 42.50 44.10 erties to follow the respective Jet-A1 specifications. The analysis of these
Density kg/m3 775–840 747.4 744.3 740.6 754.0 pathways was on the basis of the process design and comparative
Viscosity (-20oC) mm2/s 4.32 5.28 3.36 5.44
o
<8.0
assessment was made in terms of process performance, energy and
Flash Point C >38 45.6 44.4 37.9 49.7
Distillation 10% o
C <205 163.6 153.1 156.1 171.45
exergy. The process simulations results revealed that LTFT pathway
Distillation 100% o
C <300 308.5 288.7 282.9 279.30 presents the best performance as it is illustrated from the highest values
of jet fuel yields and the energy & exergy indices. This implies that the
most efficient way to produce synthetic kerosene from CO2 with ther­
Table 12
mocatalytic techniques is first to convert it into CO and then the CO/H2
KPIs from various synthetic fuels production systems and comparison. stream to be transformed into hydrocarbons via Fischer-Tropsch syn­
thesis. The alcohol-based routes presented also good performance in
Study technology Ye-jet fe-jet EJFE ηtot ηtot fuels
terms of the target product yield and the respective carbon utilization
König et al. [14] rWGS - FT 9.3% 43.9% 29.3% 28.5% 67.0% and have the potential to be competitive to the FT-based pathways if
Zang et al. [51] rWGS - FT 6.3% 46.7% 27.0% 26.7% 57.8%
certain improvements in catalysts performance (higher selectivity in
Ruokonen et al. MeOH 7.6% 21.8% 21.8% 21.8% 77.4%
[13] C12–C14 compounds) is achieved. All the proposed pathways demon­
this study CO2FT 20.8% 75.5% 51.2% 51.6% 69.2% strated improved performance compared to other similar approaches in
this study LTFT 25.4% 90.7% 70.9% 66.4% 78.1% the literature as concerns the maximization of the aviation fuel fraction,
this study MeOH 22.0% 85.8% 45.5% 45.5% 52.9% when it is considered as the desired end-product. However, a weakness
this study EtOH 17.6% 65.3% 52.4% 25.3% 79.9%
is the higher level of complexity for that cases as a certain number of
catalytic reactors and recovery units of the intermediate products are
∑
ṁi • LHVi required in order to convert methanol and ethanol into long chain
ηtot,fuels = eq 33 paraffinic fuels. An overall assessment in terms of production cost and
ṁH2,in • LHVH2
environmental impact is needed as a future work in order to get a clear
where i = jet fuel, diesel, gasoline, gas fuel etc. insight of the examined pathways prospects. Such a holistic analysis will
It makes clear that the four pathways investigated in this study address to determine the proper plant capacity and the influence of the C
present a significant advancement in maximization of jet fuel production utilization unit on the total investment compared to the rest parts of the
yield, as illustrated from the Ye-jet and fe-jet that are higher than the system i.e. the hydrogen and pure CO2 production units.
respective performance values found in the literature. Moreover, the
way that the inlet energy (hydrogen heat input, heat and electricity) are Author contribution
converted into the desired synthetic aviation fuel is more efficient. The
EtOH based route presents the highest performance in terms of hydrogen K. Atsonios: Conceptualization, Methodology, Investigation, Soft­
conversion into all the liquid fuels, even though the e-kerosene yield (Ye- ware, Validation, Visualization, Writing – original draft, J. Li: Software,
jet) is the lowest of the four new processes but not lower that the two Validation, V.J. Inglezakis: Project administration, Conceptualization,
studies from the literature. Methodology, Validation, Software, Writing – review & editing.

6. Conclusions
Declaration of competing interest
This study presents two Fischer-Tropsch based (one unconventional
The authors declare that they have no known competing financial
high temperature FT without a reverse WGS reactor in addition to a
interests or personal relationships that could have appeared to influence

13
K. Atsonios et al. Energy 278 (2023) 127868

the work reported in this paper. [8] Pratschner S, Hammerschmid M, Müller FJ, Müller S, Winter F. Simulation of a
pilot scale power-to-liquid plant producing synthetic fuel and wax by combining
fischer–tropsch synthesis and SOEC. Energies 2022;15:4134. https://doi.org/
Data availability 10.3390/en15114134.
[9] INERATEC. Sustainable e-fuels for aviation n.d. https://ineratec.de/en/e-fuels-for-
No data was used for the research described in the article. aviation/. [Accessed 14 November 2022].
[10] ExxonMobil. ExxonMobil methanol to jet technology to provide new route for
sustainable aviation fuel production n.d. https://www.exxonmobilchemical.
