Green ammonia production using current and emerging electrolysis technologies

,


Introduction
The chemical industry and the transportation sector will have to transition to green routes to produce high-demand chemicals like ammonia to align with climate policy targets.Ammonia is an extremely important chemical used in the fertilizer industry.It has also garnered interest as a green fuel for shipping [1], as well as an energy storage medium capable of overcoming hydrogen storage challenges [2].Green ammonia production is not a new concept, as it has been implemented since the late 1920s until the 1990s in Norway, utilizing AEC as a green hydrogen pathway powered by hydropower [3].However, the economic attractiveness of fossil hydrogen produced from natural gas through steam methane reforming (SMR) has made it a preferred choice for ammonia production in most locations and for volume production.Nonetheless, the mounting environmental concerns, advancements in water electrolysis technologies, and the significant cost reduction in renewable electricity over the past decade have reignited substantial interest in green ammonia production.Compared with hydrogen, ammonia has some advantages.It is much easier to liquefy and store at higher temperatures (− 33 vs. − 253 • C at ambient pressure) and its transport is at least three times cheaper than hydrogen [4].Besides, ammonia can be used in internal combustion engines [5] and gas turbines [6] with modest modifications of the core technology.The levelized cost of electricity (LCOE) from green ammonia in large-scale power plants (combustion in combined cycle gas turbines) is forecasted by Cesaro et al. [7].They found that at power plant capacity factors below 25 % (increasingly common in electricity sectors with high variable renewable electricity), the cost of ammonia should be below 400 USD/tNH 3 to effectively compete with other dispatchable, low or zero-carbon technologies, including gas, bioenergy, or coal-fired power plants equipped with CCS.In addition, published research works in the last years suggest ammonia as the future marine fuel [8,9].Many investigations exist in the literature addressing the green ammonia economy and technology readiness.Some of the relevant published works are reviewed hereunder.
Campion et al. [10] conducted a techno-economic assessment of green ammonia production, considering diverse wind and solar potentials, weather profiles, and electrolyzer technologies.Optimization findings indicated that employing a semi-islanded configuration stands out as the most cost-effective choice, potentially resulting in cost reductions of up to 23 % when compared to off-grid systems.However, it is found that this approach leads to greenhouse gas emissions comparable to those of using fossil fuels for ammonia production when considering the current electricity mix.When evaluating off-grid systems, the practice of estimating costs based on the levelized cost of electricity and capacity factors for solar or wind sources (which in turn determines operating hours) resulted in an overestimation of costs by as much as 30 %.The most economical off-grid setup has attained a production cost of 842.€/tNH 3 .Islanded green ammonia and hydrogen production and export from Saudi Arabia is optimized by Florez et al. [11].Their results showed that plant optimization can lead to production costs of ammonia of 383 USD/tNH 3 , while the single largest cost driver is the electricity cost, representing 60-70 % of the total project cost.
Frattini et al. [12] evaluated three distinct pathways for green ammonia production.These routes were benchmarked against the production of fossil ammonia through steam methane reforming and involved the utilization of renewable hydrogen obtained through SOEC, biomass gasification, and biogas reforming.They found that all the systems have the same energy consumption of 14-15 MW/tNH 3 .The CO 2 equivalent emissions are estimated to be 2.05, 3.82 and 3.59 kgCO 2 /kgNH 3 when natural gas, syngas and biogas are used as feedstock, respectively.Biomass-based ammonia can be considered carbon neutral due to the renewable character of the feedstock.However, it is important to be aware that biomass availability may become limited in future energy systems [13].
Zhang et al. [14] investigated and compared various ammonia production routes, including power-to-ammonia, biomass-to-ammonia, and methane-to-ammonia.The study concludes that power-to-ammonia represents the most efficient route, boosting LHV-based efficiency to 74 % ( Mass flow rate of produced ammonia×LHVNH3 Total consumed electricity ) but it is currently not economically attractive due to the high cost of electrolyzer stacks and high electricity prices.Nevertheless, it has the potential to become cost-competitive if the stack cost can be reduced to 920 $/m 2 .Bicer and Dincer [15] compared the feasibility of nuclear-driven SOEC and copper-chlorine cycle as green hydrogen routes to produce ammonia.It was found that employing electrolyzers has a lower environmental impact than the copper-chlorine cycle (460 vs. 580 tCO 2 eq per tNH 3 ).
