Integrating solid direct air capture systems with green hydrogen

production: Economic synergy of sector coupling

Sunwoo Kim1,2, Joungho Park3, and Jay H. Lee1*
1
Mork Family Department of Chemical Engineering and Material Sciences, University of
Southern California, 925 Bloom Walk, Los Angeles, CA 90089, USA
2
Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of
Science and Technology (KAIST), 291, Daehak-ro, Yuseong-gu, Daejeon, 34141,
Republic of Korea
3
Energy AI & Computational Science Laboratory, Korea Institute of Energy Research, 71-
2 Jang-dong, Yuseong-gu, Daejeon, 305-343, Republic of Korea
*
Corresponding author’s E-mail: [email protected]
Abstract
In the global pursuit of sustainable energy solutions, mitigating carbon dioxide (CO2) emissions
stands as a pivotal challenge. With escalating atmospheric CO2 levels, the imperative of direct
air capture (DAC) systems becomes evident. Simultaneously, green hydrogen (GH) emerges
as a pivotal medium for renewable energy. Nevertheless, the substantial expenses associated
with these technologies impede widespread adoption, primarily due to significant installation
costs and underutilized operational advantages when deployed independently. Integration
through sector coupling enhances system efficiency and sustainability, while shared power
sources and energy storage devices offer additional economic benefits. In this study, we assess
the economic viability of polymer electrolyte membrane electrolyzers versus alkaline
electrolyzers within the context of sector coupling. Our findings indicate that combining GH
production with solid DAC systems yields significant economic advantages, with
approximately a 10% improvement for PEM electrolyzers and a 20% enhancement for alkaline
electrolyzers. These results highlight a substantial opportunity to improve the efficiency and
economic viability of renewable energy and green hydrogen initiatives, thereby facilitating the
broader adoption of cleaner technologies.
Keywords: Solid direct air capture, Green hydrogen, Alkaline and PEM electrolyzer, Sector
coupling

1
Graphical abstract

2
 Main
The imperative for a sustainable global society increasingly focuses on mitigating
anthropogenic carbon dioxide (CO2) emissions.[1-3] Direct air capture (DAC) systems are
gaining recognition as a vital technology for recovering atmospheric CO2.[4-6] This
perspective is supported by a growing consensus among research entities and governments that
a transition to clean, renewable energy must be complemented by technologies capable of
actively extracting CO2 from the atmosphere.[7, 8]
Among the diverse technologies in DAC, temperature-vacuum swing adsorption
(TVSA) with solid adsorbents is emerging as a particularly promising method.[9, 10] Solid
DAC systems offer better energy efficiency compared to their liquid counterparts and
demonstrate significant operational flexibility, which is crucial for adapting to the variability
of energy sources, particularly those based on renewables.[11] Solid DAC systems operate in
two phases: adsorption and desorption.[12] During adsorption, CO2 is captured from the
atmosphere using solid adsorbents. In the desorption phase, the captured CO2 is released for
storage. This process is more energy-efficient than liquid DAC, leading to lower operational
costs.[13] Additionally, the ability of solid DAC systems to enter a shutdown state provides
unmatched adaptability to fluctuating renewable energy availability.[14] However, the high
installation costs of solid DAC facilities necessitate operating at high loads to achieve
economic viability in the standalone solid DAC system.
In tandem with DAC, the role of green hydrogen (GH) as a sustainable energy carrier
is becoming increasingly significant.[15-19] GH refers to hydrogen produced by utilizing
renewable energy sources, typically through methods like water electrolysis, including those
employing Proton Exchange Membrane (PEM) and alkaline electrolyzers.[20-22] However,
the intrinsic variability of renewable energy sources, coupled with the requirement for timely
satisfaction of hydrogen demand, can lead to oversized GH facilities and energy wastage
through curtailment.[23, 24] Consequently, the operational flexibility advantages inherent in
both technologies are not fully realized when deployed as standalone systems. Therefore,
adopting a sector coupling strategy presents a promising avenue to improve economic
feasibility and efficiency in this domain.
Our study scrutinizes the integration of solid DAC systems with GH system through
sector coupling to address existing challenges. To efficiently integrate this variable energy, a
coupling of the power sector to the residential, transport, industry, and commercial/trade sector
is often promoted, which is called sector coupling.[25] This approach ensures an optimal
alignment of energy supply and demand when paired with GH production, leveraging the
strengths of both technologies to enhance overall system efficiency and sustainability.
Additionally, sharing power sources and energy storage devices can offer further economic
benefits to the sector-coupled system.
The concept of sector coupling has been acknowledged for its potential to
synergistically integrate renewable energy systems, offering economic and energy efficiency
benefits [26, 27]. Rafael et al. explored the potential of sector coupling in carbon capture and
storage (CCS) within industrial settings such as cement production, steel manufacture, and
waste incineration.[28] By sector coupling, significant emission reductions of up to 860 Mt of

3
CO2 annually in Europe alone, equivalent to nearly 27% of European GHG emissions, are
estimated. Furthermore, the integrated approach yields notable economic benefits, with cost
savings of approximately 50% compared to standalone CCS technologies, emphasizing the
importance of system integration in achieving net-zero and net-negative industrial sectors.
Another study delved into the integration of renewable energy across Europe using the PyPSA-
Eur-Sec-30 model, concentrating on the implications of sector coupling and the expansion of
cross-border transmission networks.[29] Their research emphasizes the pivotal role of
technologies such as battery electric vehicles, power-to-gas units, and long-term thermal
energy storage in realizing a 95% decrease in CO2 emissions by 2050. Further research supports
using hydrogen for long-duration energy storage in grids dependent on variable renewable
energy, advocating an integrated approach for electricity and hydrogen supply chains.[30]
Favoring power-to-hydrogen pathways over power-to-gas-to-power strategies could enhance
renewable energy utilization and decrease overall system costs, particularly in scenarios
targeting extensive decarbonization.
Despite the growing interest in sector coupling as a means to boost renewable energy
efficiencies, there's a notable lack of research exploring the advantages of linking GH with
solid DAC systems. Furthermore, there is lack of research that comparatively evaluate the
economic feasibility of polymer electrolyte membrane (PEM) electrolyzer and Alkaline
electrolyzer under sector coupling. The minimum operating threshold for GH production varies
depending on the type of water electrolyzer. PEM electrolyzers have a lower minimum
operating threshold due to their use of a solid polymer electrolyte, which allows efficient proton
conduction and rapid response to varying electrical inputs. In contrast, alkaline electrolyzers
require a higher minimum current to maintain effective electrochemical reactions because they
use a liquid electrolyte, typically potassium hydroxide, necessitating more stable operating
conditions. These differing characteristics of electrolyzers result in varying degrees of synergy
with solid DAC in sector-coupling applications. We investigate the synergistic behavior of
each electrolyzer technology by varying the ratio between hydrogen demand and CO2 removal
targets, as well as how the installation costs of these facilities influence the economic viability
of each system and their synergy variations. This gap highlights a compelling opportunity to
not only enhance the efficiency and economic feasibility of renewable energy and green
hydrogen initiatives but also to advance the development and adoption of DAC technologies.

System description
In this research, we examine five distinct scenarios: one featuring a GH system and a
solid DAC system, both powered by renewable energy, alongside three configurations of
systems where these technologies are integrated through sector coupling. In each scenario,
wind turbines[31] and photovoltaic (PV) panels[32, 33] generate electricity from wind and
solar energy, respectively. Energy storage is facilitated through batteries[34], hydrogen tanks
(HT)[35], and thermal energy storage (TES)[14], storing electricity, hydrogen, and thermal
energy, correspondingly. Water electrolysis[36, 37], performed by electrolyzers, splits water
into oxygen and hydrogen gases, while fuel cells[38] convert hydrogen back into electricity.