Abbreviations com/en/resources/library/library-detail/101116/exxonmobil_sustainabl
e_aviation_fuel_production_en?utm_source=google&utm_medium=cpc&u
tm_campaign=cl_downstream_none&ds_k=&gclsrc=aw.
ASF Anderson-Schulz-Flory ds&&ppc_keyword=aviation fuel&gclid=CjwKCAjwo.
ATJ Alcohol to Jet [11] Sudiro M, Bertucco A. Production of synthetic gasoline and diesel fuel by
ATR autothermal reactor alternative processes using natural gas and coal: process simulation and
optimization. Energy 2009;34:2206–14. https://doi.org/10.1016/j.
CHJ catalytic hydrothermolysis energy.2008.12.009.
COP coefficient of performance [12] Navas-Anguita Z, Cruz PL, Martín-Gamboa M, Iribarren D, Dufour J. Simulation
CO2FT High-temperature Fischer-Tropsch synthesis without rWGS and life cycle assessment of synthetic fuels produced via biogas dry reforming and
Fischer-Tropsch synthesis. Fuel 2019;235:1492–500. https://doi.org/10.1016/j.
CU Carbon Utilization fuel.2018.08.147.
DME dimethyl ether [13] Ruokonen J, Nieminen H, Dahiru AR, Laari A, Koiranen T, Laaksonen P, et al.
EA ethyl acetate Modelling and cost estimation for conversion of green methanol to renewable
liquid transport fuels via olefin oligomerisation. Processes 2021;9. https://doi.org/
EJFE Energetic Jet Fuel Efficiency
10.3390/pr9061046.
EtOH ethanol [14] König DH, Baucks N, Dietrich RU, Wörner A. Simulation and evaluation of a
FT Fischer-Tropsch process concept for the generation of synthetic fuel from CO2 and H2. Energy
FT-SKA Fischer-Tropsch synthetic kerosene with aromatics 2015;91:833–41. https://doi.org/10.1016/j.energy.2015.08.099.
[15] Petersen AM, Chireshe F, Okoro O, Gorgens J, van Dyk J. Evaluating refinery
FT-SRK Fischer-Tropsch synthetic paraffinic kerosene configurations for deriving sustainable aviation fuel from ethanol or syncrude. Fuel
GHG greenhouse gas Process Technol 2021;219. https://doi.org/10.1016/j.fuproc.2021.106879.
HC-HEFA-SPK synthesized paraffinic kerosene from hydrocarbon- [16] Atsonios K, Kougioumtzis M-A, Panopoulos KD, Kakaras E. Alternative
thermochemical routes for aviation biofuels via alcohols synthesis: process
hydroprocessed esters & fatty acids modeling, techno-economic assessment and comparison. Appl Energy 2015;138.
HEFA Hydroprocessed Esters and Fatty Acids https://doi.org/10.1016/j.apenergy.2014.10.056.
HHV Higher Heating Value [17] De Falco M, Natrella G, Capocelli M, Popielak P, Sołtysik M, Wawrzyńczak D, et al.
Exergetic analysis of DME synthesis from CO2 and renewable hydrogen. Energies
IP Intermediate pressure 2022;15. https://doi.org/10.3390/en15103516.
HP High pressure [18] Szargut J. Exergy method: technical and ecological applications. 2005. UK.
LHV Lower Heating Value [19] Kotas TJ. The exergy method of thermal plant analysis. Butterworth-Heinemann;
1985. https://doi.org/10.1016/C2013-0-00894-8.
LP Low pressure [20] Wang W, Wang S, Ma X, Gong J. Recent advances in catalytic hydrogenation of
LTFT Low Temperature Fischer-Tropsch carbon dioxide. Chem Soc Rev 2011;40:3703–27. https://doi.org/10.1039/
MA methyl acetate c1cs15008a.
[21] Choi YH, Jang YJ, Park H, Kim WY, Lee YH, Choi SH, et al. Carbon dioxide Fischer-
MeOH methanol
Tropsch synthesis: a new path to carbon-neutral fuels. Appl Catal, B 2017;202:
MOGD Mobil’s olefins to gasoline and distillate 605–10. https://doi.org/10.1016/j.apcatb.2016.09.072.
MTO methanol-to-olefins [22] Rodemerck U, Holeňa M, Wagner E, Smejkal Q, Barkschat A, Baerns M. Catalyst
PEM Proton Exchange Membrane development for CO2 hydrogenation to fuels. ChemCatChem 2013;5:1948. https://
doi.org/10.1002/cctc.201200879. –55.
rWGS reverse water gas shift reaction [23] Jiang F, Liu B, Geng S, Xu Y, Liu X. Hydrogenation of CO2 into hydrocarbons:
SAF sustainable aviation fuel enhanced catalytic activity over Fe-based Fischer-Tropsch catalysts. Catal Sci
SIP synthesized iso-paraffins Technol 2018;8:4097–107. https://doi.org/10.1039/c8cy00850g.
