Techno-Economic Analysis of Processes With Integration of Fluidized Bed Heat Exchangers For H2 Production e Part 2 - Chemical-Looping Combustion

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 6 ( 2 0 2 1 ) 2 5 3 5 5 e2 5 3 7 5
Available online at www.sciencedirect.com
ScienceDirect
journal homepage: www.elsevier.com/locate/he
Techno-economic analysis of processes with

integration of fluidized bed heat exchangers for H2
production e Part 2: Chemical-looping combustion
Viktor Stenberg a,*, Vincenzo Spallina b, Tobias Mattisson a, Magnus Ryden a
a
Division of Energy Technology, Department of Space, Earth and Environment, Chalmers University of Technology,
SE-412 96, Gothenburg, Sweden
b
Department of Chemical Engineering and Analytical Science, The University of Manchester, Sackville Street, M13
9PL, Manchester, United Kingdom
highlights
Chemical-looping combustion can be integrated in a SMR plant for H2 production.

NG and biomass are evaluated as supplementary fuels to the CLC system.
Both systems show a higher hydrogen production efficiency than the conventional SMR plant.
Both systems show a lower/similar levelized hydrogen production cost than the conventional SMR plant.
Biomass fuelled plant can achieve significant negative CO2 emissions.
article info abstract
Article history: This work covers a techno-economic assessment for processes with inherent CO2 separation,
Received 27 October 2020 where a fluidized bed heat exchanger (FBHE) is used as heat source for steam reforming in a
Received in revised form hydrogen production plant. This article builds upon the work presented in Part 1 of this study by
6 April 2021 Stenberg et al. [1], where a process excluding CO2 capture was examined. Part 2 suggests two
Accepted 26 April 2021 process configurations integrating steam reforming with a chemical-looping combustion (CLC)
Available online 8 June 2021 system, thus providing inherent CO2 capture. The first system (case CM) uses natural gas as
supplementary fuel whereas the second system (case CB) uses solid biomass, which enables net
Keywords: negative CO2 emissions. In both systems, the reformer tubes are immersed in a bubbling fluidized
Chemical-looping combustion bed where heat for steam reforming is efficiently transferred to the tubes. The processes include
Hydrogen CO2 compression for pipeline transportation, but excludes transport and storage. The CLC system
Steam reforming is designed based on key parameters, such as the oxygen carrier circulation rate and oxygen
Fluidized bed heat exchanger transport capacity. The first system displays a process with net zero emissions and a hydrogen
Carbon capture production efficiency which is estimated to 76.2%, which is almost 8% higher than the conven-
Negative emissions tional process. The levelized production cost is 1.6% lower at below 2.6 V/kg H2. The second
system shows the possibility to reduce the emissions to 34.1 g CO2/MJH2 compared to the con-
ventional plant which emits 80.7 g CO2/MJH2. The hydrogen production efficiency is above 72% and
around 2% higher than the conventional process. The capital investments are higher in this plant
and the levelized hydrogen production cost is estimated to around 2.67 V/kg. The cost of CO2
avoidance, based on a reference SMR plant with CO2 capture, is low for both cases (4.3 V/tonCO2
for case CM and 2.7 V/tonCO2 for case CB).
© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
* Corresponding author.
E-mail addresses: [email protected] (V. Stenberg), [email protected] (V. Spallina), [email protected]
(T. Mattisson), [email protected] (M. Ryden).
https://doi.org/10.1016/j.ijhydene.2021.04.170
0360-3199/© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
25356 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 6 ( 2 0 2 1 ) 2 5 3 5 5 e2 5 3 7 5
Abbreviations NETL National energy technology laboratory

NG Natural gas
AFR Air-to-fuel ratio
NG feed Natural gas fed to the reformer tubes
AR Air reactor
NG fuel Natural gas used as supplementary fuel
ASU Air separation unit
NGCC Natural gas combined cycle
BEC Bare Erected Cost
O&M Operations and maintenance
BFB Bubbling fluidized bed
OC Oxygen carrier
BECCS Bio Energy Carbon Capture and Storage
OCAC Oxygen carrier aided combustion
BHF Baghouse filter
Off-gas Low pressure stream of gases obtained from the
Case A Reference hydrogen production process based on
PSA
SMR (see Part 1 of this work)
OP Oxygen polishing
Case CB Process based on CLC integrated with SMR with
RGibbs Reactor type in Aspen Plus used to calculate outlet
biomass as supplementary fuel
stream based on equilibrium (minimizes Gibbs
Case CB2 Process similar to case CB but operated with less
energy for the system)
supplementary fuel consumption
RStoich Reactor type in Aspen Plus used to calculate outlet
Case CM Process based on CLC integrated with SMR with
stream based on provided reaction stoichiometry
NG as supplementary fuel
PSA Pressure swing adsorption
Case CM2 Process similar to case CM but with FBHE placed
S/C ratio Steam-to-carbon ratio (molar basis)
in the air reactor instead
SH Superheater
Case OM Process based on OCAC integrated with SMR (see
SMR Steam methane reforming
Part 1 of this work)
TPC Total plant cost
C28 Mn-based perovskite structure oxygen carrier
VSA Vacuum swing adsorption
material, CaMn0.775Mg0.1Ti0.125O3-d
WGS Water-gas shift
CCA Cost of CO2 avoided
CCF Capital charge factor Symbols
CCS Carbon capture and storage Co Reference cost of equipment with reference
CCU Carbon capture and utilization capacity S0
CDR Carbon dioxide removal dto Reformer outside tube diameter
CFB Circulating fluidized bed E Emissions of CO2 from hydrogen production plant
CLC Chemical-looping combustion m_ Mass flow (kg/s)
CLOU Chemical looping with oxygen uncoupling Nt Number of reformer tubes
COND Condenser NOx Nitrogen oxides
CSIC Spanish National Research Council h Efficiency (%)
DES Desulfurization UD Oxygen demand (%)
DOE U.S. Department of Energy Qth Heat delivered from steam export (W)
EBTF European benchmarking task force S Capacity (scaling parameter in cost functions)
ECO Economizer T Temperature ( C)
EOR Enhanced oil recovery U Overall heat transfer coefficient (W/(m2K))
EVA Evaporator W Power (W)
FBHE Fluidized bed heat exchanger Y Yield
FR Fuel reactor
Indices/Exponents/subscripts
Fuel gas Gas stream consisting of PSA off-gas and
no arrow Case neglecting use of LP steam as heat export
supplementary fuel
d Degree of oxygen deficiency
GFF Gas-fired furnace (in conventional SMR)
bio Biomass
HHV Higher heating value
el Electricity
HP High pressure
eq Equivalent
HT High temperature
exp Export
HX Heat exchanger
o Outside
LCOH Levelized cost of hydrogen
th Thermal
LHV Lower heating value
tot Total
LP Low pressure
ref Reformer
LT Low temperature
reference Reference plant
Me Generic oxygen carrier
x,y Whole numbers
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 6 ( 2 0 2 1 ) 2 5 3 5 5 e2 5 3 7 5 25357
The most common approach for CO2 capture in the in-

Introduction dustry is to use a chemical absorption process with amines
[7,14e16]. In ammonia plants these units are commonly used
Global hydrogen production causes emissions corresponding for purification of hydrogen - instead of a pressure swing
to more than 1% of the global CO2 emissions [2]. The main adsorption - with a subsequent methanation reactor [5]. There
production route for hydrogen production is steam reforming are however several drawbacks with the use of amine ab-
of methane/natural gas and reforming of oil/naphtha. During sorption technologies for CO2 capture: i) significant energy
the next 20 years hydrogen production from fossil fuels is demand for the desorption of CO2 to regenerate the amine
considered to be the most economically feasible alternative [14]. ii) tendency for corrosion in the equipment and iii) amine
and it could play an important role for global hydrogen gen- degradation and loss which reduces its lifetime [14]. A fourth
eration during the next 20e50 years [3]. One of the most possible drawback is the potential health and environmental
interesting techniques to achieve significant CO2 emissions risks that could arise from amine losses from the plant where
reductions is by using carbon capture and storage (CCS). In the degradation products could be toxic [17,18].
order to achieve this, the industry of hydrogen production has An alternative technology which can be used for CO2 cap-
been identified as one of the most interesting sectors in regard ture is chemical-looping combustion (CLC) [19]. Chemical-
to costs for CO2 capture, policy and possibility for commercial looping combustion provides inherent separation of CO2
opportunities [4]. Voldsund et al. [5] provides an extensive from the other flue gas components such as N2 and unreacted
review of different technologies which can be used to provide O2, resulting in minimal energy penalty for gas separation and
CO2 capture for hydrogen production processes. no cost for gas separation equipment [19e21]. More informa-
Large scale centralised hydrogen production with CO2 tion about the development of CLC can be found in multiple
capture is considered feasible for the medium to long term published review articles [19,21e27]. The CO2 capture is
[6,7]. SMR based H2 production is considered as a suitable enabled by a so-called oxygen carrier (commonly metal oxide
target industry for early deployment of CCS, as has been particles) which can be oxidized in presence of air which takes
demonstrated by a large number of demonstration plants and place in the so called air reactor and subsequently reduced in
pilot projects under development [8,9] as presented in Table 1. presence of fuel which takes place in the air reactor. The flue
It is possible to capture CO2 at multiple locations in the gas from the fuel reactor contains only CO2 and steam where
SMR process, where the most mature route today is to capture the steam can be condensed to obtain a pure CO2 stream
CO2 from the syngas stream produced from the water gas shift which can be captured and transported for storage. The
reactors. This option can capture approximately 60% of the concept presents a highly attractive option for CO2 capture in
total CO2 emissions from the conventional process [5]. The general. A possible integration with SMR production has been
CO2 absorption unit can provide a high-purity stream of CO2 suggested by placing the reformer tubes immersed in the
with the drawback that the purity of the produced H2 (95e98%) dense bed of a fuel reactor (FR) operated as a fluidized bed heat
is lower than the hydrogen product stream obtained from a exchanger, and was first suggested by Ryde n et al. [28]. The
PSA [14]. Various CO2 capture options for SMR plants have process integration between SMR and CLC is recognized as
been investigated such as using amine-based solvents and what could be the only CO2 capture technology with an
membranes where the CO2 avoidance cost was estimated to increased efficiency (excluding the compression of CO2) when
40e85 V/ton depending on the natural gas and electricity price compared to conventional plants [21]. The only study which
[9]. Most of these options do however only include partial CO2 has presented a techno-economic assessment for the CLC-
capture (less than 90%). The only possibility to capture close to SMR plant is the work of Spallina et al. [29] where a CO2
100% of the CO2 emissions from the production is to capture avoidance cost of approximately 20 V/ton CO2 was estimated.
the CO2 in the flue gas stream from the reformer furnace. All the mentioned cases where CLC was integrated in a SMR
Among the possible points where CO2 can be captured in SMR plant include the use of natural gas as supplementary fuel.
plants, the CO2 avoidance cost for the case with post- In the previous works on CLC-SMR processes, a horizontal
combustion capture was found to be 69.8 V/ton in Collodi tube orientation has been assumed in most works [28e31]
et al. [9]. This indicates that it might be difficult to implement where Rahimpour et al. [32] is an exception. It is however
high capture rates from the SMR plants at a low additional standard to use vertical tubes in SMR plants today which is also
cost with post-combustion capture. the target in this work. A horizontal tube orientation is likely
Table 1 e SMR hydrogen production plants with CO2 capture.