4
Heat pumps are utilized to transform electricity into thermal energy.[14, 39] Curtailment refers
to the process of dissipating energy or idling turbines to prevent grid overload.
Given the differing current types required by each facility, efficient energy integration
and operation necessitate suitable converters and bus configurations. electrolyzers and batteries
operate on direct current (DC), whereas heat pumps and DAC plants require alternating current
(AC). We assume converter efficiencies of 98% for both DC/DC and AC/AC, and 95% for
AC/DC conversions in our analysis.[40]
In Fig. 1 (a), the focus of the GH system is on meeting hydrogen demand promptly,
with the primary energy consumption directed towards powering the water electrolyzer.
Consequently, it is most efficient to configure the water electrolyzer to operate on a DC bus.
Excess energy is stored as electricity in batteries and as hydrogen in HT. Additionally, any
surplus hydrogen may be converted back into electricity using a fuel cell, ensuring that the
system meets the minimum operational requirements of the water electrolysis process.[41] This
will be explained further in the Supplementary Materials.
In Fig. 1 (b), the solid DAC system is designed with a goal of achieving an annual
average CO2 capture target, making energy supply to the DAC plant its principal energy
demand. As such, configuring the DAC plant to utilize an AC bus is the most efficient approach.
Excess energy is stored in the form of electricity in batteries and as thermal energy in TES
systems. Additionally, the system includes a heat pump, which is subject to a minimum
operational requirement.[42] This will be explained further in the Supplementary Materials.
Fig. 1 (c) depict the integration of GH and solid DAC systems within a sector-coupled
framework. Fig. 1 (c) presents a hybrid DC-AC system, a purely DC setup and an AC setup
can be found in the Supplementary Material, Fig. B. 1 and Fig. B. 2, respectively. Across all
configurations, the systems are designed to ensure prompt satisfaction of hydrogen demand
and to meet an annual average target for CO2 removal. Each configuration employs a unique
arrangement of facilities and converters to match its electrical current requirements.
Specifically, the DC setup is optimized for GH efficiency but performs less efficiently for the
DAC plant. Conversely, the AC setup offers high efficiency for the DAC plant at the expense
of GH efficiency. The hybrid DC-AC configuration achieves high efficiency within each
respective bus but faces challenges with power exchange efficiency between DC and AC buses.
This necessitates a thorough analysis to determine the most effective configuration for
achieving the objectives of the sector-coupled system. This will be explained further in the
Supplementary Materials.

5
Fig. 1. (a) The Green Hydrogen (GH) system operating in a Direct Current (DC) configuration, designed to ensure the timely fulfillment of
hydrogen demand, (b) Renewable-based Solid DAC System: Operating under an Alternating Current (AC) configuration, aimed at achieving
an CO2 reduction target on an annual average basis throughout the project's duration, (c) Integrated sector coupling system of GH and solid
DAC, configured to operate between DC and AC, to balance the objectives of timely satisfaction of hydrogen demand and achieving annual
CO2 reduction targets over the project's duration.
6
 Evaluation metrics
The economic performance of the system is assessed using two primary metrics: the
Levelized Cost of producing Hydrogen (LCOH) and the Levelized Cost of DAC (LCOD),
defined by equation (2) and equation (3), respectively.

TAC GH TAC GH
LCOH = =
Annual GH production 1  1  (2)
   p 2hsc ,t  W 
N sc
sc t   

TAC DAC TAC DAC

LCOD = =
1
   dacsc3 ,t 
Annual CO2 reduction (3)
N sc sc t 

In the above, 𝑇𝐴𝐶 𝐺𝐻 denotes the total annualized cost (TAC) of the standalone GH system
and 𝑇𝐴𝐶 𝐷𝐴𝐶 corresponds to the TAC of standalone solid DAC system, respectively. 𝑁 𝑠𝑐
denotes the number of scenarios, 𝑝2ℎ𝑠𝑐,𝑡 presents the power rate to the electrolyzer for
scenario 𝑠𝑐 at time 𝑡 , 𝜂𝑊 indicates the conversion efficiency of the electrolyzer, and
3
𝑑𝑎𝑐𝑠𝑐,𝑡 denotes the desorption CO2 flow rate of solid DAC plant for scenario 𝑠𝑐 at time 𝑡.
For evaluating the economic synergy of sector coupling, we define two metrics. The
metric termed Improvement represents the difference in Total Annualized Cost (TAC) between
the combined system and the sum of the standalone systems. Additionally, there exists another
metric referred to as Synergy which denotes the ratio of the TAC difference to the sum of the
standalone TAC, while ensuring timely hydrogen supply and achieving the annual CO2
reduction target. The economic metrics are expressed as follows:

Improvement (MM$/yr) = TAC GH + TAC DAC − TAC Combined (4)

TAC Combined
Synergy (%) = 1 − (5)
TAC GH + TAC DAC

In the above, 𝑇𝐴𝐶 𝐶𝑜𝑚𝑏𝑖𝑛𝑒𝑑 indicates the TAC of the sector-coupled system. A Synergy value
of 0 signifies no economic advantage from the coupling, whereas a higher Synergy value
indicates increasing economic benefits.

Techno-economic analysis of standalone GH and DAC system

We conducted an economic evaluation of standalone GH and DAC systems across
different regions. We chose three locations—Antofagasta, Bundaberg, and Neom—each
offering promising renewable energy potential and distinct meteorological profiles. To
accommodate the unpredictability of renewable energy sources, we identified and
economically appraised the optimal design under three annual scenarios for solar and wind.

7
The specifics of this optimization process and detailed problem description are outlined in the
Supplementary Materials.
In Fig. 2 (a), the LCOH for PEM electrolyzers ranges from $3.2/kg H2 to $3.6/kg H2.
The primary cost component is attributed to power sources, encompassing turbines and PV
panels, followed by electrolyzers and energy storage devices. Conversely, in Fig. 2 (b), LCOH
values for alkaline electrolyzers range from $3.7/kg H2 to $4.3/kg H2, approximately 15%
higher than PEM electrolyzers. The increased minimum operating threshold of alkaline
electrolyzers escalates battery costs, alongside similar impacts on power sources and fuel cells.
However, the installation cost advantage of alkaline electrolyzers reduces the electrolyzer cost
element compared to PEM counterparts. Fig. 2 (c) depicts the LCOD in various regions,
ranging from $80/t CO2 to $86/t CO2, with DAC facilities as the largest cost driver, trailed by
batteries and power sources. The high capital expenditure associated with solid DAC facilities
necessitates operating at ~90% load to achieve economic viability in a standalone solid DAC
system.
Moreover, we conducted sensitivity analyses, varying facility costs by ±20%. In Fig.
3 (a), changes in power sources, electrolyzers, and battery costs exhibit the greatest impact on
LCOH for PEM electrolyzer GH systems, whereas Fig. 3 (b) demonstrates a different order of
influence which is power sources, battery, and electrolyzer. Fig. 3 (c) reveals that DAC, power
sources, and battery costs primarily affect LCOD. Significantly, enhanced operating flexibility
of heat pumps markedly reduces LCOD, as depicted in Fig. 4 (d). A 10% reduction in the
minimum load threshold corresponds to a ~4% LCOD reduction, equating to approximately
$3/t CO2, despite facility costs remaining constant. This underscores the surprising capacity of
heat pump flexibility to substantially lower overall costs.
Conclusively, our analysis examines how variations in power sources, battery,
electrolyzer, and DAC facilities’ cost impact on the economic feasibility and synergies of
sector-coupled systems as well as the operating flexibility of heat pump.

8
Fig. 2. Cost breakdown of standalone green hydrogen system and DAC system by varying
regions: (a) PEM electrolyzer-based standalone green hydrogen system, (b) Alkaline
electrolyzer-based standalone green hydrogen system, (c) standalone solid DAC system. The
light-blue bar indicates turbine, apricot bar represents PV panel, orange bar denotes battery,
green bar indicates electrolyzer, blue bar indicates fuel cell, grey bar denotes hydrogen tank,
dark blue bar represents solid DAC facility, red bar indicates heat pump, and purple patterned
bar denotes the thermal energy storage devices.

9
Fig. 3. (a) Sensitivity analysis of cost variations (±20%) for different components in standalone
green hydrogen and DAC systems. (a) PEM electrolyzer-based green hydrogen system, (b)
Alkaline electrolyzer-based green hydrogen system, (c) Solid DAC system, and (d) Minimum
load of heat pumps in a standalone solid DAC system. The blue bar indicates +20% cost
increased cases and red bars represents -20% cost reduction cases for each facility.