Ammonia as a hydrogen storage carrier was studied by Al-Zareer et al. [16].The hydrogen storage system was proposed to chemically store hydrogen as ammonia, which was pressurized in a multistage ammonia production system.The produced ammonia was stored in a novel conceptual design of an ammonia tanker truck equipped with an ammonia electrolyzer for the transportation and delivery of the stored hydrogen.It has been assumed that the tanker truck can either provide the ammonia to end-users who decompose the ammonia later when they need the hydrogen or deliver hydrogen directly produced from ammonia decomposition.The proposed system demonstrated energy and exergy efficiencies of 72.3 % and 71.8 %, respectively.SOEC-based green ammonia production using nitrogen generated via a series of solid electrolyte oxygen pumps was studied by Nowicki et al. [17].Solid electrolyte oxygen pumps are devices that leverage the movement of oxide ions within dense electrolyte materials to achieve the separation of oxygen from gas mixtures (here air).The specific energy consumption was found to be 9.9 kWh/kgNH 3 (efficiency of 52.1 %).Scaling options for the trade of green ammonia are studied by Egerer et al. [18] assuming the import of ammonia from Australia to Germany in 2030 as the case study.With the production costs of 30.43 €/MWh electricity supply in Australia, the levelized cost for green ammonia at the German harbor is calculated to be 109.39€/MWh (566.64 €/t).They concluded that green ammonia could achieve cost parity with natural gas-based ammonia, even at moderate gas prices (around 30 €/MWh), given a sufficiently high CO 2 tax (approximately 180 €/tCO 2 ).They also indicated that cracking ammonia to generate pure hydrogen introduces a 45 % cost increase per MWh at the destination.The impact of climate change on the cost of production of green ammonia from offshore wind at four locations in the UK was studied by Hattonet al. [19].Using an islanded green ammonia production model, the achievable cost of ammonia was evaluated to range between 935 and 1696 USD/tNH 3 .Driscoll et al. [20] evaluated offshore green ammonia production using tidal and wind energy from techno-economic perspective.They found that the addition of tidal stream capacity to wind capacity decreases the hydrogen storage requirement by 96 % and reduces the cost of ammonia by 12 %.
Table 1 summarizes the rest of the reviewed works, all dealing with different electrolysis options and different sources of renewable electricity.
The literature review highlights the extensive research on technoeconomic assessment conducted on various aspects of green ammonia production, particularly concerning existing technologies.However, it is crucial to acknowledge the rapid advancements in electrolysis and ongoing investigations into modifying existing technologies.Consequently, considering the projected future cost developments is of paramount importance for effective energy planning and policy-making.
In this work, the existing AEC and SOEC technologies are compared at their current maturity level, addressing projections for future developments through a comprehensive thermodynamic and technoeconomic analysis of green ammonia plants designed based on these two technologies.To investigate the effect of various potential technological advancements for both AEC and SOEC, high-pressure AEC is modeled in both low-temperature (LTP-AEC) and high-temperature (HTP-AEC) versions, and SOEC is modeled considering both lowpressure (LP-SOEC) and high-pressure (HP-SOEC) technologies.Furthermore, an estimation of the future cost of green ammonia, incorporating the current and projected capital expenditure (CAPEX) values of electrolysis technologies is included considering different OPEX values based on LCOE-estimates for the powering technologies.This analysis not only offers insights into anticipated costs but also provides a valuable tool for comparing the advantages of different technology improvements.The main findings of this study are related to the electrolyzers' technical characteristics, the implications of which are reflected in the economic analysis of green ammonia production.It is important to note that while this study provides general insights, a more detailed economic analysis tailored to specific locations and their respective local conditions (such as LCOE and capacity factors) is necessary to achieve solid results on green ammonia cost and conditions where it can become cost-competitive.These site-specific considerations may alter some of the economic conclusions presented here, and the economic assessments presented here should be considered as guidance rather than definitive conclusions.The technical results presented here can of course be transferred to an economic analysis for a specific site.