Project Company Location CO2 capture capacity Capture method Fate of CO2
(Mton CO2/year)
Port Arthur [10] Air Products Port Arthur, US 1 Vacuum swing adsorption Enhanced oil recovery (EOR)
(VSA)
Quest [11] Shell Alberta, Canada 1.2 Chemical absorption with Geological storage
amines
Tomakomai [12] JCCS ltd/METI Tomakomai, Japan 0.1 Chemical absorption with Geological storage
amines
ro
Port-Je ^ me [13] Air Liquide ro
Port-Je ^ me, France 0.1 Cryogenic separation Food grade CO2
not possible because of packed catalyst would tend to accu- using biomass or NG as supplementary fuel integrated with
mulate at the bottom of the tubes due to gravity, with CLC for inherent CO2 capture.
increased risk of gas bypassing the catalyst. The reformer
tubes traditionally hang in vertical tube rows to allow for good
control of the thermal expansion which can be seen as an Plant description
additional reason for the vertical orientation. This study also
includes environmentally benign oxygen carriers for chemical- The size of the two plants is identical to the previously modelled
looping combustion suitable for large-scale operation and SMR plants in Part 1 of this study [1], producing 100 000 m3n; H2 /h,
evaluates the possibility to place the reformer tubes in the air but which did not include carbon capture. The system boundary
reactor instead of the fuel reactor. A key difference compared includes the hydrogen production plant and the CO2 compres-
to previous techno-economic studies presented by Spallina sion which is assumed to take place at the plant but excludes
et al. [29] is that an estimation of the heat transfer to the CO2 transport and storage. The overall cost is however consid-
reformer tubes is included to calculate the required bed tem- ered to be dominated by the cost of CO2 capture. Greenhouse
perature in the CLC system. This should provide a better gas emissions from process prior to the plant such as extrac-
comparison with the reference SMR plant where the heat tion, refining and transport of the fuel are not included in this
transfer to the tube surface is also included (see Stenberg et al. study. Table 2 includes some of the most important character-
- Part 1 [1]). istics of the two main process configurations.
All the mentioned processes so far rely on CO2 capture
from use of fossil fuels as fuel source (CCS) which enables SMR with chemical-looping combustion (CLC-SMR) - case
processes with close to zero if not net zero CO2 emissions. In CM
addition to CCS, carbon dioxide removal technologies (CDR)
can be used to enable net negative CO2 emissions [33]. The In Part 1 of this work [1] it was found that it is possible to
importance of CDR technologies can be motivated based on integrate SMR with fluidized bed technology where a single
the findings in a recent report from the IPCC where it is fluidized bed heat exchanger was used as heat source for
concluded that all modelled pathways to limit global warming steam reforming. The results from this study clearly showed
to 1.5 C use CDR to remove 100e1000 Gton CO2 over the 21st the benefits of increased heat transfer in increasing efficiency,
century [33]. One of the most commonly suggested measures but carbon capture was not investigated. SMR could also be
for CDR which has significant CO2 abatement potential is Bio integrated CLC technology (see Fig. 1).
Energy Carbon Capture and Storage or BECCS which relies on The CLC system consists commonly of two interconnected
CCS technology using biomass as source of carbon, thus fluidized beds and solid oxygen carriers are circulating be-
enabling net negative emissions. tween the two reactors, respectively air reactor (AR) and fuel
BECCS could be implemented also in the hydrogen pro- reactor (FR). The general reactions for a generic oxygen carrier
duction process, if the process could be partially/completely (Me) in the CLC system are presented in Eqs. (1) and (2).
based on biomass. Various alternatives to use biomass as a
feedstock for hydrogen production has been reviewed by for Fuel reactor: ð2j þkÞMex Oy þCj H2k /ð2j þkÞMex Oy1 þkH2 O
example Arregi et al. [34], Salkuyeh et al. [35], Pandey et al. þ j CO2 (1)
[36] and Shahabuddin et al. [37]. One solution which is
closely related to the suggested process here is by biomass Air reactor: ð2j þ kÞMex Oy1 þ ðj þ 0:5ÞO2 /ð2j þ kÞMex Oy (2)
gasification which could be used as a feed stream to the
Several authors have reviewed hydrogen production pro-
catalytic process in the reformer tubes. However, the need
cesses which are based on chemical-looping such as Ada nez
for gas polishing from sulphur, tar and alkali species [38,39]
et al. [19], Abad [44] and Luo et al. [45].
is still a problematic issue. In this study, we aim to consider
Among the suggested processes is the idea to integrate CLC
biomass as feedstock for the combustion process required
with SMR where the process conditions on the inside of the
to supply the heat to the natural gas reforming reactions,
reformer tubes are not altered. Case A presented in Part 1 [1]
rather than also replacing the natural gas fed to the
which is defined as a fired tubular SMR plant was used as
reformer tubes. As a result significant changes on the
reference plant to compare the new plant layouts based on
reformer tube side are avoided. This process outline leads to
CLC under the same assumptions and methodology. Just as
that the plant overall can deliver significant net negative
for case A, excess heat is used to preheat feed streams and for
emissions, since all the generated CO2 can be captured from
steam production. The first consideration for this integration
the CLC process.
is the position of the reformer tubes. The tubes could be
With biomass as fuel, the CLC process is similar to bio-CLC,
placed in the air reactor, in the fuel reactor or in a separate
which has been explored previously, e.g. Refs. [40e43]. One of
external fluidized bed heat exchanger. The vertical tubes are
the biggest challenges identified so far with this process is to
placed in the fuel reactor although previous works have sug-
achieve complete/close to complete conversion of the
gested to use both fuel reactor [28e30] and air reactor [31,32].
biomass volatiles. Flue gas cleaning is well developed for CFB
The fuel will be oxidized in the dense bed of the fuel reactor by
for biomass and it is foreseen that similar steps could be used
the oxygen carrier particles. It is therefore important to
in the bio-CLC plant despite that the gas composition is
choose a suitable oxygen carrier for the system. While the
notably different.
oxygen carrier reactivity should not limit the performance of
The aim of this work is to perform an evaluation of the
the technology as discussed in Abad et al. [44], a high chemical
techno-economic and environmental performance of SMR
Table 2 e List of the most important differences in the modelling of case CM and CB compared to the reference case
presented in Part 1.
Item Plant configuration
Reference plant NG-CLC-SMR Bio-CLC-SMR
Case A Case CM Case CB
Feed Natural gas Natural gas Natural gas
Supplementary fuel Natural gas Natural gas Biomass
CO2 capture No Yes Yes
Oxygen carrier No C28 (CaMn0.775Mg0.1Ti0.125O3-d) Manganese ore
Oxygen transport (%) e 1.25 1.25
Heat of reaction for oxidation of OC (kJ/mol O2) 325 314
Steam-to-carbon ratio 2.7 2.7 4.5
Air-to-Fuel ratio in furnace/air reactor 1.09 1.1 [66] 1.1 [66]
Oxygen demand based on present volatiles (%) e e 10%
Air-to-Fuel ratio in oxygen polishing unit e e 1.05
HT WGS inlet temperature (C) 330 330 330
LT WGS inlet temperature (C) - e 205
Air blowers
Isentropic efficiency (%) [67] e 85 85
Mechanical efficiency [67] e 95 95
CO2 compression and purification
CO2 compressor stages e 3 3
Polytropic efficiency for stage 1/2/3 (%) [67] e 80/80/75 80/80/75
Mechanical efficiency (%) [67] e 95 95
Inter-cooler outlet temperature (C) [67] e 30 30
Compressor stage discharge pressures (bar) [67] e 4.4/18.7/80 4.4/18.7/80
CO2 pump efficiency (%) [67] e 75 75
CO2 pump driver efficiency (%) [67] e 95 95
Properties of delivered CO2 e 110 bar, 30 C 110 bar, 30 C
Air separation unit (ASU)
Oxygen purity (mol-%) [67,68] e e 95
ASU electricity consumption (kWh/ton O2) [69,70] e e 200
Fig. 1 e Simplified illustration of case CM where PSA off-gas and NG are used as fuel in the fuel reactor (FR).
and mechanical stability since is required since the solid is Part 1 as C28 (CaMn0.775Mg0.1Ti0.125O3-d) was chosen as a suit-
continuously circulated between fuel and air reactors. For this able bed material due to high reactivity with both methane/
purpose a manganese-based perovskite material, presented in natural gas [46,47] and biomass [48]. C28 has been amply
studied in Refs. [47,49e51] where the development process of is used as the FBHE, the fuel reactor will be operated as a fast
this material is summarized by Moldenhauer et al. [47]. The fluidized bed reactor and provide circulation of the bed material.
process flowsheet for case CM can be observed in Fig. 2. The One potential obstacle with this approach is that the required
SMR process can be observed by following the black arrows particle flux could potentially be higher than what could easily
where NG used as feed leads to syngas production (see green be achieved with the FR, which is dimensioned based on the gas
arrows) and in the end produced hydrogen. The feed streams flow of fuel gas (PSA off-gas and supplementary fuel).
to the CLC system are indicated in black whereas the hot Secondly, the choice of oxygen carrier needs to be
stream with oxygen depleted air from the AR is in purple. The considered. Low cost naturally occurring ores or industrial
flue gas stream from the FR is indicated in red from which the waste materials could be a better choice than synthetic ma-
CO2 is captured. The scheme also includes thermal integration terials [23,53,54], to compensate lower lifetime due to ash and
with the steam cycle (indicated with blue arrows). The scheme solid fuel impurities. Several natural ores were investigated by
does not include cyclones for gas-solid separation. Fossdal et al. [55] where a manganese ore was picked as the
best option for CLC application. Manganese ores contain
SMR with chemical-looping combustion with biomass (CLC-
SMR) - case CB
The second configuration is also based on chemical-looping