Economic evaluation and synergy assessment of sector-coupled system

Our study evaluates the economic advantages of a sector-coupled system integrating
solid DAC with GH production, compared against the standalone solid DAC and GH systems.
In the defined baseline scenario, the GH system is designed to fulfill a constant timely hydrogen
demand of 10 kt every year, which is the general capacities of existing steam methane
reforming plant. For the solid DAC system, we assess the combined economic impact of sector
coupling across a range of average annual CO2 removal rate, from 0.1 Mt to 1.3 Mt with 0.2
Mt increments. In terms of molar quantities, approximately 0.2 Mt of CO2 corresponds to the
10 kt of hydrogen.
During our previous session, we observed that the PEM electrolyzer demonstrated
approximately 15% greater cost-effectiveness compared to its alkaline counterpart. However,
within a sector-coupled framework, the alkaline electrolyzer exhibited a slight advantage, being
about 2% more economically efficient than the PEM variant. In the context of a standalone
green hydrogen system, a substantial disparity in economic viability existed between the two
10
electrolyzer types. Nevertheless, when integrated within a sector-coupled system, this gap
diminished notably, with the alkaline electrolyzer even surpassing the PEM variant in
economic feasibility.
In Fig. 4, alkaline electrolyzer-based system shows an ~1.7 MM$ decrease in TAC,
attributed to their lower cost, albeit requiring ~0.5 MM$ more in PV panel installations
compared to PEM electrolyzer-based system. Notably, in the standalone green hydrogen
system, the alkaline electrolyzer demanded three times larger batteries and power sources than
the PEM electrolyzer due to a high minimum load of 20%. However, within the sector-coupled
paradigm, the economic advantages of shared infrastructure, particularly batteries and power
sources with the DAC system, became significant.
Fig. 5 (a) underscores the synergy between 2-11 MM$ when PEM electrolyzers and
DAC are sector-coupled, with this synergy increasing steadily as the CO2 removal rate rises.
At an annual hydrogen production of 10 kt, the highest percentage synergy occurs when the
CO2 removal rate ranges between 0.3-0.5 Mt/yr.
Conversely, Fig. 5 (b) illustrates that the synergy within the sector-coupled scenario
of alkaline electrolyzers and DAC ranges between 7-18 MM$, with a similar trend of increasing
synergy as CO2 removal rate rises. However, the highest percentage synergy is observed when
the CO2 removal rate is at 0.1 Mt/yr, gradually decreasing as the removal rate increases.
The rationale behind the greater synergy observed with alkaline electrolyzers lies in
the costly nature of DAC installations, necessitating the optimal design of deploying numerous
batteries operating at a high average rate. Through sector coupling, the shared utilization of
facilities minimizes the need for excessive battery investment, mitigating the operational
inflexibility associated with alkaline electrolyzers. Consequently, despite their intrinsic
operational limitations, the cost advantage of alkaline electrolyzers over PEM variants becomes
more pronounced within this integrated framework.

11
Fig. 4. TAC breakdown of sector-coupled systems across different regions and CO2 removal
rates: (a) PEM electrolyzer-based sector-coupled system, (b) Alkaline electrolyzer-based
sector-coupled system. The light-blue bar indicates turbine, red bar represents PV panel, orange
bar denotes battery, green bar indicates electrolyzer, blue bar indicates fuel cell, grey bar
denotes hydrogen tank, dark blue bar represents solid DAC facility, dark red bar indicates heat
pump, and purple pattern denotes the thermal energy storage devices.

12
Fig. 5. Improvement (MM$/yr) and synergy (%) of sector-coupled system across different
regions by varying CO2 removal rate: (a) PEM electrolyzer-based sector-coupled system, (b)
Alkaline electrolyzer-based sector-coupled system. The yellow bar indicates improvement at
Neom, the grey bar denotes improvement at Bundaberg, and the green bar represents
improvement at Antofagasta. The red dashed line denotes synergy at Neom, the black dashed
line represents synergy at Bundaberg, and the green dashed line indicates synergy at
Antofagasta.

Techno-economic analysis of sector-coupled system and their synergy variation

We examine how variations in facility costs impact the economic feasibility and
synergies of sector-coupled systems, critical given uncertainties in future renewable
technology cost projections. This endeavor aims to illuminate the potential synergies within
sector-coupled systems across diverse future scenarios. Consequently, it becomes imperative
to scrutinize the economic implications and synergistic effects within our proposed sector
coupling model when there are substantial fluctuations in the costs of key facilities. This
session focuses on assessing the variations in synergy and TAC when the costs of individual
facilities escalate by 20% from the baseline scenario. The evaluation criteria comprise TAC
difference (MM$/yr) and synergy difference (%). Specifically, we analyze the impacts of a 20%
cost increase for each of the following facilities: battery, DAC, electrolyzer, and power source.
In Fig. 6 (a), a noticeable trend emerges where the TAC difference escalates with
increasing CO2 removal rates in the sector-coupled system employing PEM electrolyzer.
Specifically, a 20% surge in the power source price results in a TAC increment of
approximately 4 MM$/yr at a CO2 removal rate of 0.1 Mt/yr, and this figure rises to about 9
MM$/yr at a CO2 removal rate of 1.3 Mt/yr. Similarly, a 20% hike in DAC facility costs leads
to a TAC rise of around 1 MM$/yr at a CO2 removal rate of 0.1 Mt/yr, while this figure elevates
to about 10 MM$/yr at a CO2 removal rate of 1.3 Mt/yr. Likewise, a 20% increase in battery

13
costs results in a TAC increase of approximately 0.8 MM$/yr at a CO2 removal rate of 0.1
Mt/yr, which jumps to about 4.3 MM$/yr at a CO2 removal rate of 1.3 Mt/yr. Notably, a 20%
rise in electrolyzer costs leads to a consistent TAC increase of about 2 MM$/yr across all CO2
removal rates.
However, when considering synergy difference, a clear contrast emerges between the
two systems. Fig. 6 (c) illustrates that when battery costs are heightened compared to the base
case, the synergy difference escalates from +0.35% to +0.8% with a 20% increase in battery
costs, primarily due to the augmented CO2 removal rate. Conversely, Fig. 6(d) showcases a
decrease in synergy difference from 2.1% to 1.35% as CO2 removal rates increase. This
discrepancy arises from the fact that the TAC of the standalone Alkaline GH system is
approximately three times more sensitive to battery costs than the PEM electrolyzer, as
evidenced in Fig. 3. Moreover, Figs. 6 (a) and (b) depict that as battery costs in the sector-
coupled system rise, the TAC difference increases nearly proportionally with the CO2 removal
rate. Essentially, the sector-coupled system employing an alkaline electrolyzer demonstrates a
more substantial synergy difference compared to the system with a PEM electrolyzer.
Additionally, as CO2 removal rates escalate, the proportion of TAC of the standalone GH
system in the denominator of Synergy diminishes, indicating an inverse trend.
Regarding a 20% increase in DAC facility costs, both systems exhibit a similar
trajectory, with a synergy difference of approximately -0.3% at a CO2 removal rate of 0.1 Mt/yr
and about -0.6% at 0.5-1.3 Mt/yr. Typically, high installation costs for DAC units necessitate
opting for smaller capacities operating at higher average rates to maintain cost-effectiveness.
Consequently, as DAC unit installation costs decrease, the benefits of sector coupling become
more apparent. Reduced installation costs encourage the expansion of DAC's optimal capacity,
facilitating more dynamic DAC system operation.
When examining electrolyzers, akin to DACs, the advantages of sector coupling
diminish as electrolyzer costs rise, leading to smaller installations and higher average operating
rates. Moreover, as the CO2 removal rate escalates, the impact of increased electrolyzer costs
on synergy difference decreases due to the reduced contribution of the standalone GH system
to the sector-coupled system's TAC. However, there exists a disparity in magnitude between
the two systems: in the sector-coupled system employing PEM electrolyzer, the synergy
difference stands at -0.2% at a CO2 removal rate of 0.1 Mt/yr and -0.1% at 1.3 Mt/yr.
Conversely, in the system utilizing an alkaline electrolyzer, the synergy difference is -1.25%
at 0.1 Mt/yr and -0.4% at 1.3 Mt/yr. Notably, the PEM electrolyzer in the standalone GH system
exhibits approximately twice the sensitivity to electrolyzer costs compared to the alkaline
electrolyzer.
In a sector-coupled setup, a 20% increase in electrolyzer costs results in a TAC hike
of 2 MM$/yr for PEM electrolyzers and 1.5 MM$/yr for alkaline electrolyzers, irrespective of
the CO2 removal rate. This implies that in the sector-coupled system employing PEM
electrolyzers, the denominator TAC in the synergy difference experiences a more pronounced
increase due to the rise in electrolyzer costs compared to the numerator, in contrast to the
system with alkaline electrolyzers. Consequently, the synergy difference is smaller in the PEM
electrolyzer system compared to the alkaline electrolyzer system.