System development and mathematic modeling
The main units of the proposed green ammonia production systems are the electrolyzer (SOEC or AEC), the Haber-Bosch loop (HBL), and the air separation unit (ASU).Each unit is described and modeled individually at the component level, except the ASU for which data reported in the literature is used [26].Based on Morgan [30], an energy consumption of 108 kWh/tN 2 is assumed for the ASU, delivering pure N 2 at 8 bar.

Electrolysis technologies
The electrolysis systems examined in this study, all based on AEC and SOEC technologies, are detailed in Table 2 in terms of operating conditions.While these technologies are known, certain extensions proposed in this study surpass the current state-of-the-art for scalable systems.Notably, HTP-AEC, validated at technology readiness level 3, and HP-SOEC, demonstrated at technology readiness level 4-5, exemplify advancements beyond conventional applications [31].

Solid oxide electrolysis (SOEC)
. SOEC performs efficiently compared with other available electrolyzers, especially when there is a heat integration between the electrolyzer and HBL.Part of the required external heat for water splitting can be supplied from waste heat sources available in the HBL to increase the overall system efficiency.
A schematic diagram of a stand-alone SOEC is shown in Fig. 1.The system utilizes two heat exchangers, HEX1 and HEX2, to effectively recover the waste heat from the stack off-gas.However, to achieve the desired stack inlet temperature and provide the necessary external heat during startup, auxiliary electrical heaters, ELH1 and ELH2, are employed.To maintain a hydrogen mole fraction of 10 % in the inlet flow, a portion of the cathode off-gas is recirculated, while the remaining flow passes through HEX1 and enters the condenser, COND, where it undergoes cooling and water separation.The resulting pure hydrogen can then be combined with nitrogen and supplied to the HBL.
The stack is assumed to be operating at the thermoneutral point.The thermo-neutral point is where the current density is raised to the extent that the internal resistive heating from various internal resistances (ohmic, activation and concentration) balances the heat needed for the water splitting reaction.Part of the thermal energy required to produce steam (at the stack operating temperature) can be supplied from the HBL.Maximizing the heat integration between SOEC and HBL minimizes electricity consumption by the electrical heaters and increases the overall power-to-ammonia efficiency.
To consider the possibility of selling the oxygen as a by-product and to protect the electrode through start-up and thermally managing the stack, part of the generated oxygen is recirculated with an outgoing bleed instead of sweeping air [32].However, the recirculation of hot pure oxygen is a safety challenge that has to be addressed (requiring industrial protocols to be developed).It imposes stringent material requirements on the construction materials used in the oxygen loop.It should be noted that this assumption has a minimal impact on the assessed efficiencies and costs, as the power consumption of an alternative air blower would be extremely low compared with other components.
The chemical reactions at the fuel (cathode), at the oxygen (anode) electrodes of the cell, and the overall reaction are as follows: Cathode: Anode: Overall: The cell voltage, E SOEC , can be calculated as follows [33]: where, the reversible potential, E r , can be obtained by the Nernst equation [34].η ohm , η conc , and η act are the ohmic overpotential, concentration overpotential, and activation overpotential, respectively.Table 3 lists the equations utilized to calculate these quantities.More details about SOEC modeling and cell performance validation can be found in the Appendix.Finally, the generated hydrogen, ( ṅH2 ) SOEC , and SOEC power consumption, ẆSOEC , can be written as: J is the current density and ṅH2 the hydrogen production rate per cell cross-sectional area.
where, A cell and N cell are the active cell area and number of the cells in a stack, respectively.