combustion, but the supplementary fuel source is instead
solid biomass. The idea is to substitute this fossil fuel supply
to the process with a fuel which is considered to be carbon
neutral fuel resource which will enable net negative emissions
(see Fig. 3). The captured CO2 originating from the carbon in
the biomass represents the amount which is considered as a
negative emission.
Compared to case CM there are several differences. Firstly,
the fluidized bed heat exchanger (FBHE) is placed in the air
reactor because the presence of biomass in the fuel reactor can
lead to the formation of ash melts (slagging). Chlorine and other
elements present in the biomass can also contribute to a cor-
rosive environment under reducing conditions [52]. These ef-
fects could potentially damage the FBHE tubes and reduce the
heat transfer coefficient. It should be acknowledged that ash
components such as K and Ca or others could possibly be
transferred to the AR with the bed material and released here. Fig. 3 e Simplified illustration of case CB where PSA off-gas
However, this would at most be a minor problem similar to what and biomass are completely burnt in the FR.
occurs in CFB plants for steam generation. When the air reactor
Fig. 2 e Schematic illustration of case CM where the reformer is integrated with a FBHE unit in the FR of a CLC system.
typically 30e60% manganese in oxide form [54]. The material be operated as a circulating fluidized bed (CFB). An oxygen
is cheap, environmentally friendly, abundant and manganese polishing unit fed with pure oxygen is used to ensure com-
oxide, which can be expected to make up a significant part of plete fuel conversion downstream of the FR.
the Mn ore, has a high oxygen carrier capacity (7e10 wt-% Many kinds of biomass should be able to use in this pro-
depending on temperature) in relation to iron for example cess. A basic feature should however be that the fuel is
(3.3 wt-%) [56,57]. Manganese ore has shown to have high gas affordable and available in large quantities. Examples of
conversion rates for syngas in CLC systems (95e100%) possible fuels are for example wood chips, GROT (tops,
[53,58,59] and up to 80% with methane [58]. In operation with branches and stumps) or other harvested forest residues
biomass in CFB units today where silica sand is used as bed which cannot be used by the pulp & paper industry and
material the temperature is limited to below 900 C, mainly sawdust, a by-product of sawmills industry. These fuels are
due to formation alkali silicates which tend to become sticky commonly used in combined heat and power plants and are
thus causing the bed material to adhere together and cause considered to be suitable for this process.
agglomeration [60]. This issue is however not expected with In order to obtain a CO2-rich gas stream suitable for pipe-
bed materials as for example manganese ore or ilmenite, line transportation, flue gas cleaning is required. In case CM it
which have been shown to have less agglomeration ten- is assumed that no flue gas cleaning is required, and that the
dencies with biomass at high bed temperatures [57]. An CO2 is ready for transportation after condensing the steam
overview of challenges with fluidized bed combustion fuelled and CO2 compression. In case CB, oxygen polishing and a
with biomass is presented by Khan et al. [61]. conventional biomass-fired CFB flue gas cleaning system
Thirdly, in order to further reduce the CO2 emissions, this including feeding lime and active carbon for removal of water-
process is designed to have a higher H2 yield than the refer- soluble gases (including SO2, HCl, HF, NH3) as well as bag
ence process to increase the NG-to-H2 yield. This increases the house filters to remove fine particulate matter are considered.
biomass thermal input to the process. The two measures The resulting process outline can be observed in Fig. 4. The
chosen in this work was to use a higher steam-to-carbon ratio scheme is similar to the one presented for case CM in Fig. 2
in the reformer compared to the plant in case CM and a LT where the main differences being the addition of oxygen
WGS reactor is added. polishing, ASU, LT WGS reactor and minor changes in the
Fourthly, it is a challenge to ensure complete conversion of steam cycle. The scheme does not include flue gas cleaning
the biomass in the fuel reactor. In order to achieve this suc- steps other than oxygen polishing.
cessful biomass char gasification and volatiles combustion
has to take place. Since biomass contains a lot of volatiles
which, when released can have poor contact with the oxygen Model description
carrier [23,62] and result in unburnt fuel leaving the FR. This in
turn increases the required thermal input to the FR as well as Aspen Plus model design
the need for downstream conditioning.
Among the possible solutions to achieve this, several The two processes were studied using a similar approach as
different processes could be considered such as using in-bed described in part 1 of this study [1] in regard to the process
feeding, Chemical looping with oxygen uncoupling (CLOU) simulation in Aspen Plus and differences in the overall model
materials [63,64], recycling of flue gas [65] and multi-stage fuel are described in this section. Assumptions related to both
reactors [65]. This work however does not focus on specifically steam reforming process, strategy for thermal integration,
addressing this challenge since it is focused on the overall steam cycle, pressure drops and compression of the H2 prod-
techno-economic assessment. Therefore, the FR is assumed to uct are the same as in Part 1. Heat surplus in the processes is
Fig. 4 e Schematic illustration of case CB where the reformer is integrated with a FBHE unit in the AR of a CLC system.
used for preheating the feed stream to the reformer, HP steam the same. All units with chemical reactions (fluidized beds,
production as well as some additional LP steam production WGS reactors, oxygen polishing unit etc.) are modelled as
was included in case CB. The choice to also include additional RGibbs equilibrium reactors apart from the furnace in case CB
LP steam production in case CB resulted from the fact that this which is modelled as a RStoich reactor.
case operates with a higher S/C ratio which in turn leads to The delivered CO2 stream has 110 bar pressure and a
more heat available in the syngas and flue gas streams. This maximum temperature of 30 C [16,67,73] which is suitable for
heat is recovered in the steam cycle thus resulting in a higher transportation by pipeline. The CO2 compression is carried out
feedwater demand. The key differences in the model as- as suggested by the EBTF-report serving as a benchmark for
sumptions for case CM and CB compared to case A are pre- evaluation of CO2 capture technologies [67].
sented in Table 2. The biomass included in the process was based on the ul-
The outlet temperature of the FBHE units is defined based on timate analysis of wood chips presented by Hannula et al. [74]
the estimation of the heat transfer coefficient from bed to tube (see Table A3 in Appendix). Based on this data an equivalent
in the same manner as for case OM in Part 1 of this work. Both biomass composition was estimated which represents the
air reactor (AR) and fuel reactor (FR) were modelled as RGibbs fuel composition of the biomass used in the Aspen Plus model
reactors where the heat required for steam reforming was where both mass and energy balance was conserved.
withdrawn from FR in case CM and from AR in case CB. The air In addition to case CM and CB, two alternative configura-
to fuel ratio is based on the amount of reduced oxygen carrier tions of these processes (one for each case) were considered in
fed to the air reactor. The relation between oxidized oxygen this work. For case CM an alternative placement of the FBHE in
carrier and the amount of fuel fed to the fuel reactor is based on the AR instead of in the FR was attempted in CM2. For case CB
having an oxygen transport of 1.25 wt-% to ensure long-term it was considered interesting to study the effect of operating
durability. The oxygen transport represents the mass of oxy- with process parameters similar to case CM with a lower S/C
gen consumed in the FR in relation to the total mass of oxygen ratio (2.7) and no LT WGS reactor resulting in a lower hydrogen
carrier at the outlet of the AR expressed as a percentage. yield (case CB2).
The models for case CM and CB both include oxygen car-
riers which are defined as conventional solid components in Reactor design
ASPEN PLUS. In order to model the reactions of the oxygen
carrier, the heat of reaction in the oxidation and reduction The design of the FBHE units is carried out in the same
reactions in AR and FR belong to the key characteristics to approach as presented for case OM in Part 1 [1] as well as the
include. In the Aspen Plus model 20 mol-% nickel oxide (NiO) estimation of the number of reformer tubes where all cases
is used on 80 mol-% inert magnesium aluminate (MgAl2O4) are operating with the same residence time of gas in the tubes
support material. The use of NiO here was for modelling as in Part 1. The aim with the FBHE reactor design is to present
purpose only. Differences in heat of reaction for nickel oxide a design which presents a reasonable flue gas temperature
and the oxygen carrier materials C28 (used in case CM) and and supplies the required heat to the steam reforming. The
manganese ore (used in case CB) were compensated for and flue gas temperature in the reformer furnace is for all cases
included in the Aspen Plus model [56]. calculated based on the procedure presented in Part 1 [1]. Fig. 5
The biomass char is modelled as graphite is therefore in Part 1 [1] explains the calculation procedure where an
assumed to be combusted completely in the FR. Biomass char estimation of the bed-ot-tube heat transfer for each case
could be transferred to the air reactor (and burnt there) or lost makes up an essential part.
with the fuel reactor exhaust gas but this risk is neglected In the FBHE, an even number of tube passes for the vertical
since biomass char is more reactive than other char and that reformer tube bundle is assumed. A ratio of 2:1 between width
sufficient residence times should be reached [71]. It was and depth was assumed for both FBHE and CFB. The used
assumed that 10% of the volatiles entering the fuel reactor oxygen carrier in the FBHE was assumed to have the same
(CH4, CO and H2) remain unconverted and therefore converted properties as presented in Part 1 with a mean particle size of
downstream. This assumption is close to the recently pub- 180 mm and a bulk density of 1600 kg/m3. The CFB units are
lished experimental results for various biomass-fuelled CLC assumed to use a superficial gas velocity of 5 m/s which is
[72]. Steam released from the biomass drying and CO2-rich limited by the risk of furnace erosion for these units [75]
PSA off-gas (approximately 67 vol-% CO2) is assumed to act as whereas the FBHE reactors are operated with a superficial gas
gasification agents. Additional gasification gas could be pro- velocity of 0.5 m/s. The gas velocities are estimated based on
vided based on the large quantity of LP steam which is the operating temperature and pressure for these units where
generated at the plant. the pressure is close to atmospheric conditions in the fluid-
The required oxygen could be fed to a separate unit or be ized bed units. The FBHE is as previously mentioned placed in
included in the FR cyclone where good gas-solid contact can the FR in case CM and in the AR in case CM2.
be expected. The oxygen polishing unit (post-oxidation Based on the most commonly used oxygen carriers and
chamber) is modelled as an adiabatic RGibbs reactor placed fuels in CLC systems the heat of reaction is always more
directly after the fuel reactor operating with 5% excess oxy- exothermic in the AR compared to the FR, resulting in a higher
gen. The air separation unit (ASU) is modelled according to temperature in the AR. This argues for placing the FBHE in the
literature data [69,70]. It should be mentioned that the elec- AR based on a larger temperature driving force for heat ex-
tricity consumption could be higher. change. There are however several reasons for placing the
For both case CM and CB, the Aspen Plus model for the FBHE in the FR instead. Firstly, the volume of gas fed to the AR
reforming as in the two cases presented in Part 1 are generally roughly two times higher than the gas flow to the FR which
additions related to case CB in particular which mainly results

from the use of biomass as supplementary fuel in the process
(see Fig. 5).
The hydrogen production efficiency for case CB is esti-
mated based on Eq. (3).
m_ H2 ,LHVH2
hH2 ¼ (3)
m_ biomass ,LHVbiomass þ m_ NG;tot ,LHVNG
The equivalent hydrogen production efficiency is esti-

mated for case CB based on Eqs. (4) and (5) where the equiv-
alent mass flow of natural gas is estimated considering the
biomass input to the system. The increased difficulty to
handle biomass as a fuel is incorporated into other model
Fig. 5 e Mass and energy streams included within the
assumptions.
system boundary in case CM and CB.
m_ biomass ,LHVbiomass
m_ NG;eq;bio ¼ (4)
LHVNG
means that total cross-section for the system can be reduced

Qth Wel
by placing the FBHE, where the gas velocity is lower, in the m_ NG;eq ¼ m_ NG;feed þ m_ NG;eq;bio
hth;reference ,LHVNG hel;reference ,LHVNG
unit with the smaller gas flow. Secondly, the heat transfer is
(5)
expected to be higher in the FR [76] based on that the flue gas
has a higher thermal conductivity based on a higher steam The heat delivered for steam export, the hydrogen yield
content. Thirdly, the amount of solid material which can be and the equivalent hydrogen production efficiencies are
circulated through the riser is limited by the available cross evaluated in the same way as in Part 1.
section which would be smaller in the FR since it is based on Upstream CO2 emissions for producing the fuels to case CM
the gas flow. In summary, the best placement is based on this (natural gas) and case CB (biomass) are not considered in this
considered to be in the FR in case CM but results for the estimation. In case CB, the captured CO2 from the biomass
opposite design are also presented in the results section (see combustion is a net negative emission.
case CM2).
In this work it was assumed that the particle flux in the Economic assessment
CFB unit should not exceed 50 kg/(m2s) which was observed
in a 12 MW CFB boiler with silica sand with a mean diameter The approach for the cost assessment is provided in Part 1
of 0.32 mm as bed material [77]. More recent experiments where the levelized cost of hydrogen (LCOH) is calculated in
with ilmenite as bed material [78] with a mean particle the same manner. LCOH includes both capital costs for pro-
diameter of around 0.2 mm in the same CFB boiler displayed cess equipment as well as operational costs. The same Capital
that particle fluxes of around 62 kg/(m2s) operating with a gas Charge Factor (CCF) was used for these plants and the same
velocity of around 3 m/s in the furnace but in order to make a Bottom-Up Approach is used as starting point for the plant
more conservative estimate the limit in this model is set to cost estimation. The conversion rate from USD to V was
50 kg/(m2s). assumed to be 0.90 while the rate was 0.11 from SEK to V.
The height of CFB reactors in these plants were estimated Additional data required for the cost estimation for case CM
based on the relationship of ratio between height and cross and CB is presented in Tables 3 and 4.
sectional area in a suggested 1000 MWth CLC plant [71]. The In addition to the units presented in Table 3, a LT WGS
bed inventory in fast fluidized CFB unit was estimated by reactor is used in case CB. The cost of the LT WGS reactor is
assuming a settled (unfluidized) bed height of 0.8 m [71]. estimated to be four times as expensive as the cost of the HT
The reactor design of the oxygen polishing (OP) reactor is WGS reactor. This is based on that the required residence
mainly based on the work of Lyngfelt & Leckner [71] where a time of the gas in the LT WGS reactor is expected to be 3e5
design of a large scale CLC plant is presented. The residence times longer than in the HT WGS reactor [81] which means
time in the OP reactor is set to 0.5 s and the diameter of the that the required volume should be 3e5 larger. The reactor
reactor is assumed as 4 m. Based on these assumptions the cost is expected to be directly related to the reactor volume
length of the OP reactor is determined which differs slightly and it is therefore assumed that the LT WGS reactor cost is
between case CB and CB2. equal to four times the cost of the HT WGS reactor (cost
All the other equipment has been considered using the estimation for this reactor is presented in Part 1). In the
same methods as presented in Part 1 of this work [1] including estimation of the fixed costs for the LT WGS catalyst it is
air and fuel gas pre-heating. assumed that the price per unit volume is the same but that
four times as much catalyst as a result of having a larger
Thermodynamic assessment volume.
The cost of the oxygen polishing reactor was estimated
The performance indicators for case CM and CB are similar to based on the calculated dimensions (see section Reactor
the indicators used in Part 1 (see Table 3 [1]) but there are some design) and assuming that the cost of the reactor can be
Table 3 e Cost calculation assumptions for the units not included in Part 1.
Equipment unit Scaling parameter Ref. capacity, So Ref erected cost (MV) Scale factor Cost year
Cyclone Volumetric flow [m3std =sec] 47.74 2.24 [79] 0.6 2002
CO2 compressor Compressor power [MW] 13 9.95 [67] 0.67 2008
Air separation unit (ASU) Oxygen produced (kg/s) 28.9 26.6 [80] 0.7 2008
oxygen carrier bed material was instead estimated based on