14
 An intriguing finding emerges when the cost of the power source facility (turbine, PV
panel) is increased by 20%. In the system utilizing PEM electrolyzer, a negative synergy
difference is observed when the CO2 removal rate is below 0.5 Mt/yr, but a positive synergy
difference emerges beyond that threshold. To elucidate this, it's crucial to recognize that the
power source facility serves dual roles in terms of sector coupling synergy. Firstly, it supplies
variable electricity necessary for the dynamic operation of DAC and electrolyzer systems.
Essentially, more renewable energy supply is required to enable the dynamic operation of these
systems, thereby enhancing sector coupling synergy. Secondly, the power source facility
facilitates cost reduction by preventing over-sizing through shared facilities like batteries.
Consequently, as costs escalate, synergy difference increases akin to batteries.
Therefore, in both systems, as CO2 removal rates rise, synergy difference escalates due to the
relatively larger impact of the second factor compared to the first.
Interestingly, in both systems, at each CO2 removal rate, the sum of synergy
differences for a 20% increase in the installed price of a major facility nearly equals zero. Thus,
irrespective of overarching technological advancements or regressions, the synergistic effects
exhibit a notable constancy when juxtaposed with those discerned in the foundational scenario.
In essence, the advantages conferred by sector coupling endure notwithstanding the evolution
of renewable energy technology.
The economic benefits of sector coupling are notably influenced by easing the
minimum operational constraints of the heat pump, which serves the DAC unit by converting
electricity into heat. In the baseline scenario, the large-scale heat pump's minimum operational
threshold is set at 0.4[42], with subsequent analyses exploring the effects of reducing this
threshold to 0.1 overcoming scale-up issues.[14, 39] Examining Figs. 7 (a) and (b), it's evident
that the TAC decreases as the minimum load of the heat pump diminishes, thereby enhancing
operating flexibility. The TAC ratio, defined as the ratio of TAC when the minimum load of
the heat pump is reduced to that when it's set at 40% (the base case), reflects this trend. A
smaller TAC ratio indicates a greater reduction in TAC compared to the base case. In both
systems, as the CO2 removal rate rises, the reduction in TAC ratio intensifies with decreasing
minimum load of the heat pump. For instance, at a CO2 removal rate of 0.1 Mt/yr, a 10%
increase in the operating flexibility of the heat pump leads to approximately a 1% decrease in
the TAC ratio. This reduction increases to about 2.7% at a CO2 removal rate of 1.3 Mt/yr.
Turning to Figs. 7 (c) and (d), the overall Synergy trend in the base case remains
consistent, albeit with slight variations. Specifically, in the sector-coupled system employing
PEM electrolyzer, synergy increases by up to 1% when the CO2 removal rate is below 0.9 Mt/yr
due to enhanced operating flexibility of the heat pump. Conversely, for CO2 removal rates
exceeding 1.1 Mt/yr, synergy experiences a minor decline as the operating flexibility of the
heat pump increases. Conversely, in the system with an alkaline electrolyzer, while the overall
trend aligns, synergy attributed to the operating flexibility of the heat pump gradually
diminishes as the CO2 removal rate rises. This phenomenon arises from the fact that as heat
pump flexibility increases, the synergy decreases due to improved economics of the standalone
DAC system. This observation emphasizes the symbiotic dynamics within the components of
the sector-coupled system and highlights the critical need for system-wide optimization to fully
leverage the collective advantages.

15
Fig. 6. TAC difference (MM$/yr) variation resulting from a 20% increased cost in the cost of
each facility while varying CO2 removal rate in the sector-coupled system: (a) PEM
electrolyzer-based sector-coupled system, (b) Alkaline electrolyzer-based sector-coupled
system, Synergy difference (%) variation resulting from a 20% increase in the cost of each
facility while varying the CO2 removal rate in the sector-coupled system: (c) PEM electrolyzer-
based sector-coupled system, (d) Alkaline electrolyzer-based sector-coupled system (light-blue
color indicates 0.1 Mt CO2/yr case, yellow color represents 0.5 Mt CO2/yr case, brown color
denotes 0.9 Mt CO2/yr case, and light-green color indicates 1.3 Mt CO2/yr).

16
Fig. 7. TAC ratio referenced in the base case across different heat pump minimum loads by
varying CO2 removal rate for (a) PEM electrolyzer-based sector-coupled system and (b)
Alkaline electrolyzer-based sector-coupled system. Synergy (%) across different heat pump
minimum loads by varying CO2 removal rate for (c) PEM electrolyzer-based sector-coupled
system and (d) Alkaline electrolyzer-based sector-coupled system (black color indicates the
base case, blue represents 30% minimum load of the heat pump case, red denotes 20%
minimum load of the heat pump case, and green represents 10% minimum load of the heat
pump case).

17
 Conclusion
This study presents an extensive optimization-based approach for integrating solid
Direct Air Capture (DAC) systems with Green Hydrogen (GH) production, aimed at addressing
the dual challenges of reducing anthropogenic CO2 emissions and enhancing the efficiency of
renewable energy usage. We concentrated on evaluating the economic benefits of sector
coupling within three distinct regions—Antofagasta, Bundaberg, and Neom—each
characterized by unique renewable energy capacities and meteorological conditions.

The key findings derived from our analysis are as follows:

 The combination of GH production with solid DAC systems delivers significant
economic benefits: ~10% for PEM electrolyzer, ~20% for alkaline electrolyzer.
 Alkaline electrolyzer superior PEM electrolyzer in the sector-coupled system although
the economic feasibility of standalone GH system was ~15% lower.
 Economic advantages derived from sector coupling are significantly affected by the
specific ratio of annual hydrogen production to the system's average annual CO2
reduction goal.
 The cost-effectiveness of sector coupling is enhanced by increased battery cost,
decreased DAC facility and electrolyzer cost, and impact of power sources cost
depends on CO2 removal targets.
 The robustness of sector coupling in enhancing efficiency and sustainability is
underscored by its enduring benefits, which remain significant irrespective of future
developments in renewable energy technology.
 Increased flexibility in operating constraints for heat pumps boosts the economic
viability of sector coupling, particularly for lower CO2 removal targets.
In summary, our study proposes the sector coupling strategy of integrating GH production with
solid DAC systems, and offers valuable insights into its economic advantages, contributing to
the broader discourse on sustainable energy solutions. Sector coupling gives more opportunity
to alkaline electrolyzer compared to PEM electrolyzer. These findings serve as a resource for
policymakers, industry stakeholders, and researchers committed to advancing a low-carbon
future.

Methods
The Two-stage Stochastic Programming (2SSP) approach is utilized to determine the
optimal system capacity and operational strategies that effectively manage uncertainties.[43,
44] This methodology requires decision-makers to make two sequential sets of decisions. The
first-stage decisions dictate the capacity of the various facilities, while the second-stage

18
decisions revolve around managing the energy system under uncertainties such as wind speed
variability, solar Global Horizontal Irradiance (GHI). Initially, first-stage decisions are made
with an incomplete knowledge of future uncertainties, focusing on system design choices such
as capacity and infrastructure investments. Subsequently, second-stage operational decisions
are made in response to the actual realization of weather scenarios, based on the infrastructure
established in the first stage. These decisions are aimed at optimizing capacity investment and
resilient energy management to real-world conditions.
The goal of our model is to minimize the total annualized cost (TAC), encompassing
the expenses related to installation, operation, and maintenance as below:

r  (1 + r ) L
i

min TAC =   IC  X i  ( CRF + OM )  where CRF =

i i i i
(1)
(1 + r ) L − 1
i
i

In above equation, for facility 𝑖, 𝐼𝐶 𝑖 stands for its investment cost, 𝑋𝑖 denotes the
capacity, 𝐶𝑅𝐹 𝑖 signifies the capital recovery factor, and 𝑂𝑀𝑖 represents the operating and
maintenance cost, assumed to proportional to the installation cost. The economic parameters
of facilities are summarized in below Table 1.
Equation (1) outlines the objective function, with constraints defined by energy
balance equations and mathematical models of the facilities detailed in the Supplementary
Materials, equation (A1-28). For example, the constraints for the standalone GH system are
specified in equation (A1-12), whereas the standalone DAC system is governed by equation
(A1-6) and (A13-25). The sector-coupled system's constraints are delineated in equation (A1-
25) along with the energy balance equations for each current configuration outlined in equation
(A26-28).
Details on generating weather and demand scenarios are provided in the
Supplementary materials. The IBM CPLEX solver is used to address the optimization
challenge. The mathematical programming codes are publicly accessible via the GitHub link
provided in the Data availability section.

Table 1. Economic parameters of facilities. (benchmarked in 2040)

Lifetime O&M cost

Component Investment cost Reference
(year) (% of installation cost)

Wind turbine 30 0.95 MM$/MW 2.4% [20, 45]

PV panel 35 0.6 MM$/MW 2% [20, 45]

Battery (8 h) 10 0.161 MM$/MWh 2.5% [20, 45]

PEM electrolyzer 10 0.44 MM$/MW 2% [20, 46]

19
 Alkaline
10 0.33 MM$/MW 2% [20, 46]
electrolyzer

Fuel cell 5 0.96 MM$/MW 2% [38]

Hydrogen tank 25 0.25 MM$/t H2 1% [20]

Solid DAC plant 30 2.76 MM$/t CO2 • h 4% [47]

Heat pump 25 0.0633 MM$/MW 0.36% [47]

Thermal energy
30 0.025 MM$/MWhth 0.02% [14]
storage

Data availability
The datasets generated and the original codes of mathematical programming are available from
the public GitHub by the following link: https://github.com/SunBooGermany/DAC-GH-
System. The detail parameters value and the description will be updated after the final
acceptance or reasonable request from editors and reviewers.

Conflicts of interest
There are no conflicts of interest to declare

Acknowledgements
The authors acknowledge that this work was supported by the KAIST-Aramco CO2
management center.

1. Liu, S., et al., Direct air capture of CO2via cyclic viologen electrocatalysis. Energy &
Environmental Science, 2024. 17(3): p. 1266-1278.
2. Sharifian, R., et al., Electrochemical carbon dioxide capture to close the carbon cycle.
Energy & Environmental Science, 2021. 14(2): p. 781-814.
3. Holmes, H.E., M.J. Realff, and R.P. Lively, Water management and heat integration
in direct air capture systems. Nature Chemical Engineering, 2024. 1(3): p. 208-215.
4. Jacobson, M.Z., The health and climate impacts of carbon capture and direct air
capture. Energy & Environmental Science, 2019. 12(12): p. 3567-3574.