Alkaline electrolysis (AEC).
Currently, AEC is the primarily employed technology for large-scale green hydrogen production, as it is the most developed and mature electrolysis technology [41].However, relatively high internal cell resistance decreases the overall efficiency of the system.For AEC, the optimum operating point can be determined through a balance between CAPEX and OPEX, as shown in Ref. [31].This study investigates two AEC technologies.Currently, highpressure low temperature (~80 • C) AEC is being commercialized and scaled up.Operation at high pressure improves the efficiency, as bubble formation is retarded and the internal resistance in this way can be kept down [42,43].Bubble formation and transportation results in extra losses [42].Not only the gas dissolution but also the blocking of the reaction site with the gas phase between the electrolyte and electrodes leads to increased resistance.The oxygen evolution process, however, still causes high internal resistance.Increasing the temperature to

150-200
• C can significantly lower the overpotential needed for this.This is done in the high temperature, high-pressure alkaline technology (HTP-AEC).
The operating load current density for both AEC technologies is assumed to be around 0.5 A/cm 2 , which is higher than the thermoneutral point for the current technologies.Therefore, both LTP-AEC and HTP-AEC operate under exothermic conditions.Fig. 2 illustrates the flow diagram of AEC.HEX1 and HEX2 are used for cooling to keep the stack inlet temperature fixed at the operating temperature.
Chemical reactions in the cathode and anode sides of the cell and the overall reaction are as follows: Cathode: Anode: Overall: The required cell voltage for this electrochemical reaction can be determined by the cell thermodynamics as follows: here, ΔG is the change in Gibs free energy of the reaction, ΔH the reaction enthalpy and ΔS the reaction entropy.
Anode exchange current density [37] J0,a = γ a Cathode concentration overpotential [38] η conc,c = RT 2F ln Ohmic overpotential [37] η ohm = J T B ohm exp In this research, the polarization curve reported by Green Hydrogen Systems [44] is used to represent the state-of-the-art performance of a LTP-AEC system.For HTP-AEC, in-house data from the EEEHy project [45] have been used.More details about the data utilized for both LTPand HTP-AEC modeling can be found in Ref. [31].
The rate of produced hydrogen and power consumption by the AEC are as follows: Where J is the current density running through the electrolyzer, E AEC the cell potential, A cell the cell area and N cell the number of cells.

Haber-Bosch loop (HBL)
Haber-Bosch loop is the primarily used industrial process of ammonia production, which utilizes a mixture of hydrogen and nitrogen as feedstock to synthesize ammonia at temperatures around 450-600 • C and pressures higher than 100 bar [46].Fig. 3 shows the flow diagram of the HBL at the system level.In this research, the HBL was modeled based on the study performed by Flórez-Orrego et al. [47] for large-scale ammonia production.As the figure shows, the ammonia synthesis plant consists of a three-bed reactor equipped with a heat exchanger to transfer heat between the first bed inlet and outlet flows and two heat recovery steam generators (HRSG) after the second and third beds to produce steam.In addition, the waste heat of the compressor intercoolers is utilized to produce steam or preheat the feed water for electrolysis.The inlet temperature of each bed should be high enough for the reaction to proceed fast but not to the point where the conversion is limited by equilibrium.Flórez-Orrego et al. [47] have optimized the performance of the HBL by tuning the inlet bed temperature of each bed for the specific syngas compositions.To achieve a higher ammonia conversion per pass, they divided the reactor into three sequential catalyst beds with intercooler.Table 4 outlines the assumptions and input data considered in the HBL modeling.
The following reaction, which occurs at higher temperatures (~300-600 • C) in the presence of appropriate catalysts, describes the ammonia synthesis: The reaction rate of ammonia synthesis over a catalyst can be expressed as follows [14]: ) where P is the partial pressure of the compound (bar), R the universal gas constant (J/mol K), ρ cat the catalyst density (kg/m 3 ), f a correction factor (4.75), k 1 and k -1 the pre-exponential factors for the forward and reverse reaction paths, respectively.KMR 111 is assumed as a prereduced iron-based ammonia synthesis catalyst with a density of 2474 kg/m 3 [48].