Table 4 e List of Process O&M costs not included in Part 1.
the presence of ash-forming elements (although none such
Fixed O&M costs species are assumed in the Aspen Plus model of case CB). As a
Labour costs, including CO2 capture [84] 1.8 MV result, the regeneration rate (need to replace with fresh oxy-
Variable O&M costs
gen carrier material) is based on the thermal input of biomass
Cost NG (Sweden) [85] 11.67 V/GJ
Cost biomass (Sweden) [82] 183 SEK/MWh
fed to the system and 3 kg/MWth was assumed to be a suitable
Flue gas cleaning regeneration rate for biomass based on industrial practise [87].
Lime consumption [83] 75 kg/h The result of this estimation shows that the lifetime of the bed
Cost of lime 250 V/ton material in case CB and case CM are similar (8732 h and 9000 h
Active carbon consumption [83] 7.5 kg/h respectively). Part of the reason for this can be traced to the
Cost of active carbon 1000 V/ton
fact that only part of the thermal input to the fuel reactor
Oxygen carrier (OC)
consists of biomass.
Cost manganese ore [71] 225 V/ton
Lifetime of C28 in case CM [86] 9000 hours The cost of refractory has only been considered for the
Regeneration rate of oxygen 3 kg/MWth fluidized bed reactors, cyclones and the oxygen polishing
carrier in case CB [87] reactor. The assumed lifetime of the reformer tubes is
Estimated lifetime of OC in case CB 8732 hours 100 000 h for both case CM and CB which is the same as in case
Cost of landfill [88] 70.6 SEK/ton used OC OM in Part 1.
estimated in a similar way as the fluidized bed units (see Results and discussion
additional information in Part 1).
The CO2 avoidance cost (CCA) was estimated according to Thermodynamic evaluation
Eq. (7) and represents the added cost to capture CO2 for a
certain case (i) compared to a reference plant with no capture The composite curves for case CM and CB are included in Fig. 6
which is represented by case A in Part 1 [1]. The cost is pre- and Fig. 7 respectively where the sections for the flue gas,
sented in the unit V/ton CO2. syngas and oxygen depleted air streams are included.
Several differences can be observed between case CM
LCOHi LCOHcase A
CCA ¼ (7) and CB when consulting Figs. 6 and 7. The flue gas section
ECO2 ; case A ECO2 ;i
looks similar overall although the heat load is overall larger
Just as in Part 1, the economic value of LP steam has been for case CB resulting in more excess heat generation. This
neglected in this analysis since there is generally little use of can be explained by that the overall thermal input to the
this low-value heat source at a refinery. The biomass cost is furnace is higher in case CB which in turn can be explained
assumed to be 183 SEK/MWh (approx. 20 $/MWh) which was by that the heating value of the biomass is lower than for
the average price for wood chips during 2015 for industries in the natural gas and that the fuel conversion in the FR is
Sweden [82]. Feedstock preparation costs and the handling incomplete in case CB.
costs are not included in this analysis. In the syngas section more heat is available in the syngas
The cost of the biomass derived flue gas cleaning was stream in case CB compared to CM mainly as a result of a higher
estimated using industrial data from Valmet based on the S/C ratio. The section with the oxygen depleted air is similar for
targeted fuel and assumed flue gas flow to approximately case CM and CB where the outlet temperature is higher in case
1.7 MV [83] in case CB and CB2. This contact was also used to CM. The amount of LP steam to export is equal to 7.6 kg/s in
estimate some of the variable costs including necessary case CM and 9.2 kg/s in case CB which are still significantly
chemicals for continuous removal of contaminants. The lower than in the conventional plant (see case A in Part 1).
electricity consumption for the baghouse filter and the feeding The results from the thermodynamic evaluation of case
system for the lime and active carbon were estimated to be CM and CB including the thermal efficiency to produce
negligible and as a result not included in the variable costs. hydrogen and the CO2 emissions from the plants are pre-
Other relevant O&M costs are found in Part 1 and presented in sented in Table 5 together with the result for case CM2 and
Table 4. CB2 and the two cases from Part 1 which are included for
The lifetime of the oxygen carrier in case CM is estimated comparison.
to the same value as was assumed in case OM which was CO2 compression adds a significant electricity consump-
based on loss of fines from operation with a 10 kWth CLC unit tion in the cases with CO2 capture. The compression work is
for natural gas [89]. The cost of C28 is the same as presented in highest for case CB and CB2 since additional CO2 is com-
Part 1 of this work, 1225 V/ton. For case CB the lifetime of the pressed resulting overall in electricity demand in case CM,
Fig. 6 e Hot and cold composite curves for the syngas system, flue gas and oxygen depleted air section in the plant where
SMR is integrated with CLC with NG as supplementary fuel (case CM).
Fig. 7 e Hot and cold composite curves for the syngas system, flue gas and oxygen depleted air section in the plant where
SMR is integrated with CLC with biomass as supplementary fuel (case CB).
CM2, CB and CB2. The electricity consumption of the ASU adds balance over the fuel reactor is overall endothermic. In case
additional electricity consumption for case CB and CB2. CB the net heat of reaction is exothermic in both FR and AR.
In case CB the H2 yield is significantly higher compared to It can be observed that the off-gas stream is the same in all
the other cases as a result of a higher S/C ratio and the use of a cases apart from case CB where a higher hydrogen yield is
LT WGS reactor which results in that the NG feed stream can achieved resulting in less combustible compounds such as
be reduced while still producing same hydrogen amount. This CH4 and CO in the off-gas which is compensated by adding
is also the reason why the furnace duty transferred to the 5.9 kg/s of biomass.
reformer tubes is slightly smaller in case CB although there is In terms of CO2 emissions, case CM has no specific CO2
more inert steam in the tubes and that more endothermic emissions (100% capture) and the equivalent CO2 emissions
steam reforming reaction takes place. are negative for the process if the excess heat can be utilized.
When comparing the NG consumption of case OM and CM The CO2 emissions are in this case considered as negative
it can be observed that the consumption increases when since it assumes that the utilization of the excess heat means
adding CO2 capture, but it is still clearly lower than for case A. replacing heat which would otherwise have resulted in CO2
The difference is small between CM and CM2 and the con- emissions. In case CB the specific CO2 emissions from the
sumption is significantly lower in case CB where no NG is used plant are net negative (without considering possible use of
as supplementary fuel and the NG feed stream is reduced. excess heat nor electricity) and resulting in emissions which
The heat balance over the FR in case CM considering the are 0.29 Mton CO2/year which is 142% lower than for the
fuel oxidation and the OC reduction indicates that the net heat benchmark plant (Case A).
of reaction should be exothermic but since a significant When consulting the presented hydrogen production effi-
amount of heat is also extracted to steam reforming the heat ciencies in Table 5 it can be observed that the energy penalty for
Table 5 e Energy streams and performance parameters of the plant configurations (positive numbers imply production and
negative numbers imply consumption).
Case A Case OM Case CM Case CM2 Case CB Case CB2
Electricity production/consumption
Air blowers MWel 1.0 0.6 2.2 2.4 2.3
H2 compressors MWel 3.8 3.8 3.8 3.8 3.8 3.8
CO2 compressors MWel 8.8 8.8 11.5 10.3
Steam turbines MWel 16.9 8.0 10.2 10.2 12.6 13.3
Pumps MWel 0.8 0.5 0.6 0.6 0.8 0.7
Pumps for heat rejection MWel 0.1 0.2 0.4 0.4 0.6 0.4
Air separation unit MWel 0.7 0.8
Net electric power MWel 12.3 2.6 4.1 5.6 7.3 5.3
Heat production/consumption
Total cooling requirement MWheat 12.3 22.1 55.6 55.3 73.4 54.3
Steam export (160 C, 6 bar) kg/s 18.8 4.1 7.6 7.7 9.2 13.2
Furnace duty transferred to reformer tubes MW 97.9 97.9 97.9 97.9 96.0 97.9
Efficiency and emission data
Hydrogen yield after reformer, YH2,ref % 68.1 68.1 68.1 68.1 79.8 68.1
Hydrogen yield after WGS, YH2,WGS % 78.5 78.5 78.5 78.5 93.1 78.5
Total NG consumption m_ NG;tot kg/s 9.10 8.15 8.45 8.43 6.43 7.64
Thermal input off-gas MWLHV 113.5 113.5 113.5 113.5 54.3 113.5
Thermal input supplementary fuel MWLHV 68.0 23.9 37.8 36.6 115.4 53.0
Total thermal input to furnace MWLHV 181.4 137.3 151.2 150.1 169.7 166.5
H2 production efficiency hH2 % 70.7 79.0 76.2 76.4 72.2 73.3
H2 production efficiency hH2;eq % 83.9 82.1 78.4 78.2 73.9 78.0
H2 production efficiency h0eq;H2 % 75.8 80.2 75.4 75.1 70.1 72.0
CO2 specific emissions ECO2 gCO2 /MJH2 produced 80.7 72.2 0 0 34.1 15.7
Equivalent CO2 specific emissions ECO2 ;eq gCO2 /MJH2 produced 68.0 69.5 2.1 1.7 35.9 20.0
Equivalent CO2 specific emissions E0CO2 ;eq gCO2 /MJH2 produced 75.3 71.1 0.8 1.3 32.3 14.9
Total plant CO2 emissions Mton CO2/year 0.69 0.61 0 0 0.29 0.13
adding CO2 capture is limited when comparing case OM and correction factor closer to 1 as a result of less dense tube
case CM. The main reasons for the energy penalty are a higher spacing compared to case OM and CM. It can be observed that
air consumption as a result of a higher AFR, air reactor being the estimated flue gas temperatures from the reformer
heated to a higher temperature than the FBHE in case OM and a furnace is significantly lower in the cases with FBHE as a result
lower fuel gas inlet temperature (400 C vs 266 C). In this regard of the high heat transfer in the fluidized beds. CB2 has almost
it is also interesting to point out that the placement of the FBHE identical heat transfer coefficients as in case CB where the
seems to have a limited effect on the hydrogen production ef- minor differences in gas composition in the AR flue gas has
ficiency since case CM2 has only a slightly higher efficiency. almost no effect on the estimated heat transfer coefficient on
Despite the mentioned energy penalties, the hydrogen produc- the outside of the reformer tubes when comparing case CB
tion efficiency hH2 is 7.7e8.0% higher in case CM and CM2 than and CB2.
for the reference SMR plant. It can be observed that case CB has The estimated temperature levels for the dual-fluidized
a lower thermodynamic efficiency than case CM. This is mainly systems are considered to be suitable for combustion of both
explained by the incomplete fuel conversion in the fuel reactor gaseous and solid fuel in the FR based on previous operational
and a higher S/C ratio. The equivalent hydrogen production ef- experience [23]. The design specifications for the reformer
ficiency is lower for case CM, CM2, CB and CB2 compared to case furnaces in case CM, CM2 and CB are presented in Table 7
A and OM which to a large extent can be explained by the sig- together with the values for the cases presented in Part 1.
nificant electricity consumption for CO2 compression. It can be observed that the FBHE reactor size is of the same
order of magnitude for case OM presented in Part 1. The di-
Reactors design mensions are however different as a result of different gas
flows. The gas flow fed to the FBHE in case CM is approxi-
The results of the heat transfer calculations required for FBHE mately 25% of the gas flow passing through the FBHE in case
reactor design for case CM, CM2 and CB are presented in Table OM. In order to maintain a tube pitch above two tube di-
6 as well as the corresponding values for reformer furnaces in ameters (2Dto) for this case it is therefore necessary to use 2
the two cases presented in Part 1. tube passes instead of 4 as in case OM. This results in a higher
It can be observed in Table 6 that all cases with FBHE units bed height and a higher pressure drop over the bed. This
display high heat transfer coefficients although some differ- pressure drop should be acceptable since the PSA off-gas is
ences can be seen. The radiative heat transfer coefficient is assumed to be delivered at a pressure above 1.5 bar. The ob-
similar for the different fluidized bed heat exchangers. In the tained particle flux in the CFB is below the set limit of 50 kg/
estimation of the corrected bed-to-tube heat transfer coeffi- (m2s) for case CM. The fact that the cross-sectional area of the
cient it can be observed that case CM2, CB and CB2 has a FBHE is significantly smaller in case CM compared to case OM,
Table 6 e Heat transfer calculations in the reformer furnaces.