[24] de Klerk A. Fischer-tropsch refining. PhD. Univeristy of Pretoria; 2008.
WGS water-gas shift reaction [25] Steynberg A, Dry M. Fischer-tropsch technology. Elsevier; 2004.
[26] Chen X, Chen Y, Song C, Ji P, Wang N, Wang W, et al. Recent advances in
Appendix B. Supplementary data supported metal catalysts and oxide catalysts for the reverse water-gas shift
reaction. Front Chem 2020;8. https://doi.org/10.3389/fchem.2020.00709.
[27] Adams TA, Barton PI. Combining coal gasification and natural gas reforming for
Supplementary data to this article can be found online at https://doi. efficient polygeneration. Fuel Process Technol 2011;92:639–55. https://doi.org/
org/10.1016/j.energy.2023.127868. 10.1016/j.fuproc.2010.11.023.
[28] Scott AJ, Adams II TA. Biomass-gas-and-nuclear-to-liquids (BGNTL) processes Part
I: model development and simulation. Can J Chem Eng 2018;96:1853–71.
References [29] Link F, Ahad N, de Klerk A. Low-pressure hydrocracking of wax over Pt/SiO2-
Al2O3to produce kerosene for synthetic jet fuel. ACS (Am Chem Soc) Symp Ser
[1] Kyriakopoulos GL, Streimikiene D, Baležentis T. Addressing challenges of low- 2021;1379:311–52. https://doi.org/10.1021/bk-2021-1379.ch012. American
carbon energy transition. Energies 2022;15. https://doi.org/10.3390/ Chemical Society.
en15155718. [30] Zhang Y, Dou B, Liu X, Fan H, Geng C, Liu X, et al. Experimental and theoretical
[2] EEA. Greenhouse gas emissions by aggregated sector n.d. https://www.eea.europa. study of the adsorption of mixed low carbon alcohols and acids from Fischer
eu/data-and-maps/daviz/ghg-emissions-by-aggregated-sector-5#tab-dash Tropsch synthesis wastewater by activated carbon. Fuel 2023;339:126928. https://
board-02. doi.org/10.1016/j.fuel.2022.126928.
[3] Eurostat. Greenhouse gas emissions falling in most source sectors n.d. https://ec. [31] Zhang R, Huang Y, Zheng K, Xu C. Design and control of fraction cutting for the
europa.eu/eurostat/en/web/products-eurostat-news/-/ddn-20220823-1. separation of mixed alcohols from the Fischer–Tropsch aqueous by-products. Chin
[4] Woodroffe JD, Harvey BG. High-performance, biobased, jet fuel blends containing J Chem Eng 2022;50:143–54. https://doi.org/10.1016/j.cjche.2022.08.005.
hydrogenated monoterpenes and synthetic paraffinic kerosenes. Energy Fuel 2020; [32] Gogate MR. Methanol-to-olefins process technology: current status and future
34:5929–37. https://doi.org/10.1021/acs.energyfuels.0c00274. prospects. Petrol Sci Technol 2019;37:559–65. https://doi.org/10.1080/
[5] Wang M, Dewil R, Maniatis K, Wheeldon J, Tan T, Baeyens J, et al. Biomass-derived 10916466.2018.1555589.
aviation fuels: challenges and perspective. Prog Energy Combust Sci 2019;74: [33] Metzger ED, Brozek CK, Comito RJ, Dincă M. Selective dimerization of ethylene to
31–49. https://doi.org/10.1016/J.PECS.2019.04.004. 1-butene with a porous catalyst. ACS Cent Sci 2016;2:148–53. https://doi.org/
[6] Zhang L, Dang Y, Zhou X, Gao P, Petrus van Bavel A, Wang H, et al. Direct 10.1021/acscentsci.6b00012.
conversion of CO2 to a jet fuel over CoFe alloy catalysts. Innovation 2021;2. [34] Zhang X, Zhong J, Wang J, Gao J, Liu A. Trimerization of butene over Ni-doped
https://doi.org/10.1016/j.xinn.2021.100170. zeolite catalyst: effect of textural and acidic properties. Catal Lett 2008;126:
[7] Marchese M, Giglio E, Santarelli M, Lanzini A. Energy performance of Power-to- 388–95. https://doi.org/10.1007/s10562-008-9642-y.