20
5. Küng, L., et al., A roadmap for achieving scalable, safe, and low-cost direct air carbon
capture and storage. Energy & Environmental Science, 2023. 16(10): p. 4280-4304.
6. Breyer, C., et al., Direct Air Capture of CO2: A Key Technology for Ambitious Climate
Change Mitigation. Joule, 2019. 3(9): p. 2053-2057.
7. Bui, M., et al., Carbon capture and storage (CCS): the way forward. Energy &
Environmental Science, 2018. 11(5): p. 1062-1176.
8. Bachman, E., et al., Rail-based direct air carbon capture. Joule, 2022. 6(7): p. 1368-
1381.
9. Young, J., et al., The impact of binary water–CO2 isotherm models on the optimal
performance of sorbent-based direct air capture processes. Energy & Environmental
Science, 2021. 14(10): p. 5377-5394.
10. Sabatino, F., et al., A comparative energy and costs assessment and optimization for
direct air capture technologies. Joule, 2021. 5(8): p. 2047-2076.
11. Breyer, C., M. Fasihi, and A. Aghahosseini, Carbon dioxide direct air capture for
effective climate change mitigation based on renewable electricity: a new type of
energy system sector coupling. Mitigation and Adaptation Strategies for Global Change,
2019. 25(1): p. 43-65.
12. Young, J., et al., Process-informed adsorbent design guidelines for direct air capture.
Chemical Engineering Journal, 2023. 456.
13. Erans, M., et al., Direct air capture: process technology, techno-economic and socio-
political challenges. Energy & Environmental Science, 2022. 15(4): p. 1360-1405.
14. Yang, L. and X. Wu, Net-zero carbon configuration approach for direct air carbon
capture based integrated energy system considering dynamic characteristics of CO2
adsorption and desorption. Applied Energy, 2024. 358.
15. Shirizadeh, B., et al., Towards a resilient and cost-competitive clean hydrogen economy:
the future is green. Energy & Environmental Science, 2023. 16(12): p. 6094-6109.
16. Liu, H., et al., Pathway toward cost-effective green hydrogen production by solid oxide
electrolyzer. Energy & Environmental Science, 2023. 16(5): p. 2090-2111.
17. Staffell, I., et al., The role of hydrogen and fuel cells in the global energy system. Energy
& Environmental Science, 2019. 12(2): p. 463-491.
18. Park, J., et al., Techno-economic analysis of solar powered green hydrogen system
based on multi-objective optimization of economics and productivity. Energy
Conversion and Management, 2024. 299.
19. Won, W., et al., Design and operation of renewable energy sources based hydrogen
supply system: Technology integration and optimization. Renewable Energy, 2017. 103:
p. 226-238.
20. Terlouw, T., et al., Large-scale hydrogen production via water electrolysis: a techno-
economic and environmental assessment. Energy & Environmental Science, 2022.
15(9): p. 3583-3602.
21. Voronova, A., et al., Systematic degradation analysis in renewable energy-powered
proton exchange membrane water electrolysis. Energy & Environmental Science, 2023.
16(11): p. 5170-5184.
22. Hu, K., et al., Comparative study of alkaline water electrolysis, proton exchange
membrane water electrolysis and solid oxide electrolysis through multiphysics
modeling. Applied Energy, 2022. 312.
23. Park, J., et al., Green hydrogen to tackle the power curtailment: Meteorological data-
based capacity factor and techno-economic analysis. Applied Energy, 2023. 340.

21
24. Joos, M. and I. Staffell, Short-term integration costs of variable renewable energy:
Wind curtailment and balancing in Britain and Germany. Renewable and Sustainable
Energy Reviews, 2018. 86: p. 45-65.
25. Ramsebner, J., et al., The sector coupling concept: A critical review. WIREs Energy
and Environment, 2021. 10(4).
26. Bernath, C., G. Deac, and F. Sensfuß, Impact of sector coupling on the market value of
renewable energies – A model-based scenario analysis. Applied Energy, 2021. 281.
27. Gea-Bermúdez, J., et al., The role of sector coupling in the green transition: A least-
cost energy system development in Northern-central Europe towards 2050. Applied
Energy, 2021. 289.
28. Castro-Amoedo, R., et al., On the role of system integration of carbon capture and
mineralization in achieving net-negative emissions in industrial sectors. Energy &
Environmental Science, 2023. 16(10): p. 4356-4372.
29. Brown, T., et al., Synergies of sector coupling and transmission reinforcement in a cost-
optimised, highly renewable European energy system. Energy, 2018. 160: p. 720-739.
30. He, G., et al., Sector coupling via hydrogen to lower the cost of energy system
decarbonization. Energy & Environmental Science, 2021. 14(9): p. 4635-4646.
31. Chauhan, A. and R.P. Saini, A review on Integrated Renewable Energy System based
power generation for stand-alone applications: Configurations, storage options, sizing
methodologies and control. Renewable and Sustainable Energy Reviews, 2014. 38: p.
99-120.
32. Bukar, A.L., et al., A rule-based energy management scheme for long-term optimal
capacity planning of grid-independent microgrid optimized by multi-objective
grasshopper optimization algorithm. Energy Convers Manag, 2020. 221: p. 113161.
33. Bukar, A.L., C.W. Tan, and K.Y. Lau, Optimal sizing of an autonomous
photovoltaic/wind/battery/diesel generator microgrid using grasshopper optimization
algorithm. Solar Energy, 2019. 188: p. 685-696.
34. Liu, T., et al., Techno-economic feasibility of solar power plants considering PV/CSP
with electrical/thermal energy storage system. Energy Conversion and Management,
2022. 255.
35. Andersson, J. and S. Grönkvist, Large-scale storage of hydrogen. International Journal
of Hydrogen Energy, 2019. 44(23): p. 11901-11919.
36. Buttler, A. and H. Spliethoff, Current status of water electrolysis for energy storage,
grid balancing and sector coupling via power-to-gas and power-to-liquids: A review.
Renewable and Sustainable Energy Reviews, 2018. 82: p. 2440-2454.
37. Grigoriev, S.A., et al., Current status, research trends, and challenges in water
electrolysis science and technology. International Journal of Hydrogen Energy, 2020.
45(49): p. 26036-26058.
38. Nayak-Luke, R.M. and R. Bañares-Alcántara, Techno-economic viability of islanded
green ammonia as a carbon-free energy vector and as a substitute for conventional
production. Energy & Environmental Science, 2020. 13(9): p. 2957-2966.
39. Wiegner, J.F., et al., Optimal Design and Operation of Solid Sorbent Direct Air Capture
Processes at Varying Ambient Conditions. Industrial & Engineering Chemistry
Research, 2022. 61(34): p. 12649-12667.
40. Kim, S., et al., Multi-period, multi-timescale stochastic optimization model for
simultaneous capacity investment and energy management decisions for hybrid Micro-
Grids with green hydrogen production under uncertainty. Renewable and Sustainable
Energy Reviews, 2024. 190.

22
41. Niblett, D., et al., Review of next generation hydrogen production from offshore wind
using water electrolysis. Journal of Power Sources, 2024. 592.
42. Meesenburg, W., et al., Optimizing control of two-stage ammonia heat pump for fast
regulation of power uptake. Applied Energy, 2020. 271.
43. You, F., J.M. Wassick, and I.E. Grossmann, Risk management for a global supply chain
planning under uncertainty: Models and algorithms. AIChE Journal, 2009. 55(4): p.
931-946.
44. Yeh, K., et al., Two stage stochastic bilevel programming model of a pre-established
timberlands supply chain with biorefinery investment interests. Computers & Chemical
Engineering, 2015. 73: p. 141-153.
45. Vimmerstedt, L., et al., Annual Technology Baseline: The 2022 Electricity Update.
2022, National Renewable Energy Lab.(NREL), Golden, CO (United States).
46. Böhm, H., et al., Projecting cost development for future large-scale power-to-gas
implementations by scaling effects. Applied Energy, 2020. 264.
47. Fasihi, M., O. Efimova, and C. Breyer, Techno-economic assessment of CO2 direct air
capture plants. Journal of Cleaner Production, 2019. 224: p. 957-980.