SMR SMR-OCAC SMR-CLC SMR-CLC Bio-CLC-SMR (case CB) Bio-CLC-SMR
(Case A) (Case OM) (case CM) (case CM2) (case CB2)
Bed voidage e 0.67 0.67 0.67 0.67 0.67
Heat transfer coefficient
Inside (W/(m2K)) 1384 1384 1384 1384 1296 1384
Wall (W/(m2K)) 2872 2872 2872 2872 2837 2872
Outside
Convective (W/(m2K)) 0 497 598 432 429 432
Radiative (W/(m2K)) 138 307 305 307 295 307
Total outside without e 804 902 739 724 739
correction for tube bundle
(W/(m2K))
Tube correction factor e 0.83 0.85 0.93 0.92 0.93
(W/(m2K))
Outside with correction 138 671 764 686 665 686
(W/(m2K))
Logarithmic mean 450 144 134 142 122 142

temperature difference
(C)
Overall heat transfer coefficient 118 370 396 374 359 342
(Uo) (W/(m2K))
Required heat transfer area (m2) 1841 1841 1841 1841 2192 1841
Flue gas temperature (C) 1201 935 929 934 922 934
Table 7 e Design results for the reformers (GFF and FBHE).

Case A OM CM CM2 CB CB2
Number of reformer tubes 428 428 428 428 511 428
GFF/FBHE
Width (m) 8.7 30.0 16.0 24.3 24.9 24.5
Depth (m) 4.5 15.0 8.0 12.2 12.4 12.2
Height (m) 11.4 3.4 6.8 6.8 6.8 6.8
Bed height (m) e 2.85 5.7 5.7 5.7 5.7
Tube passes 1 4 2 2 2 2
Pressure drop (bar) e 0.25 0.5 0.5 0.5 0.5
CFB-unit
Width (m) e e 8.0 3.7 6.4 6.1
Depth (m) e e 4.0 1.8 3.2 3.1
Height (m) e e 11.0 2.3 7.0 6.6
Pressure drop (bar) 0.1 0.1 0.1 0.1
Particle flux (kg/(m2s)) 29 133 46 47
Oxygen carrier inventory
FBHE (ton) e 1943 1081 2615 2712 2648
kg/MWth e 19 854 7150 26 718 28 263 27 051
CFB (ton) e e 41 9 26 24
kg/MWth e e 352 47 149 146
Circulation rate (kg/s) e e 908 889 936 900
for example, should be seen as an advantage of this process similar to case CB despite that more reformer tubes are pre-
since space at the plant may be limited. A result of smaller sent in the latter (which is a result of larger gas flow), with very
reactor dimensions also results in a less OC inventory in case limited effect on the reactor dimensions and required oxygen
CM compared to case OM. carrier inventory.
In case CM2, CB and CB2 the cross-section of the FBHE is The CFB unit in CM2 is significantly smaller (96% in terms
larger compared to case CM mainly as a result of the place- of volume) than in case CM since the gas flow to the CFB is
ment in the air reactor where the gas flow is larger. Despite the smaller. The resulting particle flux is however higher than the
larger gas flow, it is not possible to use 4 tube passes but maximum value considered and a possible bottleneck for case
instead 2 tube passes are used for case CM2, CB and CB2. The CM2 has been identified.
estimated pressure drop for the three mentioned units are It could be possible to operate with a higher degree of
therefore the same as for case CM. Case CB2 is overall very reduction of the oxygen carrier than assumed in this work but
it may lead to fragmentation. This could for example be also more expensive since these are operated with a higher
possible by using an oxygen carrier with higher oxygen carrier CO2 flow rate. It should also be said that some flue gas
capacity (e.g. Cu-based material). This would then result in a cleaning is most likely needed also for all the other systems
lower circulation rate. A separate riser could also be used to (which has not been included in this work), but the result of
provide circulation as proposed in 10e100 kW units [90,91]. In this work should still provide a good comparison between
that case additional gas could be used to circulate part of the the requirements of case CB in relation to the other cases.
particles to maintain a particle flux below the limit in the CFB The variable costs however are actually the lowest for this
riser. case despite that this case has the highest cost for electricity
The specific bed inventory for the different FBHE units are import (3.4 MV/y). The main reason is the lower cost of the
similar to the value in case OM. For the air reactors it is in line fuel feedstock (NG þ biomass) justified by the higher H2 yield
with what is observed in larger CLC pilots today [23] whereas as explained in the section before (23 MV/year compared to
the FBHE has a significantly higher value, but this is reason- case A). The lower cost for oxygen carrier make up could also
able considering the purpose of the unit. justify the use of a synthetic and more expensive material if
this would benefit the overall performance and operation
Economic assessment strategies.
It should in this regard be accentuated that the price of
The results from the economic analysis are documented in the biomass has a high impact on the LCOH for case CB
Table 8 which includes individual component costs, fixed and (here assumed to be roughly half the price of the NG). The
variable costs which together add up to the levelized hydrogen effect of the biomass fuel price is investigated further in the
production costs for all presented cases. economic sensitivity analysis presented in section
In a comparison of the costs of the reference SMR plant Economic sensitivity analysis. The cost of landfilling is in
(case A) and case CM it is possible to see that the equipment this case negligible.
cost is higher for case CM which to a large extent is depen- When comparing case CB2 with CB the most significant
dent on the cost of the CO2 compressor. Additional in- reduction in equipment cost can be traced to the absence of
vestments for air reactor, cyclone and a larger cooling water the LT WGS reactor in case CB2 while other equipment cost
system corresponding to an additional 21.1 MV/year are also such as the FBHE, and HXs are less relevant. In terms of var-
contributing in this regard. It can however be noted that the iable costs, case CB2 presents a higher fuel cost (þ9.0 MV/y)
cost for the heat exchangers is lower for case CM compared and lower oxygen carriers make-up. The cost for oxygen car-
to both reference plant and case OM. When considering the rier make-up is even lower in case CB2 which is originating
variable costs, it should be noted that oxygen carrier make- from a lower thermal input of biomass compared to case CB.
up cost is 42% lower in case CM compared to case OM as a The LCOH is fairly similar for case CB and CB2. Case CB2 could
result of less bed material use. Case CM still has a higher be seen as more interesting from an economic point of view
production cost than case OM (þ6%), but it could be said that while case CB appears as the best option from an environ-
the incremental cost for CO2 capture is small compared to mental point of view.
the possibility to reduce the NG cost compared to case A. This The cost of CO2 avoidance is negative for all but one of
results in that the levelized hydrogen production cost is the suggested cases when comparing it to the reference
lower than for the reference SMR plant and the estimated plant case A, despite that these include CO2 capture and
CO2 avoidance cost is 4.28 V/tonCO2. This option looks very compression work. In this regard, it should however be said
interesting to enable hydrogen production with cost-efficient that the cost of CO2 avoidance, with respect to case OM,
CO2 capture. show a positive cost (þ16.9e20.5 V/tonCO2) for case CM and
When comparing the equipment costs of CM and CM2, it CM2, while the CB cases are ranging between þ14.4e17.7
can be observed that the FBHE is cheaper in case CM as V/tonCO2. This results are still better than existing technol-
expected but the cost of the CFB and cyclones is higher. The ogies proposed for H2 production with CO2 capture
total cost of these two units is higher in case CM (þ1.0 MV). [16,92e95] thus underlining the incremental environmental
The cost of the flue gas HX and air blower is however more and economic advantage of a biomass co-fuelled plant.
expensive in case CM2 which results in a slightly higher There are two main reasons which explains why the CO2
equipment cost in that case. When considering the variable avoidance cost is so low for these processes:
costs, it can be seen that case CM2 has as a lower NG cost
(0.4 MV) but higher oxygen carrier make up cost (þ1.6 MV) The cost of the fuel makes up some 60e70% of the levelized
and higher electricity consumptions (þ0.7 MV). The net production cost in general so measures where the fuel
result of these differences is that case CM has a slightly consumption of NG can be limited or where cheaper fuels
lower LCOH. can be incorporated in the process such as in case CB, the
In case of CB, the overall equipment costs are significantly variable costs can be decreased significantly which can
higher (þ12.5 MV/year compared to CM) due to the addition compensate for more expensive capital investments.
of a LT WGS, a larger bigger FBHE reactor with more reformer CLC is known to provide low energy penalty and low in-
tubes as well as the ASU and flue gas cleaning. The cost of cremental costs for CO2 capture and this study highlights
HXs is also remarkably higher (þ8.6 MV) which results from this opportunity where the cost is still higher than for case
higher amount of heat exchanged and larger flowrate (as OM which is a very similar process but not including any
represented in Figs. 6 and 7). The cost of CO2 compressors is CO2 capture.
Table 8 e Cost comparison including the cases with CO2 capture and case A and OM presented in Part 1.
Hydrogen production process Case A Case OM Case CM Case CM2 Case CB Case CB2
Bare Erected Cost (MV (% of total BEC))
Desulfurization 0.71 (0.8%) 0.71 (0.8%) 0.71 (0.7%) 0.71 (0.7%) 0.64 (0.5%) 0.71 (0.6%)
Pre-reformer 6.16 (6.9%) 6.16 (6.8%) 6.16 (6.2%) 6.16 (6.2%) 5.42 (4.0%) 6.16 (5.6%)
PSA unit 20.72 (23.3%) 20.72 (23.0%) 20.72 (20.8%) 20.72 (20.8%) 20.22 (14.8%) 20.72 (18.7%)
H2 compressor 1.86 (2.1%) 1.86 (2.1%) 1.86 (1.9%) 1.86 (1.9%) 1.87 (1.4%) 1.87 (1.7%)
HT WGS reactor 5.32 (6.0%) 5.32 (5.9%) 5.32 (5.3%) 5.32 (5.3%) 4.74 (3.5%) 5.32 (4.8%)
LT WGS reactor e e e e 18.98 (13.9%) e
Steam turbine 6.23 (7.0%) 4.17 (4.6%) 4.43 (4.4%) 4.45 (4.5%) 5.64 (4.1%) 5.84 (5.3%)
Pump 0.24 (0.3%) 0.20 (0.2%) 0.22 (0.2%) 0.22 (0.2%) 0.31 (0.2%) 0.29 (0.3%)
Syngas cooling HX 10.57 (11.9%) 13.05 (14.5%) 11.62 (11.7%) 11.59 (11.6%) 12.52 (9.2%) 10.96 (9.9%)
Flue gas/oxygen depleted air cooling HX 24.95 (28.0%) 21.78 (24.1%) 18.78 (18.9%) 19.84 (19.9%) 26.64 (19.4%) 23.60 (21.3%)
Cooling water system 3.44 (3.9%) 5.10 (5.7%) 9.47 (9.5%) 9.43 (9.4%) 11.39 (8.4%) 9.32 (8.4%)
Reformer furnace/FBHE 8.82 (9.9%) 10.88 (12.1%) 9.78 (9.8%) 10.71 (10.7%) 12.30 (9.0%) 10.78 (9.7%)
Air blower e 0.28 (0.3%) 0.19 (0.2%) 0.47 (0.5%) 0.49 (0.4%) 0.47 (0.4%)
CFB e e 1.45 (1.5%) 0.41 (0.4%) 0.99 (0.7%) 0.95 (0.9%)
Cyclones e e 1.33 (1.3%) 0.48 (0.5%) 0.84 (0.6%) 0.77 (0.7%)
CO2 compressor þ pump e e 7.45 (7.5%) 7.44 (7.5%) 8.93 (6.6%) 8.29 (7.5%)
Air separation unit (ASU) e e e e 2.28 (1.7%) 2.68 (2.4%)
Oxygen polishing unit e e e e 0.50 (0.4%) 0.49 (0.4%)
Flue gas cleaning e e e e 1.70 (1.2%) 1.7 (1.5%)
Bare Erected Cost (BEC) 89.0 90.2 99.5 99.8 136.2 110.9
Total Plant Cost (TPC) 210.1 212.9 234.8 235.5 321.4 261.7
Total Plant Cost (MV/year) 30.3 30.7 33.9 34.0 46.4 37.8
O&M fixed costs (MV (% of total O&M fixed costs))
Labor costs 1.20 (8.0%) 1.20 (8.7%) 1.80 (11.8%) 1.80 (11.7%) 1.80 (8.8%) 1.80 (10.8%)
Maintenance cost 5.25 (35.0%) 5.32 (38.7%) 5.87 (38.5%) 5.89 (38.3%) 8.04 (39.3%) 6.54 (39.3%)
Insurance cost 4.20 (28.0%) 4.26 (30.9%) 4.70 (30.8%) 4.71 (30.7%) 6.43 (31.4%) 5.23 (31.5%)
HT&LT WGS catalyst 0.19 (1.2%) 0.19 (1.3%) 0.19 (1.2%) 0.19 (1.2%) 0.93 (4.6%) 0.19 (1.1%)
Reformer tube replacement 1.22 (8.1%) 0.61 (4.4%) 0.61 (4.0%) 0.61 (4.0%) 0.73 (3.5%) 0.61 (3.7%)
Pre-reformer and reformer catalyst 1.31 (8.7%) 1.31 (9.5%) 1.31 (8.6%) 1.31 (8.5%) 1.61 (7.9%) 1.31 (7.9%)
Desulfurization catalyst 0.33 (2.2%) 0.33 (2.4%) 0.33 (2.1%) 0.33 (2.1%) 0.28 (1.3%) 0.33 (2.0%)
Internal þ external insulation 1.33 (8.8%) 0.56 (4.1%) 0.45 (3.0%) 0.54 (3.5%) 0.64 (3.1%) 0.62 (3.7%)
Total O&M fixed costs (MV/year) 15.0 13.8 15.2 15.4 20.5 16.6
O&M variable costs (MV (% of total O&M variable costs))
Natural gas 140.2 (99.5%) 125.6 (97.7%) 130.2 (97.0%) 129.79 (95.3%) 99.09 (81.2%) 117.68 (90.7%)
Biomass e e e e 17.78 (14.6%) 8.16 (6.3%)
Electricity e e 1.90 (1.4%) 2.62 (1.9%) 3.40 (2.8%) 2.45 (1.9%)
Cooling water make-up 0.06 (0%) 0.10 (0.1%) 0.26 (0.2%) 0.26 (0.2%) 0.34 (0.3%) 0.25 (0.2%)
Process water 0.71 (0.5%) 0.71 (0.5%) 0.69 (0.5%) 0.71 (0.5%) 0.72 (0.6%) 0.69 (0.5%)
Oxygen carrier make-up e 2.1 (1.6%) 1.20 (0.9%) 2.82 (2.1%) 0.55 (0.5%) 0.25 (0.6%)
Cost of landfill e e e e 0.19 (0.2%) 0.09 (0.1%)
Cost of lime & active carbon e e e e 0.21 (0.2%) 0.21 (0.2%)
Total O&M variable cost (MV/year) 140.9 128.5 134.2 136.2 122.1 129.8
LCOH (V/Nm3H2 ) 0.236 0.219 0.233 0.235 0.240 0.234