Liquid applications integrating biogas upgrading, reverse water gas shift, solid [35] Li X, Han D, Wang H, Liu G, Wang B, Li Z, et al. Propene oligomerization to high-
oxide electrolysis and Fischer-Tropsch technologies. Energy Convers Manag X quality liquid fuels over Ni/HZSM-5. Fuel 2015;144:9–14. https://doi.org/
2020;6:100041. https://doi.org/10.1016/J.ECMX.2020.100041. 10.1016/J.FUEL.2014.12.005.

14
K. Atsonios et al. Energy 278 (2023) 127868

[36] Lu WZ, Teng LH, Xiao W De. Simulation and experiment study of dimethyl ether [44] Fazeli H, Panahi M, Rafiee A. Investigating the potential of carbon dioxide
synthesis from syngas in a fluidized-bed reactor. Chem Eng Sci 2004;59:5455–64. utilization in a gas-to-liquids process with iron-based Fischer-Tropsch catalyst.
https://doi.org/10.1016/J.CES.2004.07.031. J Nat Gas Sci Eng 2018;52:549–58. https://doi.org/10.1016/j.jngse.2018.02.005.
[37] Lu P, Yang G, Tanaka Y, Tsubaki N. Ethanol direct synthesis from dimethyl ether [45] Meurer A, Kern J. Fischer–tropsch synthesis as the key for decentralized
and syngas on the combination of noble metal impregnated zeolite with Cu/ZnO sustainable kerosene production. Energies 2021;14. https://doi.org/10.3390/
catalyst. Catal Today 2014;232:22–6. https://doi.org/10.1016/J. en14071836.
CATTOD.2013.10.042. [46] Avidan AA. Gasoline and distillate fuels from methanol. Stud Surf Sci Catal 1988;
[38] Atsonios K, Panopoulos KD, Kakaras E. Thermocatalytic CO2 hydrogenation for 36:307–23. https://doi.org/10.1016/S0167-2991(09)60524-3.
methanol and ethanol production: process improvements. Int J Hydrogen Energy [47] Atsonios K, Panopoulos KD, Kakaras E. Thermocatalytic CO2 hydrogenation for
2016;41:792–806. https://doi.org/10.1016/J.IJHYDENE.2015.12.001. methanol and ethanol production: process improvements. Int J Hydrogen Energy
[39] Wang Y, Wang K, Zhang B, Peng X, Gao X, Yang G, et al. Direct conversion of CO2 2016;41. https://doi.org/10.1016/j.ijhydene.2015.12.001.
to ethanol boosted by intimacy-sensitive multifunctional catalysts. ACS Catal 2021; [48] Tenhumberg N, Büker K. Ecological and economic evaluation of hydrogen
11:11742–53. https://doi.org/10.1021/acscatal.1c01504. production by different water electrolysis technologies. Chem Ing Tech 2020;92:
[40] Geleynse S, Brandt K, Garcia-Perez M, Wolcott M, Zhang X. The alcohol-to-jet 1586–95. https://doi.org/10.1002/cite.202000090.
conversion pathway for drop-in biofuels: techno-economic evaluation. [49] Standard, Alternative Fuel Specifications. Biofuels for aviation: feedstocks,
ChemSusChem 2018;11:3728–41. https://doi.org/10.1002/cssc.201801690. technology and implementation. 2016. p. 351–7. https://doi.org/10.1016/B978-0-
[41] Dowson GRM, Haddow MF, Lee J, Wingad RL, Wass DF. Catalytic conversion of 12-804568-8.00023-8.
ethanol into an advanced biofuel: unprecedented selectivity for n-butanol. Angew [50] Boehm RC, Coburn AA, Yang Z, Wanstall CT, Heyne JS. Blend prediction model for
Chem Int Ed 2013;52:9005. https://doi.org/10.1002/anie.201303723. 8. the freeze point of jet fuel range hydrocarbons. Energy Fuels 2022;36:12046–53.
[42] Saeidi S, Amin NAS, Rahimpour MR. Hydrogenation of CO2 to value-added https://doi.org/10.1021/acs.energyfuels.2c02063.
products - a review and potential future developments. J CO2 Util 2014;5:66–81. [51] Zang G, Sun P, Elgowainy AA, Bafana A, Wang M. Performance and cost analysis of
https://doi.org/10.1016/j.jcou.2013.12.005. liquid fuel production from H2and CO2based on the Fischer-Tropsch process.
[43] Landau Mv, Vidruk R, Herskowitz M. Sustainable production of green feed from J CO2 Util 2021;46. https://doi.org/10.1016/j.jcou.2021.101459.
carbon dioxide and hydrogen. ChemSusChem 2014;7:785–94. https://doi.org/
10.1002/cssc.201301181.

15