23
 Integrating solid direct air capture systems with green hydrogen
production: Economic synergy of sector coupling

Supporting information

Sunwoo Kim1,2, Joungho Park3, and Jay H. Lee1*

1
Mork Family Department of Chemical Engineering and Material Sciences, University of
Southern California, 925 Bloom Walk, Los Angeles, CA 90089, USA
2
Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of
Science and Technology (KAIST), 291, Daehak-ro, Yuseong-gu, Daejeon, 34141,
Republic of Korea
3 Energy AI & Computational Science Laboratory, Korea Institute of Energy Research, 71-
2 Jang-dong, Yuseong-gu, Daejeon, 305-343, Republic of Korea
*
Corresponding author’s E-mail: [email protected]
Table of contents
A. Models used………………………………………………………………………………2
A. 1. Wind turbine
A. 2. Solar PV panel
A. 3. Battery
A. 4. Water electrolyzer
A. 5. Fuel cell
A. 6. Gaseous hydrogen tank
A. 7. Electrified heat pump
A. 8. Thermal energy storage
A. 9. Solid DAC plant
B. Sector coupled system description of DC configuration and AC configuration…………10
C. Weather and demand scenarios generation………………………………………………12
D. Influence of number of weather scenario used in Two-stage stochastic programming ....13
E. Influence of the duration of battery in the economic viability…………………………...14
F. Reference…………………………………………………………………………………15

24
A. Models used
A. 1. Wind turbine
The model for a 5-MW turbine from the National Renewable Energy Laboratory
(NREL) is utilized, formulated as follows:[1]

 C  X T  (  WS sc ,t 3 +  ) for WS ci  WS sc ,t  WS r
PscT,t =  C  X T  P rated for WS r  WS sc ,t  WS co (A1)
0 for WS sc ,t
 WS or WS
ci sc ,t
 WS co

𝑇
In this equation, 𝑃𝑠𝑐,𝑡 represents the power output from the turbine, 𝜂𝐶 denotes the efficiency
of the appropriate converter, and 𝑋𝑇 represents the capacity of turbine. This model includes
critical wind speed thresholds: the cut-in speed (𝑊𝑆 𝑐𝑖 ), rated speed (𝑊𝑆 𝑟 ), and cut-out speed
(𝑊𝑆 𝑐𝑜 ). The parameters 𝛼 and 𝛽 describe the turbine’s power output characteristics for
wind speeds lying between the cut-in and cut-out thresholds. The power increases with wind
speed until reaching the rated power rate (𝑃𝑟𝑎𝑡𝑒𝑑 ). The turbine ceases to generate power if the
wind speed falls below the cut-in speed or surpasses the cut-out speed.

A. 2. Solar PV panel
The power generated from a solar photovoltaic (PV) panel is defined by the following
equation:[2, 3]

  NOCT − 20   GHI
sc ,t
P P
sc ,t =   X P  1 +    T sc ,t + GHI sc ,t 
C
− T ref   (A2)
  800   1000
𝑃
In this equation, 𝑃𝑠𝑐,𝑡 represents the power generated from the PV panel, 𝜂𝐶 denotes the
efficiency of the appropriate converter, and 𝑋𝑃 is the PV panel’s capacity. γ (-0.0037 K-1), is
the temperature coefficient, affecting the panel’s performance relative to the cell temperature
(𝑇 𝑠𝑐,𝑡 ). 𝐺𝐻𝐼 𝑠𝑐,𝑡 stands for the global horizontal solar irradiance (GHI) at a specific scenario
and time, the normal operating cell temperature (𝑁𝑂𝐶𝑇) is set at 45 ℃, and 𝑇 𝑟𝑒𝑓 (298.15 K)
represents the reference ambient temperature.

A. 3. Battery
This study employs a 8-hour battery system, designed to complete a full charge and
discharge cycle within a 8-hour timeframe, as detailed in eqn (A3) and (A4). The dynamics of
energy storage within the battery (𝐸𝑆𝑆𝑠𝑐,𝑡 , unit: MWh) are governed by various factors such as
the self-discharging rates (𝜎 𝐵 ), the charging (𝑐ℎ𝑠𝑐,𝑡 , unit: MW) and discharging (𝑑𝑐ℎ𝑠𝑐,𝑡 , unit:
MW) rates, and the overall battery efficiency (𝜂𝐵 ), as elaborated in eqn (A5). The self-
discharging rate is assumed to be 0.0083% per hour and the efficiency is assumed to be 95%.
Furthermore, we stipulate that the battery’s storage level should be maintained within 15% to
95% of its rated capacity, as expressed in eqn (A6).[4]
25
 0  chsc ,t  0.125  X B (A3)

0  dchsc,t  0.125  X B (A4)

 
ESSsc ,t = (1 −  B )  ESSsc ,t −1 +  C  B  chsc ,t − C
1
 dchsc ,t   t (A5)
   B


0.15  X B  ESSsc ,t  0.95  X B (A6)

A. 4. Water electrolyzer
This study utilizes a proton exchange membrane (PEM) water electrolyzers and
Alkaline electrolyzers to split water into hydrogen and oxygen gases. The PEM electrolyzers
employ a solid polymer membrane to conduct protons from the anode to the cathode,
effectively preventing the mixing of gases. In contrast, Alkaline electrolyzers employ an anion
exchange membrane for gas separation. This distinction in membrane type fundamentally alters
the electrolysis process, as alkaline electrolyzers utilize an aqueous alkaline solution as the
electrolyte. By utilizing a liquid electrolyte, alkaline electrolyzers eliminate the need for the
expensive and fragile proton exchange membrane, potentially offering cost advantages and
improved durability compared to PEM electrolyzers. Despite of the high cost, PEM
electrolyzers have notable benefits by high operational efficiency and swift response times to
demand fluctuations. Furthermore, the PEM electrolyzer is distinguished by its ability to
produce high-purity hydrogen gas and its capacity to operate efficiently at low minimum loads
(𝑚𝑖𝑛𝑊 )—significantly lower than those required by alkaline electrolyzers (20%), with a
typical minimum load of 5%.[5, 6]. The assumed energy efficiency ( 𝜂𝑊 ) for the PEM
electrolyzer is 48 kWh per kilogram of hydrogen, and 50 kWh per kilogram of hydrogen for
alkaline electrolyzer. The unit of 𝑝2ℎ𝑠𝑐,𝑡 and the unit of 𝑋𝑊 is MW. The operational
parameters and constraints of the PEM electrolyzer are detailed in the subsequent equation.

minW  X W  p 2hsc ,t  X W (A7)

A. 5. Fuel cell
This research utilizes a PEM fuel cell for the generation of electricity from hydrogen
stored. Fuel cells are innovative devices that directly convert the chemical energy of a fuel into
electrical energy, presenting clear advantages over conventional combustion technologies. A
primary advantage of hydrogen fuel cells is their environmental friendliness; they emit only
water as a byproduct, significantly contributing to the reduction of CO2 emissions and other air
26
pollutants.[7] For the purposes of this study, the energy conversion efficiency (𝜂𝐹𝐶 ) of the PEM
fuel cells is assumed to be 20.4 kWh per kilogram of hydrogen. The unit of both ℎ2𝑝𝑠𝑐,𝑡 and
𝑋𝐹𝐶 is MW. The operational parameters and constraints governing the use of the PEM fuel
cell are outlined in the equation that follows.

0  h2 psc ,t  X FC (A8)

A. 6. Gaseous hydrogen tank

This research utilizes a gaseous HT system, designed to store hydrogen under high
pressure to efficiently accommodate large volumes of hydrogen gas.[8] The volume of
hydrogen stored in the HT (𝐻𝑇𝑠𝑐,𝑡 , unit: ton of hydrogen) is determined by various factors,
including the self-leakage rate (𝜎 𝐻𝑇 ), the rate of hydrogen production from the electrolyzer,
and the rate of hydrogen consumption by the fuel cell, as detailed in eqn (A9). For every week,
𝐻
there is a demand for hydrogen (𝐷𝑠𝑐,𝑤 ) to be satisfied. Consequently, the balance equation for
the HT is adjusted accordingly to reflect this demand, as shown in eqn (A10). The self-leakage
rate of the HT is assumed to be 0.0104% per hour, accounting for the minimal loss of stored
hydrogen over time due to imperfections in the tank's containment system. Additionally, it is
crucial that the HT's storage level does not surpass its designed capacity to ensure safety and
maintain the integrity of the storage system, as indicated in eqn (A11). This constraint is
essential for preventing overpressure scenarios and ensuring the operational reliability of the
hydrogen storage system

 C 
HTsc ,t = (1 −  HT )  HTsc ,t −1 +  W  p 2hsc ,t − FC  h2 psc ,t   t
1
(A9)
  

 C 
HTsc ,t = (1 −  HT )  HTsc ,t −1 +  W  p 2hsc ,t − FC  h2 psc ,t   t − DscH, w
1
(A10)
  

0  HTsc,t  X HT (A11)

The energy balance equation for the GH system can be expressed as below:

PscT,t + PscP,t − chsc ,t + dchsc ,t − outsc ,t +  C  h 2 psc ,t − p 2hsc ,t = 0 (A12)