LCOH (V/kgH2 ) 2.629 2.441 2.589 2.619 2.667 2.598
CO2 avoidance cost compared to case A e e 4.28 1.09 2.71 2.69
(V/tonCO2 )
CO2 avoidance cost compared to case OM 16.9 20.5 17.7 14.4
(V/tonCO2 )
Thermodynamic sensitivity analysis OM, so no variation of basic thermodynamic parameters was

considered.
A sensitivity analysis was conducted for case CB where the It can be observed that operation with a lower oxygen de-
hydrogen production efficiency and the LCOH was estimated mand results in a significant improvement in the hydrogen
when varying the oxygen demand in the fuel reactor between production efficiency and a lower LCOH despite that the costs
0 and 20% (see Table 9). This is considered to be one of the for the OP and ASU are not substantial when compared to
biggest uncertainties in the process and the effect of this was other process units. In addition, the amount of HX surface
investigated in this section. Case CM is overall similar to case area is reduced as well as the cost for steam turbines. These
stream from selling LP steam which is slightly higher in case

Table 9 e Results from the sensitivity analysis where hH2
CM. A more considerable difference when comparing the
and LCOH (V/Nm3H2 ) are estimated as well as the change in
LCOH with a higher price to emit CO2 where case CM presents
relation to case CB as a percentage (%) for varied oxygen
demand. a process with no net emissions of CO2 and therefore no
additional cost making its implementation interesting for
UD ¼ 0% UD ¼ 20%
large-scale deployment.
Oxygen demand (UD ¼ 10%) hH2 þ3.4% 3.5% Case CB is more sensitive than the other cases to the var-
74.7% 69.7%
iables affected by the capital investment. Case CB has a lower
LCOH 2.1% þ1.3%
consumption of NG which makes it less exposed to a volatile
0.235 0.243
NG price and could be used as strategy to reduce the variable
costs and consider possible limitations in biomass availability
are the main reasons for the cost reduction in the equipment and transportation. The price of the biomass in the sensitivity
cost. It should also be mentioned that the consumption of analysis is varied based on the approximate price in Sweden
biomass is reduced by almost 2 MV when no oxygen demand of recycled wood (lower range) and wood pellets (higher range)
is assumed. An increased oxygen demand results in the in 2015 [96].
opposite effect. In a recently published work of Gogolev et al. The sensitivity of the LCOH show some effect for the cost of
[48], C28 was mixed with ilmenite particles in a 100 kWth CLC oxygen carrier in case the price is 10 times higher than the one
system. This experiment resulted in an oxygen demand of considered resulting in an increase of 6.2% and 2.6% respec-
around 5% with black pellets as fuel which could be compared tively for case CM and CB configurations. An increase in price
to previous experiments with two different manganese ore of this magnitude is not expected for case CM but for case CB
which resulted in an oxygen demand of 25e30%. This kind of this could occur based on using synthesized materials instead
results indicate that depending on the performance of the of ores. The use of a different oxygen carrier should be justi-
oxygen carrier, it could be worth to choose a more expensive fied by better performance in terms of fuel conversion, life-
oxygen carrier with a higher reactivity with biomass to time and stability which would possibly compensate the
partially or completely substitute the suggested ore. marginal cost required for the oxygen carrier make up.
At present the carbon tax is only targeting CO2 from com-
Economic sensitivity analysis bustion of fossil fuels thus incentivizing emission reductions
down to zero, but no further [97]. The implementation of more
In the economic sensitivity analysis, the levelized hydrogen strict policies in terms of CO2 emissions, including negative
production cost was estimated based on varying key param- emissions, could enable the faster implementation of BECCS
eters, which do not affect the hydrogen production efficiency technologies such as the processes presented here. This could
of the processes but only the economics of the plants. The facilitate larger investments in biomass-based plant located in
results from this analysis are presented in Table 10. countries like Sweden, Canada and Norway. In this work it is for
It can be observed that the trends for case CM are similar as example seen that a carbon tax of around 13e14 V/ton CO2 is
presented for case OM in Part 1 of this study. One difference enough to present LCOH-values for case CB and CB2 which are
between case CM and OM concern the potential revenue identical to case OM which does not include any CO2 capture.
Table 10 e Parameters for the economic sensitivity analysis with studied ranges where the results from using the lowest
and highest value within the range is presented. Both estimated LCOH (V/Nm3H2 ) and price change in % in relation to the
respective base case.
Operating parameter Range LCOH for Case CM LCOH for Case CB
Low Mid High Low Mid High
Cost of FBHE including reformer tubes (% of estimated cost) 50e200 1.2% Base þ2.4% 1.5% Base þ2.9%
Base ¼ 100 0.2298 0.2381 0.2362 0.2466
Plant availability 50e100 þ21.1% Base 2.6% þ27.6% Base 3.5%
(%) Base ¼ 90 0.2817 0.2264 0.3058 0.2314
Capital Charge Factor (CCF) (%/year) 5e20 12.1% Base þ7.1% 16.1% Base þ9.5%
Base ¼ 14.4 0.2045 0.2491 0.2012 0.2623
Reformer tube lifetime 25 000e100000 þ1.0% e Base þ1.2% e Base
(hours) Base ¼ 100 000 0.2349 0.2424
Price of LP steam 0e40 Base e 4.75% Base e 5.6%
(V/ton) Base ¼ 0 0.2215 0.2263
Natural gas price 5e20 40.6% Base þ50.7% 30.0% Base þ37.4%
(V/GJ) Base ¼ 11.67 0.1382 0.3504 0.1678 0.3294
Biomass price 97e287 e e e 4.4% Base þ5.4%
(SEK/MWh) Base ¼ 183 0.2291 0.2525
Oxygen carrier price 0e1000 0.7% Base þ6.2% 0.3% Base þ2.6%
(% of assumed price) Base ¼ 100 0.2208 0.236 0.2390 0.2460
Cost to emit CO2 0e100 Base e þ0.0% Base e 15.3%
(V/ton) Base ¼ 0 0.2323 0.2029
heat transfer to tube compared to the conventional process

Conclusions which results in a possibility to lower fuel consumption
which makes up a significant part of the levelized cost, ii) a
Steam methane reforming (SMR) was integrated with chemical- limited incremental cost for CO2 capture with a technology
looping combustion (CLC) to two main processes for large-scale based on CLC and iii) the possibility in the second process to
hydrogen production with CO2 capture. The plants are assessed use a cheaper fuel. Both case CM and CB are considered for
from a techno-economic point of view and compared with a industrial implementation where case CM is more easily
conventional SMR plant with gas-fired furnace. The first of the implemented where less new units are needed. Case CB
suggested processes involves using NG as supplementary fuel to adds a few more complexities related to using biomass as
the furnace, whereas the second is using biomass. fuel input but it has the added benefit of enabling net
The first process results in net zero emissions and a hydrogen negative emissions.
production efficiency of above 76%, which is almost 8% higher
than for the conventional process. The levelized hydrogen pro-
duction cost of 2.6 V/kg is 1.6% lower than for the conventional Declaration of competing interest
process. The possibility to place the reformer tubes in the air
reactor was also investigated for this case. From a thermody- The authors declare that they have no known competing
namic point of view this could be interesting, but it would be financial interests or personal relationships that could have
limited by design constraints in the bed material circulation. appeared to influence the work reported in this paper.
The second process achieves net negative emissions of 34.1
tonCO2/MJH2 (142% lower than State-of-the-Art SMR plants). The
hydrogen production efficiency is above 72% (around 2% higher
than the conventional process). Despite that this process has a Acknowledgements
higher equipment cost compared to the other processes, the
levelized hydrogen production cost is estimated to around 2.67 The work has been supported by the Swedish Energy Agency
V/kg, only 1.4% higher than the conventional SMR plant resulting (project 40559e1 - Heat to endothermic industrial processes
in a CO2 avoidance cost around 2.7 V/tonCO2. This is to a large with new efficient combustion method in fluidized bed) and
extent a result of the possibility to use a cheaper biomass feed- Haldor Topsøe A/S.
stock in the SMR plant and the fact that low CO2 avoidance cost
can be expected with CLC, which makes its integration with SMR
Appendix
very interesting.
The promising results for both suggested processes can
to a large extent be explained by three factors: i) improved
Table A1 e Stream data for case CM (see Fig. 2).

Flow T P Composition (mol-%)
kg/s C bar C1 C2 C3 C4 CO2 CO O2 N2 Ar H2 H2O NiO Ni MgAl2O4
CM1 8.45 15 40.0 89.0 7.0 1.0 0.11 2.0 0.89
CM2 7.64 15 40.0 89.0 7.0 1.0 0.11 2.0 0.89
CM3 29.62 345 34.3 22.9 1.8 0.3 0.5 0.2 0.5 73.8
CM4 29.62 447 33.3 24.1 2.5 0.2 6.5 66.6
CM5 29.62 900 32.3 3.3 5.2 10.7 0.2 50.2 30.5
CM6 29.62 330 31.6 3.3 5.2 10.7 0.2 50.2 30.5
CM7 29.62 414 31.3 3.3 12.8 3.1 0.2 57.8 22.8
CM8 19.80 30 30.1 4.3 16.6 4.0 0.2 74.8 0.1
CM9 2.50 30 70.0 100.0
CM10 17.28 30 1.4 13.0 50.8 12.2 0.6 22.9 0.4
CM11 0.81 15 40.0 89.0 7.0 1.0 0.1 0.89
CM12 18.09 266 1.4 18.3 0.5 47.4 11.4 0.6 21.3 0.4
CM13 53.94 15 1.0 0.04 20.95 78.08 0.93
CM14 53.94 400 1.4 0.04 20.95 78.08 0.93
CM15 42.59 1008 1.0 0.05 2.7 96.4 1.6
CM16 42.59 150 1.0 0.05 2.7 96.4 1.6
CM17 29.44 929 1.0 56 0.4 0.5 0.2 43.0
CM18 29.44 188 1.0 56 0.4 0.5 0.2 43.0
CM19 22.45 30 110.0 98.1 0.7 0.8 0.4
CM20 29.53 15 1.0 100.0
CM21 29.53 485 92.0 100.0
CM22 7.56 160 6.0 100.0
CM23 896.5 929 1.0 10.0 10.0 80.0
CM24 907.8 1008 1.0 20.0 80.0
Table A2 e Stream data for case CB (see Fig. 4).