System efficiency
The LHV-based efficiency of the entire system can be written as: where ṁNH3 is the produced green ammonia mass flow rate (kg/s), LHV NH3 is the lower heating value of ammonia (kJ/kg) and Ẇtotal is the total consumed power by the system (kW), including electrolyzer, air separation unit, electrical heaters, compressors and pumps.

Techno-economic analysis
To conduct the economic analysis, the required initial investment cost; as the capital expenditure (CAPEX) of each individual component is estimated together with maintenance and manufacturing costs.Manufacturing costs are the costs incurred during the production of a product.These costs include direct (labor costs, direct supervisory and clerical labor and laboratory charges) fixed (local taxes and plant overhead) and general (distribution, R&D and administration) manufacturing costs.The most frequently utilized cost functions in the literature are considered for the employed components to estimate the  overall system CAPEX, as detailed below.Different expenses can be found in the literature for electrolyzers.Common is, however, that the electrolysis stack cost is expected to drop significantly over the years.This is due to the expected development of electrolysis technologies and, in particular, the expected cost reductions from mass manufacturing.SOEC with the lowest maturity is expected to experience the highest relative cost reduction [49].It is generally assumed that the HBL plant and ASU are mature technologies and are not expected to undergo significant changes in terms of CAPEX in the foreseeable future.
Here, literature values are used to estimate the cost of AEC and SOEC in 2020, 2030, and 2050.The chosen base case values are listed in Table 5.More details about the CAPEX estimation for AEC and SOEC and their stack replacement cost can be found in Ref. [31].
Also, it is assumed that the entire CAPEX of a HP-SOEC is 20 % higher than a LP-SOEC [31].The reason is that more robust pipes and vessels will be used to withstand the higher pressures.
No cost prediction is available for HTP-AEC by 2020, as this technology is not yet established.However, it is assumed that this technology will be developed by 2030 and follow the same CAPEX trend as the SOEC technology.Both technologies rely on ceramic cells, and coatings and special steels are needed to withstand the corrosive environments.
After estimating the CAPEX of the entire system, the present value (PV) of future monetary transactions, specifically the cost of stack replacement, is calculated as follows [52]: where SRC is the stack replacement cost at the end of the mth year and i is the annual interest rate.In this way, the present value of stack replacement cost can be calculated and considered as part of the initial CAPEX for the electrolyzer.After estimating the required CAPEX for the entire system and the present value of the stack replacement cost, the amortized CAPEX can be expressed in terms of €/year as follows: CRF is the capital recovery factor and is defined as: where, EL is the system economic lifetime.
In addition to estimating the amortized CAPEX, it is necessary to estimate the OPEX of the system in terms of €/year.This should include costs related to electricity consumption, maintenance of the entire system, and manufacturing costs.The annual maintenance cost is assumed to be 2 % of the whole system CAPEX [21].Manufacturing costs are adopted from Reference [53], consisting of direct, fixed and general manufacturing costs.Direct manufacturing cost mainly depends on the number of operators per shift (NOL) which is a function of the number of processing steps involving the handling of particulate solids and the number of nonparticulate processing steps.More details can be found in Reference [53].Details of manufacturing costs and the rest of the inputs utilized in the techno-economic assessment are listed in Table 6.The evaluated CAPEX for electrolyzers encompasses comprehensive turn-key prices, including manufacturing, engineering, piping, procurement, construction, contingencies, financing, and other relevant factors.Consequently, the electrolyzer CAPEX is subtracted from the overall gross initial cost (CAPEX total in Table 6) to estimate the manufacturing expenses.
Once the amortized CAPEX and annual OPEX of the system have been estimated, the cost of green ammonia can be determined using the following equation:

Table 5
Cost functions of the main components.The costs of the electrolysis units are "tun key" estimates at plant level and includes thus balance of plant components and additional costs as outlined in Ref. [31].