A. 7. Electrified heat pump

The electrified heat pump in this study is designed to convert electrical power into
thermal energy, specifically generating superheated steam, by harnessing electricity. It operates
with a minimum operating ratio (𝑝2𝑡 𝑚𝑖𝑛 ) of its rated capacity (𝑋𝐻𝑃 , unit: MW), which is a
27
critical parameter for ensuring efficient operation without underutilization of its potential, as
elaborated in eqn (A13), where the unit of 𝑝2𝑡𝑠𝑐,𝑡 is also MW. Initially, 𝑝2𝑡 𝑚𝑖𝑛 is set at 0.4
for the base case analysis, indicating that the heat pump must operate at least 40% of its rated
capacity.[9] To explore the economic implications of varying operational efficiencies, further
analysis includes scenarios where 𝑝2𝑡 𝑚𝑖𝑛 is reduced to 0.1, assessing the impact of increased
flexibility on system economics.[10, 11] The heat pump's coefficient of performance (COP) is
assumed to be 3.41, signifying that for every unit of electrical energy consumed, the heat pump
produces 3.41 units of thermal energy.[12]

p 2t min  X HP  p 2tsc ,t  X HP (A13)

A. 8. Thermal energy storage

This study incorporates the use of a superheated steam-based TES system, designed to
operate at 40% of its rated capacity (𝑋𝑇𝐸𝑆 , unit: MWth) for both charging and discharging
process, as expressed in eqn (A14) and (A15). The quantity of thermal energy stored in the
TES (𝑇𝐸𝑆𝑠𝑐,𝑡 , unit: MWhth) is influenced by a variety of factors including the self-discharge
rate (𝜎 𝑇𝐸𝑆 ), charging (𝑇𝑐ℎ𝑠𝑐,𝑡 , unit: MWth) and discharging (𝑇𝑑𝑐ℎ𝑠𝑐,𝑡 , unit: MWth) rates, and
the overall efficiency of the TES system (𝜂𝑇𝐸𝑆 ), as elaborated in eqn (A16). The self-discharge
rate accounting for thermal energy losses over time even without active discharge, is set at 0.5%
per hour, while the overall efficiency is assumed to be 98%, indicating a high level of energy
conservation during storage operations.[10] Additionally, it is stipulated that the TES system's
energy storage level is regulated to stay within 10% to 90% of its rated capacity, as expressed
in eqn (A17). This range ensures optimal operation and longevity of the storage system by
preventing conditions that could lead to underutilization or overstrain of the TES capabilities,
thereby maintaining its effectiveness and reliability in energy storage and retrieval operations.

Tchsc ,t  0.4  X TES (A14)

Tdchsc ,t  0.4  X TES (A15)

 
TES sc ,t = (1 −  TES )  TES sc ,t −1 +  TES  Tchsc ,t − TES  Tdchsc ,t   t
1
(A16)
  

0.1 X TES  TESsc ,t  0.9  X TES (A17)

28
A. 9. Solid DAC plant
This study employs a solid DAC plant, comprising numerous modularized units, each
with a rated capacity of 50 t CO2 per year.[13] These modularized units are designed to operate
independently, allowing for flexibility in their operation. There is a distinct sequential
constraint between the two operational phases of the DAC process, as well as the possibility of
shutdowns, as illustrated in below figure:

Fig. A. 1. Sequential batch process of the solid DAC system, detailing the operational phases
and shutdown periods. In the figure, (a) indicates the adsorption phase where CO2 is captured
from the atmosphere, (b) represents a shutdown period following the adsorption phase,
allowing for any necessary system adjustments or maintenance, (c) denotes the desorption
phase during which the captured CO2 is released from the adsorbent material for storage or
use, and (d) represents a shutdown period after desorption, preparing the system for the next
adsorption cycle. This sequence ensures the efficient capture and release of CO2 in a cyclic
manner.

Fig. A. 1 illustrates the operational phases of the solid DAC system, focusing on the
dynamics of CO2 capture and the associated energy consumption. Fig. A. 1 (a) showcases the
adsorption phase, where ventilation is activated to increase air flow over the adsorbent,
enhancing CO2 capture given its low atmospheric concentration.[10] During this phase, CO2 is
adsorbed to the solid absorbent and the remaining air is expelled, which requires electrical
energy for the ventilation system. Fig. A. 1 (b) details the post-adsorption shutdown. After
adsorption, closing the air passages of the modular unit traps CO2 within. This stage does not
consume energy, presenting an opportunity for operational pause when renewable energy
29
supply is low, retaining the captured CO2 for subsequent desorption. Fig. A. 1 (c) describes the
desorption phase, where the adsorbent is heated to approximately 100 ℃ to release the captured
CO2.[11] A vacuum pump is employed to aid in CO2 collection, marking this phase as the most
energy-intensive, consuming both electrical and thermal energy. Fig. A. 1 (d) covers the post-
desorption shutdown, an energy-free phase allowing for operational breaks following CO2
release, preparing the system for the next adsorption cycle.
The sequential operational relationship between these phases is further elaborated in
Fig. A. 2 and defined through eqn (A18-23). A unit in state (a) can progress to state (b) or (c);
from state (b), it can remain or move to state (c); in state (c), it can shift to state (d) or revert to
state (a); and from state (d), it can either stay or transition back to state (a). This cyclical and
strategic progression between states underscores the system's operational flexibility and energy
management in solid DAC processes.

Fig. A. 2. Description of variables and their interrelations across the various phases of solid
DAC system, highlighting the energy consumption characteristics of each phase.

30
 0  dac1sc ,t , dacsc2 ,t , dacsc3 ,t , dacsc4 ,t  2  X DAC (A18)

dac1sc ,t + dacsc2 ,t + dacsc3 ,t + dacsc4 ,t = 2  X DAC (A19)

dact1+1, sc  dact3, sc + dact4, sc (A20)

dact3+1, sc  dact1, sc + dact2, sc (A21)

dact2+1, sc = dact2, sc + dact1, sc − dact3+1, sc (A22)

dact4+1, sc = dact4, sc + dact3, sc − dact1+1, sc (A23)

𝑘
In above equations, 𝑑𝑎𝑐𝑠𝑐,𝑡 (unit: t CO2/hr) denotes the capable CO2 flow rate of DAC plant
at phase 𝑘, and 𝑋𝐷𝐴𝐶 (unit: t CO2/hr) represents the capability of DAC plant, reducing CO2
through the entire process encompassing the adsorption and the desorption. Here, the
multiplication by 2 accounts for the complete cycle time.
In this analysis, given our focus on achieving CO2 reduction targets ranging from 0.1
Mt to 1.3 Mt per year, the total count of modular units required falls within the thousands to
tens of thousands range. Consequently, while the variable representing the quantity of solid
DAC system units is inherently discrete, for the purposes of this study, it is treated as a
continuous variable to simplify calculations and modeling. Furthermore, it is assumed that all
units operate with uniform energy efficiency and cycle times, with both the adsorption and
desorption phases lasting 1 hour each.
For the adsorption phase, the energy consumption is estimated at 0.07 MWh per ton
of CO2 captured (𝑒 1 ). During the desorption phase, the energy requirements are assumed to be
0.14 MWh of electrical energy (𝑒 3 ) and 1.286 MWh of thermal energy (ℎ3 ) per ton of CO2.[12]
These assumptions allow for a simplified yet realistic approximation of the energy inputs
necessary for the operation of the solid DAC system. The thermal and electricity energy
balance equations for the solid DAC system can be expressed in eqn (A24) and eqn (A25),
respectively.

31
  C  COP  p 2tsc ,t − h3  dacsc3 ,t − Tchsc ,t + Tdchsc ,t = 0 (A24)

PscT,t + PscP,t − chsc ,t + dchsc ,t − outsc ,t − e1  dac1sc ,t − e3  dacsc3 ,t − p 2t sc ,t = 0 (A25)

Energy balance equations for the combined system vary according to the specific
current configuration and are summarized below:
DC or AC configuration:

PscT,t + PscP,t − chsc,t + dchsc,t − outsc,t +  C  h2 psc,t − p 2hsc,t − e1  dac1sc,t

(A26)
−e3  dacsc3 ,t − p 2tsc ,t = 0

AC-DC configuration:

PscP,t − chsc ,t + dchsc ,t +  C  h 2 psc ,t − p 2hsc ,t − D 2 Asc ,t +  C  A2 Dsc ,t = 0 (A27)

PscT,t − chsc2 ,t + dchsc2 ,t − outsc ,t +  C  D 2 Asc ,t − A2 Dsc ,t − e1  dac1sc ,t

(A28)
−e3  dacsc3 ,t − p 2tsc ,t = 0

B. Sector coupled system description of DC configuration and AC configuration.

The diagrams illustrate the sector coupled system of DC configuration and AC configuration
are presented here. Among the current conversion configurations analyzed, the DC to AC setup
is identified as the most efficient for the integrated system, as demonstrated in Fig. B. 1-2. The
economic advantages stemming from this sector-coupling depend significantly on the current
configuration as illustrated in Fig. B. 3. The AC-DC hybrid current configuration superior the
other two current configurations for all regions and CO2 removal rate in both electrolyzers.