Flow T P Composition (mol-%)
kg/s C bar C1 C2 C3 C4 C(s) CO2 CO O2 N2 Ar H2 H2O NiO Ni MgAl2O4
CB1 6.43 15 40.0 89.0 7.0 1.0 0.11 2.0 0.89
CB2 37.27 341 34.3 15.3 1.2 0.2 0.3 0.2 0.3 82.4
CB3 37.27 443 33.3 5.7 2.1 0.1 6.2 75.7
CB4 37.27 900 32.3 1.0 5.7 7.2 0.1 42.7 43.3
CB5 37.27 330 32.3 1.0 5.7 7.2 0.1 42.7 43.3
CB6 37.27 394 31.6 1.0 11.7 1.2 0.1 48.7 37.3
CB7 37.27 217 31.3 1.0 12.8 0.1 0.1 49.8 36.2
CB8 19.16 30 29.8 1.5 20.1 0.2 0.2 77.9 0.1
CB9 2.50 30 70.0 100.0
CB10 16.65 30 1.1 5.0 67.3 0.6 0.6 26.1 0.5
CB11 5.90 63 1.1 9.2 36.4 0.5 14.6 7.7 17.9 13.8
CB12 22.56 62 1.1 6.8 15.2 39.3 6.5 3.2 0.3 22.6 6.0
CB13 56.33 15 1.0 0.04 20.95 78.08 0.93
CB14 56.33 400 1.4 0.04 20.95 78.08 0.93
CB15 34.40 926 1.0 0.6 59.2 0.6 2.9 0.3 2.0 34.4
CB16 35.31 1126 1.0 59.5 2.9 0.4 37.1
CB17 35.31 180 1.0 59.5 2.9 0.4 37.1
CB18 28.36 110 30.0 94.6 4.7 0.7
CB19 44.48 922 110.0 0.05 2.3 96.4
CB20 44.48 150 1.0 0.05 2.3 96.4
CB21 39.97 15 1 100.0
CB22 38.72 485 92 100.0
CB23 9.15 160 6 100.0
CB24 948.14 922 1.0 20.0 80.0
CB25 936.29 926 1.0 10.0 10.0 80.0
CB26 0.90 15 48.0 95.0 5.0
[6] Riis T, Hagen EF, Vie PJS, Ulleberg Ø. Hydrogen production