32
Fig. B. 1. Sector coupled system that merges Green Hydrogen (GH) production and solid Direct
Air Capture (DAC), operating entirely in a Direct Current (DC) configuration. This system is
engineered to meet hydrogen demand promptly and achieve a specified annual CO 2 reduction
goal throughout the entirety of the project.

Fig. B. 2. Sector coupled system combining Green Hydrogen (GH) production with solid Direct
Air Capture (DAC), functioning in an Alternating Current (AC) configuration. This
arrangement aims to ensure the prompt fulfillment of hydrogen demand while also meeting
33
Fig. B. 3. Improvement (MM$/yr) and Synergy (%) of sector coupling for each current
configuration across different regions by varying CO2 removal rate: (a) PEM electrolyzer-
based sector coupled system, (b) Alkaline electrolyzer-based sector coupled system.

C. Weather and demand scenarios generation.

In this study, we conducted an extensive analysis in three regions to evaluate the improvement
and characteristics of the proposed sector coupled system. We utilized hourly solar GHI data
and wind speed data at observatory height obtained from National Solar Radiation Database
(NSRDB) as summarized in Table C. 1.[14] The wind speed at 100 meters above the ground
was compensated by using the Wind Atlas[15] to standardize these measurements for
comparison. The wind speeds were adjusted to a reference height using the formula below:[16]

h hobservatory
ws sc ,t
= ws
sc ,t
observatory  ln  ln roughness (C1)
z roughness z
𝑠𝑐,𝑡
In above equation, 𝑤𝑠𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑜𝑟𝑦 is the wind speed at the observatory, ℎ is the hub height,
𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑜𝑟𝑦
ℎ indicates the observatory height, and 𝑧 𝑟𝑜𝑢𝑔ℎ𝑛𝑒𝑠𝑠 denotes the roughness factor.

Table. C. 1. Summary of selected years and locations of the solar GHI and wind speed
data.[17]

Latitude Longitude Year

Antofagasta -23.42 -70.49 2019–2021

Bundaberg -24.87 152.45 2018–2020
Neom 27.97 35.50 2017–2019
*The spatial resolution is 2km by 2km and the temporal resolution is hourly.

34
 D. Influence of number of weather scenario used in Two-stage stochastic programming
The current state of the art regarding the integration of weather data into the economic
evaluation and design process of renewable energy systems lacks uniformity, primarily due to
regional disparities and the specific objectives of the system in question. The incorporation of
extensive seasonal data offers the advantage of fostering resilience and accuracy in system
design and economic assessment. However, this approach amplifies the complexity of the
optimization task exponentially, thus necessitating a careful balance.
In our recent session, we undertook an empirical study utilizing three annual weather data to
assess economic viability. We juxtaposed the economic outcomes derived from one-year and
two-year data averages against the comprehensive three-year dataset. Notably, employing a
solitary year's data yielded discrepancies ranging from 2% to 6%, indicating inadequacy,
whereas leveraging two years' data demonstrated an error margin within 1%. Consequently,
we opted for the utilization of the complete three-year seasonal dataset to ensure robustness
and accuracy in our economic evaluations.

Fig. D. 1. Difference (%) variation by used annual scenario number: (a) PEM electrolyzer-
based standalone green hydrogen system, (b) Alkaline electrolyzer-based standalone green
hydrogen system, (c) Standalone DAC system, (d) Sector coupled system.

35
 E. Influence of the duration of battery in the economic viability
We assess the economic viability of each system by varying the duration of battery usage. As
elaborated in the preceding section, the duration of battery storage varies based on the energy
capacity of the battery stack and the power rate capabilities of the system. Longer durations
correlate with reduced power rate capabilities and lower unit costs per megawatt-hour (MWh),
indicating a trade-off. A summary of the economic parameters is presented in Table E. 1.
We evaluate the economic feasibility of different battery durations and compute the TAC ratio
relative to the economic value derived from utilizing a 4-hour battery for each system. As
depicted in Fig. E. 1, the PEM electrolyzer-based standalone GH system favors a 6-hour battery
duration, although the discrepancy among durations is minimal. Conversely, the alkaline
electrolyzer exhibits significant variance among battery durations, with the 6-hour option
offering the lowest cost. For both the standalone DAC system and the sector-coupled system,
an 8-hour battery duration yields the lowest cost.

Table E. 1. Economic parameter for each duration of battery (benchmarked in 2040).

Battery (4 h) 10 0.19 MM$/MWh 2.5% [18, 19]

Battery (6 h) 10 0.171 MM$/MWh 2.5% [18, 19]

Battery (8 h) 10 0.161 MM$/MWh 2.5% [18, 19]

Battery (10 h) 10 0.155 MM$/MWh 2.5% [18, 19]

36
Fig. E. 1. TAC ratio of each system across different regions by varying duration of battery: (a)
PEM electrolyzer-based standalone GH system, (b) Alkaline electrolyzer-based standalone GH
system, (c) Standalone DAC system, (d) Sector coupled system.

F. Reference.
1. Chauhan, A. and R.P. Saini, A review on Integrated Renewable Energy System based
power generation for stand-alone applications: Configurations, storage options, sizing
methodologies and control. Renewable and Sustainable Energy Reviews, 2014. 38: p.
99-120.
2. Bukar, A.L., et al., A rule-based energy management scheme for long-term optimal
capacity planning of grid-independent microgrid optimized by multi-objective
grasshopper optimization algorithm. Energy Convers Manag, 2020. 221: p. 113161.
3. Bukar, A.L., C.W. Tan, and K.Y. Lau, Optimal sizing of an autonomous
photovoltaic/wind/battery/diesel generator microgrid using grasshopper optimization
algorithm. Solar Energy, 2019. 188: p. 685-696.
4. Liu, T., et al., Techno-economic feasibility of solar power plants considering PV/CSP
with electrical/thermal energy storage system. Energy Conversion and Management,
2022. 255.
5. Buttler, A. and H. Spliethoff, Current status of water electrolysis for energy storage,
grid balancing and sector coupling via power-to-gas and power-to-liquids: A review.
Renewable and Sustainable Energy Reviews, 2018. 82: p. 2440-2454.
37
6. Grigoriev, S.A., et al., Current status, research trends, and challenges in water
electrolysis science and technology. International Journal of Hydrogen Energy, 2020.
45(49): p. 26036-26058.
7. Nayak-Luke, R.M. and R. Bañares-Alcántara, Techno-economic viability of islanded
green ammonia as a carbon-free energy vector and as a substitute for conventional
production. Energy & Environmental Science, 2020. 13(9): p. 2957-2966.
8. Andersson, J. and S. Grönkvist, Large-scale storage of hydrogen. International Journal
of Hydrogen Energy, 2019. 44(23): p. 11901-11919.
9. Meesenburg, W., et al., Optimizing control of two-stage ammonia heat pump for fast
regulation of power uptake. Applied Energy, 2020. 271.
10. Yang, L. and X. Wu, Net-zero carbon configuration approach for direct air carbon
capture based integrated energy system considering dynamic characteristics of CO2
adsorption and desorption. Applied Energy, 2024. 358.
11. Wiegner, J.F., et al., Optimal Design and Operation of Solid Sorbent Direct Air Capture
Processes at Varying Ambient Conditions. Industrial & Engineering Chemistry
Research, 2022. 61(34): p. 12649-12667.
12. Fasihi, M., O. Efimova, and C. Breyer, Techno-economic assessment of CO2 direct air
capture plants. Journal of Cleaner Production, 2019. 224: p. 957-980.
13. McQueen, N., et al., A review of direct air capture (DAC): scaling up commercial
technologies and innovating for the future. Progress in Energy, 2021. 3(3).
14. Sengupta, M., et al., The National Solar Radiation Data Base (NSRDB). Renewable
and Sustainable Energy Reviews, 2018. 89: p. 51-60.
15. Davis, N.N., et al., The Global Wind Atlas: A high-resolution dataset of climatologies
and associated web-based application. Bulletin of the American Meteorological
Society, 2023.
16. Wever, N., Quantifying trends in surface roughness and the effect on surface wind
speed observations. Journal of Geophysical Research: Atmospheres, 2012. 117(D11).
17. Kim, S., et al., Techno-economic analysis for design and management of international
green hydrogen supply chain under uncertainty: An integrated temporal planning
approach. Energy Conversion and Management, 2024. 301.
18. Terlouw, T., et al., Large-scale hydrogen production via water electrolysis: a techno-
economic and environmental assessment. Energy & Environmental Science, 2022.
15(9): p. 3583-3602.
19. Vimmerstedt, L., et al., Annual Technology Baseline: The 2022 Electricity Update.
2022, National Renewable Energy Lab.(NREL), Golden, CO (United States).

38