Table A3 e Ultimate analysis of the biomass included in and storage - R&D priorities and gaps. IEA - Hydrogen Co-
the Aspen Plus model. ordination Group; 2006.
Wood chips [74] [7] Metz B, Davidson O, de Coninch H, Loos M, Meyer L. IPCC
special report on carbon dioxide capture and storage.
Ultimate analysis (wt-% dry basis)
Prepared by working Group III of the Intergovernmental
C 50.7
Panel on Climate Change. In: Press CU; 2005. Cambridge,
H 6.2
United Kingdom and New York, NY, USA.
N 0.1
[8] European zero emission technology and innovation platform
Cl 0
(ETIP ZEP). In: Commercial scale feasibility of clean
S 0.01
hydrogen; 2017.
O 42.8
[9] Collodi G, Azzaro G, Ferrari N, Santos S. Techno-economic
Ash 0.2
evaluation of deploying CCS in SMR based merchant H2
Moisture content (wt-%) 6.9
production with NG as feedstock and fuel. Energy Procedia
HHV 20.58
2017;114:2690e712.
[10] Preston CK. The carbon capture project at air products. Port
Arthur Hydrogen Facility; 2018.
references [11] Preston CK. The shell quest carbon capture and storage
project. IEAGHG; 2018.
[12] Tanaka Y, Sawada Y, Tanase D, Tanaka J, Shiomi S,
n M. Techno-
[1] Stenberg V, Spallina V, Mattisson T, Ryde Kasukawa T. Tomakomai CCS demonstration project of
economic analysis of H2 production processes using fluidized Japan, CO2 injection in process. Energy Procedia
bed heat exchangers with steam reforming e Part 1: oxygen 2017;114:5836e46.
carrier aided combustion. Int J Hydrogen Energy [13] Pichot D, Granados L, Morel T, Schuller A, Dubettier R,
2020;45:6059e81. Lockwood F. Start-up of port-je ro
^ me CRYOCAP™ plant:
[2] IEA. Technology. Roadmap: hydrogen and fuel cells. 2015. optimized cryogenic CO2 capture from H2 plants. Energy
Paris, France. Procedia 2017;114:2682e9.
[3] Liu K, Song C, Subramani V. Hydrogen and syngas [14] Fahim MA, Alsahhaf TA, Elkilani A. Chapter 11 - hydrogen
production and purification technologies. Hoboken, New production. Fundamentals of petroleum refining.
Jersey: John Wiley & Sons, Inc.; 2010. Amsterdam: Elsevier; 2010. p. 285e302.
[4] (IEA) IEA. Energy technology perspectives 2015. 2015. Paris, [15] Rath LK. Assessment of hydrogen production with CO2
France. capture. Baseline state-of-the-art plants. DOE/NETL-2010/
[5] Voldsund M, Jordal K, Anantharaman R. Hydrogen 1434, Volume 1; 2010.
production with CO2 capture. Int J Hydrogen Energy [16] IEAGHG. Techno-economic evaluation of SMR based
2016;41:4969e92. standalone (merchant) plant with CCS. 2017.
[17] Shao R, Stangeland A. Amines used in CO2 capture - health from different biomass gasification processes. Int J Hydrogen
and environmental impacts. 2009. Energy 2018;43:9514e28.
[18] Gjernes E, Helgesen LI, Maree Y. Health and environmental [36] Pandey B, Prajapati YK, Sheth PN. Recent progress in
impact of amine based post combustion CO2 capture. Energy thermochemical techniques to produce hydrogen gas from
Procedia 2013;37:735e42. biomass: a state of the art review. Int J Hydrogen Energy
[19] Ada nez J, Abad A, Garcı́a-Labiano F, Gaya n P, de Diego LF. 2019;44:25384e415.
Progress in chemical-looping combustion and reforming [37] Shahabuddin M, Krishna BB, Bhaskar T, Perkins G. Advances
technologies. Prog Energy Combust Sci 2012;38:215e82. in the thermo-chemical production of hydrogen from
[20] Abanades JC, Arias B, Lyngfelt A, Mattisson T, Wiley DE, Li H, biomass and residual wastes: summary of recent techno-
et al. Emerging CO2 capture systems. Int J Greenhouse Gas economic analyses. Bioresour Technol 2020;299:122557.
Control 2015;40:126e66. [38] Moud PH, Andersson KJ, Lanza R, Pettersson JBC, Engvall K.
[21] Boot-Handford ME, Abanades JC, Anthony EJ, Blunt MJ, Effect of gas phase alkali species on tar reforming catalyst
Brandani S, Mac Dowell N, et al. Carbon capture and storage performance: initial characterization and method
update. Energy Environ Sci 2014;7:130e89. development. Fuel 2015;154:95e106.
[22] Mattisson T, Keller M, Linderholm C, Moldenhauer P, [39] Alstrup I, Rostrup-Nielsen JR, Røen S. High temperature
Ryde n M, Leion H, et al. Chemical-looping technologies using hydrogen sulfide chemisorption on nickel catalysts. Appl
circulating fluidized bed systems: status of development. Catal 1981;1:303e14.
Fuel Process Technol 2018;172:1e12. [40] Lyngfelt A, Brink A, Langørgen Ø, Mattisson T, Ryde n M,
[23] Lyngfelt A, Linderholm C. Chemical-looping combustion of Linderholm C. 11,000h of chemical-looping combustion
solid fuels e status and recent progress. Energy Procedia operationdwhere are we and where do we want to go? Int J
2017;114:371e86. Greenhouse Gas Control 2019;88:38e56.
[24] Li J, Zhang H, Gao Z, Fu J, Ao W, Dai J. CO2 capture with [41] Mendiara T, Garcı́a-Labiano F, Abad A, Gaya n P, de Diego LF,
chemical looping combustion of gaseous fuels: an overview. Izquierdo MT, et al. Negative CO2 emissions through the use
Energy Fuels 2017;31:3475e524. of biofuels in chemical looping technology: a review. Appl
[25] Fan LS, Zeng L, Wang W, Luo S. Chemical looping processes Energy 2018;232:657e84.
for CO2 capture and carbonaceous fuel conversion - prospect [42] Ada nez J, Abad A, Mendiara T, Gaya n P, de Diego LF, Garcı́a-
and opportunity. Energy Environ Sci 2012;5:7254e80. Labiano F. Chemical looping combustion of solid fuels. Prog
[26] Lyngfelt A. Oxygen carriers for chemical looping combustion Energy Combust Sci 2018;65:6e66.
- 4 000 h of operational experience. Oil & Gas Sci Technol e [43] Zhao X, Zhou H, Sikarwar VS, Zhao M, Park A-HA, Fennell PS,
Revue d’IFP Energies nouvelles 2011;66:161e72. et al. Biomass-based chemical looping technologies: the
[27] Hossain MM, de Lasa HI. Chemical-looping combustion (CLC) good, the bad and the future. Energy Environ Sci
for inherent CO2 separationsda review. Chem Eng Sci 2017;10:1885e910.
2008;63:4433e51. [44] Abad A. Chemical looping for hydrogen production. In:
[28] Ryde n M, Lyngfelt A. Using steam reforming to produce Fennell P, Anthony B, editors. Calcium and chemical looping
hydrogen with carbon dioxide capture by chemical-looping technology for power generation and carbon dioxide (CO2)
combustion. Int J Hydrogen Energy 2006;31:1271e83. capture. Cambridge, U.K: Woodhead Publishing; 2015.
[29] Spallina V, Shams A, Battistella A, Gallucci F, Annaland MvS. p. 327e67.
Chemical looping technologies for H2 production with CO2 [45] Luo M, Yi Y, Wang S, Wang Z, Du M, Pan J, et al. Review of
capture: thermodynamic assessment and economic hydrogen production using chemical-looping technology.
comparison. Energy Procedia 2017;114:419e28. Renew Sustain Energy Rev 2018;81:3186e214.
[30] Ortiz M, Gaya n P, de Diego LF, Garcı́a-Labiano F, Abad A, [46] Hallberg P. Mixed oxide oxygen carriers for chemical-looping
Pans MA, et al. Hydrogen production with CO2 capture by combustion. Gothenburg, Sweden: Chalmers University of
coupling steam reforming of methane and chemical-looping Technology; 2017.
combustion: use of an iron-based waste product as oxygen [47] Moldenhauer P, Hallberg P, Biermann M, Snijkers F,
carrier burning a PSA tail gas. J Power Sources Albertsen K, Mattisson T, et al. Oxygen-carrier development
2011;196:4370e81. of calcium manganiteebased materials with perovskite
[31] Pans MA, Abad A, de Diego LF, Garcı́a-Labiano F, Gaya n P, structure for chemical-looping combustion of methane.
Ada nez J. Optimization of H2 production with CO2 capture by Energy Technol 2020;8:2000069.
steam reforming of methane integrated with a chemical- [48] Gogolev I, Linderholm C, Gall D, Schmitz M, Mattisson T,
looping combustion system. Int J Hydrogen Energy Pettersson JBC, et al. Chemical-looping combustion in a
2013;38:11878e92. 100kW unit using a mixture of synthetic and natural oxygen
[32] Rahimpour MR, Hesami M, Saidi M, Jahanmiri A, Farniaei M, carriers e operational results and fate of biomass fuel alkali.
Abbasi M. Methane steam reforming thermally coupled with Int J Greenhouse Gas Control 2019;88:371e82.
fuel combustion: application of chemical looping concept as [49] Abad A, Garcı́a-Labiano F, Gaya n P, de Diego LF, Ada nez J.
a novel technology. Energy Fuels 2013;27:2351e62. Redox kinetics of CaMg0.1Ti0.125Mn0.775O2.9d for chemical
[33] IPCC. Global Warming of 1.5 C. An IPCC Special Report on the looping combustion (CLC) and chemical looping with oxygen
impacts of global warming of 1.5 C above pre-industrial uncoupling (CLOU). Chem Eng J 2015;269:67e81.
levels and related global greenhouse gas emission pathways. [50] Mattisson T, Ada nez J, Mayer K, Snijkers F, Williams G,
In: The context of strengthening the global response to the Wesker E, et al. Innovative oxygen carriers uplifting
threat of climate change, sustainable development, and chemical-looping combustion. Energy Procedia
efforts to eradicate poverty. Geneva, Switzerland: World 2014;63:113e30.
Meteorological Organization; 2018. p. 32. [51] Jing D, Jacobs M, Hallberg P, Lyngfelt A, Mattisson T.
[34] Arregi A, Amutio M, Lopez G, Bilbao J, Olazar M. Evaluation of Development of CaMn0.775Mg0.1Ti0.125O3-d oxygen carriers
thermochemical routes for hydrogen production from produced from different Mn and Ti sources. Mater Des
biomass: a review. Energy Convers Manag 2018;165:696e719. 2016;89:527e42.
[35] Salkuyeh YK, Saville BA, MacLean HL. Techno-economic [52] Uusitalo MA, Vuoristo PMJ, Ma € ntyla
€ TA. High temperature
analysis and life cycle assessment of hydrogen production corrosion of coatings and boiler steels in reducing chlorine-
containing atmosphere. Surf Coating Technol [71] Lyngfelt A, Leckner BA. 1000 MWth boiler for chemical-
2002;161:275e85. looping combustion of solid fuels e discussion of design and
[53] Leion H, Mattisson T, Lyngfelt A. Use of ores and industrial costs. Appl Energy 2015;157:475e87.
products as oxygen carriers in chemical-looping combustion. [72] Mendiara T, Gaya n P, Garcı́a-Labiano F, de Diego LF, Perez-
Energy Fuels 2009;23:2307e15. Astray A, Izquierdo MT, et al. Chemical looping combustion
[54] Matzen M, Pinkerton J, Wang X, Demirel Y. Use of natural of biomass: an approach to BECCS. Energy Procedia
ores as oxygen carriers in chemical looping combustion: a 2017;114:6021e9.
review. Int J Greenhouse Gas Control 2017;65:1e14. [73] Biede O. Encap - WP 1.1 Deliverable D1.1.1 & D1.1.2 -
[55] Fossdal A, Bakken E, Øye BA, Schøning C, Kaus I, reference cases and guidelines for technology concepts
Mokkelbost T, et al. Study of inexpensive oxygen carriers for Vattenfall A/S Report No, vol. 55431; 2008.
chemical looping combustion. Int J Greenhouse Gas Control [74] Hannula I, Kurkela E. A parametric modelling study for
2011;5:483e8. pressurised steam/O2-blown fluidised-bed gasification of wood
[56] Ryden M, Leion H, Mattisson T, Lyngfelt A. Combined oxides with catalytic reforming. Biomass Bioenergy 2012;38:58e67.
as oxygen-carrier material for chemical-looping with oxygen [75] Basu P. Heat transfer. Combustion and gasification in
uncoupling. Appl Energy 2014;113:1924e32. fluidized beds. CRC Press; 2006. p. 173e209.
[57] Ryden M, Hanning M, Corcoran A, Lind F. Oxygen carrier [76] Andeen BR, Glicksman LR. Heat transfer to horizontal tubes
aided combustion (OCAC) of wood chips in a semi- in shallow fluidized beds. St Louis, MO: ASME-AIChe Heat
commercial circulating fluidized bed boiler using manganese Transfer Conference; 1976.
ore as bed. Material. Appl Sci 2016;6:347. [77] Johnsson F, Leckner B. Vertical distribution of solids in a CFB-
[58] Sundqvist S, Arjmand M, Mattisson T, Ryde n M, Lyngfelt A. furnace. Proc Int Conf Fluid Bed Combustion 1995:671e9.
Screening of different manganese ores for chemical-looping [78] Stenberg V, Lind F, Ryde n M. Measurement of bed-to-tube
combustion (CLC) and chemical-looping with oxygen surface heat transfer coefficient to a vertically immersed u-
uncoupling (CLOU). Int J Greenhouse Gas Control tube in bubbling loop seal of a CFB boiler. Powder Technol
2015;43:179e88. 2021;381:652e64.
[59] Arjmand M, Leion H, Mattisson T, Lyngfelt A. Investigation of [79] Nexant Inc. Equipment design and cost estimation for small
different manganese ores as oxygen carriers in chemical- modular biomass systems, synthesis gas cleanup, and
looping combustion (CLC) for solid fuels. Appl Energy oxygen separation equipment: task 1: cost Estimates of Small
2014;113:1883e94. Modular Systems. 2006. San Francisco.
[60] Hupa M, Karlstro € m O, Vainio E. Biomass combustion [80] Manzolini G, Macchi E, Gazzani M. CO2 capture in integrated
technology development e it is all about chemical details. gasification combined cycle with SEWGS e Part B: economic
Proc Combust Inst 2017;36:113e34. assessment. Fuel 2013;105:220e7.
[61] Khan AA, de Jong W, Jansens PJ, Spliethoff H. Biomass [81] Ratnasamy C, Wagner JP. Water gas shift catalysis. Catal Rev
combustion in fluidized bed boilers: potential problems and 2009;51:325e440.
remedies. Fuel Process Technol 2009;90:21e50. [82] Energimyndigheten. Tra € dbra
€ nsle- och torvpriser Nr 3/2017.
[62] Leckner B. 14 - atmospheric (non-circulating) fluidized bed Sveriges officiella statistik - statistiska Meddelanden EN 0307
(FB) combustion. In: Scala F, editor. Fluidized bed SM 1703. 2017.
technologies for near-zero emission combustion and [83] Gustavsson Tony, Valmet AB. Personal communication. 2019.
gasification. Woodhead Publishing; 2013. p. 641e68. [84] Campanari S, Chiesa P, Manzolini G, Bedogni S. Economic
[63] Lyngfelt A. Chemical-looping combustion of solid fuels e analysis of CO2 capture from natural gas combined cycles using
status of development. Appl Energy 2014;113:1869e73. Molten Carbonate Fuel Cells. Appl Energy 2014;130:562e73.
[64] Mattisson T, Lyngfelt A, Leion H. Chemical-looping with [85] Eurostat. Energy price statistics. http://ec.europa.eu/
oxygen uncoupling for combustion of solid fuels. Int J eurostat/statistics-explained/index.php?title¼Archive:
Greenhouse Gas Control 2009;3:11e9. Energy_price_statistics. [Accessed 4 September 2017].
[65] Gaya n P, Abad A, de Diego LF, Garcı́a-Labiano F, Ada
nez J. [86] Hallberg P, Hanning M, Ryde n M, Mattisson T, Lyngfelt A.
Assessment of technological solutions for improving Investigation of a calcium manganite as oxygen carrier
chemical looping combustion of solid fuels with CO2 during 99h of operation of chemical-looping combustion in a
capture. Chem Eng J 2013;233:56e69. 10kWth reactor unit. Int J Greenhouse Gas Control
[66] Stenberg V, Ryde n M, Mattisson T, Lyngfelt A. Exploring 2016;53:222e9.
novel hydrogen production processes by integration of [87] Moldenhauer P, Linderholm C, Ryde n M, Lyngfelt A. Avoiding
steam methane reforming with chemical-looping CO2 capture effort and cost for negative CO2 emissions using
combustion (CLC-SMR) and oxygen carrier aided industrial waste in chemical-looping combustion/
combustion (OCAC-SMR). Int J Greenhouse Gas Control gasification of biomass. Mitigation and Adaptation Strategies
2018;74:28e39. for Global Change. 2019.
[67] Franco F, Anantharaman R, Bolland O, Booth N, van Dorst E, [88] Hogg D. Costs for municipal waste management in the EU -
Ekstrom C, et al. D4.9 European Best practice guidelines for final report to directorate general environment. European
assessment of CO2 capture technologies. 2011. Commission; 2002.
[68] Smith AR, Klosek J. A review of air separation technologies [89] Hallberg P, Ka€ llen M, Jing D, Snijkers F, van Noyen J, Ryden M,
and their integration with energy conversion processes. Fuel et al. Experimental investigation of CaMnO3-d based oxygen
Process Technol 2001;70:115e34. carriers used in continuous chemical-looping combustion.
[69] Darde A, Prabhakar R, Tranier J-P, Perrin N. Air separation Int J Chem Eng 2014;2014:9.
and flue gas compression and purification units for oxy-coal [90] Ada nez J, Gaya n P, Celaya J, de Diego LF, Garcı́a-Labiano F,
combustion systems. Energy Procedia 2009;1:527e34. Abad A. Chemical looping combustion in a 10 kWth
[70] Beysel G. Enhanced cryogenic air separation A proven prototype using a CuO/Al2O3 oxygen Carrier: effect of
process applied to oxyfuel: future prospects. 1st Oxyfuel operating conditions on methane combustion. Ind Eng Chem
Combustion Conference; 2009. Res 2006;45:6075e80.
[91] Linderholm C, Schmitz M. Chemical-looping combustion of [95] IPCC.. In: Metz B, Davidson O, de Coninck HC, Loos M,
solid fuels in a 100kW dual circulating fluidized bed system Meyer LA, editors. IPCC special report on carbon dioxide
using iron ore as oxygen carrier. J Environ Chem Eng capture and storage. Prepared by working group III of the
2016;4:1029e39. intergovernmental panel on climate change [; 2005. p. 442.
[92] IEAGHG. Project costs - industrial applications - QUEST. In: Cambridge, United Kingdom and New York, NY, USA.
CCS cost network - 2016 workshop; 2016. Cambridge, [96] Energimyndigheten. Tra € dbra
€ nsle och torvpriser fo
€ r tredje
Massachusetts, USA. kvartalet. 2018 [In Swedish]. 2018-11-29 ed2018.
[93] International Energy Agency. Technology roadmapdcarbon [97] Fridahl M, Lehtveer M. Bioenergy with carbon capture and
capture and storage. 2013. storage (BECCS): global potential, investment preferences,
[94] Carbon Counts Company. CCS roadmap for industry: high- and deployment barriers. Energy Res & Social Sci
purity CO2 sources: final draft sectoral assessment. 2010. 2018;42:155e65.

Techno-Economic Analysis of Processes With Integration of Fluidized Bed Heat Exchangers For H2 Production e Part 2 - Chemical-Looping Combustion

Uploaded by

Copyright:

Available Formats

Techno-Economic Analysis of Processes With Integration of Fluidized Bed Heat Exchangers For H2 Production e Part 2 - Chemical-Looping Combustion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Techno-Economic Analysis of Processes With Integration of Fluidized Bed Heat Exchangers For H2 Production e Part 2 - Chemical-Looping Combustion

Uploaded by

Copyright:

Available Formats

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 6 ( 2 0 2 1 ) 2 5 3 5 5 e2 5 3 7 5

Available online at www.sciencedirect.com

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

Techno-economic analysis of processes with

Chemical-looping combustion can be integrated in a SMR plant for H2 production.

article info abstract

Abbreviations NETL National energy technology laboratory

The most common approach for CO2 capture in the in-

Table 1 e SMR hydrogen production plants with CO2 capture.

The second configuration is also based on chemical-looping

additions related to case CB in particular which mainly results

The equivalent hydrogen production efficiency is esti-

means that total cross-section for the system can be reduced

oxygen carrier bed material was instead estimated based on

Table 6 e Heat transfer calculations in the reformer furnaces.

Logarithmic mean 450 144 134 142 122 142

Table 7 e Design results for the reformers (GFF and FBHE).

O&M fixed costs (MV (% of total O&M fixed costs))

O&M variable costs (MV (% of total O&M variable costs))

LCOH (V/Nm3H2 ) 0.236 0.219 0.233 0.235 0.240 0.234

Thermodynamic sensitivity analysis OM, so no variation of basic thermodynamic parameters was

stream from selling LP steam which is slightly higher in case

heat transfer to tube compared to the conventional process

Table A1 e Stream data for case CM (see Fig. 2).

Table A2 e Stream data for case CB (see Fig. 4).

[6] Riis T, Hagen EF, Vie PJS, Ulleberg Ø. Hydrogen production

You might also like