System layouts
Figs. 4 and 5 show the green ammonia producing systems designed based on AEC and SOEC, respectively.Current densities of 0.5 A/cm 2 and the one corresponding to the thermoneutral point are assumed for AEC and SOEC, respectively.The operating conditions and the technical parameters of the HBL and ASU are the same for all systems.This baseline enables a consistent comparison between the different routes for hydrogen generation of each process concept.
In Fig. 4, the AEC and HBL are connected such that the produced hydrogen is combined with nitrogen and sent to the HBL.A multi-stage compressor (COMP1) equipped with intercoolers is modeled to pressurize the HBL feedstock.During steady-state operation, the HTP-AEC operates under exothermic conditions and there is no need for heat integration (rather, there is a need for cooling or extracting waste heat to be sold if possible).
Fig. 5 shows the layout of the SOEC-based green ammonia plant.The main difference with the AEC-based plant is that here there is a heat integration between SOEC and the HBL.The following couplings have been made.
• Waste heat from the multi-stage compressor (COMP1) intercoolers is utilized to generate steam for the LP-SOEC system in the preheater (PH).However, the heat from the intercoolers will only be able to preheat the feed water (without vaporizing it) for the HP-SOEC system, as evaporation occurs at 234 • C (saturated steam at 30 bar).• The heat content in the reactor off-gas (REAC2 and REAC3) is also used to evaporate part of the feed water using the heat recovery steam generator 1 and 2 (HRSG1 and HRSG2).• Furthermore, the heat content in the SOEC plant off-gas is utilized to vaporize any remaining water in the feed and to superheat the generated steam if possible.
For the pressurization, it should be noticed that the cathode (or fuel) side of the SOEC in this study is pressurized to obtain a lower open circuit voltage as compared to pressurizing both oxygen and fuel sides.As reported in Ref. [54], there is little difference between single-sided and double-sided pressurization as the single-sided mode hinders the OCV increase and reduces the ASR.This would eventually lead to a more efficient system, as the generated oxygen is not pressurized and hence there is no efficiency penalty via an increased electromotive force (EMF) of the cell.It is assumed that the stacks can tolerate the imposed pressure difference.One could also operate with the oxygen at the same pressure as the H 2 , but this case was not treated here.
The electrical heater 1 (ELH1) vaporizes the remaining feed water and/or superheats the generated steam up to electrolyzer operating temperature.Although the SOEC plant operates at the thermoneutral point, ELH1 and ELH2 are needed to compensate for the slight amount of heat that is lost in waste heat recovery from SOEC off-gas.For start-up conditions, ELH1 supplies all the required energy to produce steam and heat up the system.

Thermodynamic results
An overview of the technical results obtained from the detailed thermodynamic analysis of each system is shown in Table 7.The systems are designed to produce 430 kton green ammonia per year under steadystate conditions.In Table 7, the results are normalized to be per produced ton of ammonia, as no non-linear scaling effects enter the models.
Referring to this table, the system equipped with HTP-AEC performs more efficiently than the system designed based on LTP-AEC.The benefits of HTP-AEC compared with LTP-AEC include enhancing the rates of electrochemical reactions at the electrode surfaces and increasing the ionic conductivity of the electrolyte, which results in low power consumption by the electrolyzer and higher system efficiency.
SOEC is more efficient than AEC, but especially the heat integration between SOEC and HBL leads to efficiency improvement.
Increasing SOEC pressure decreases the heat generated by the compressors and, as a result, the heat from the intercoolers.Thus, ELH1 requires more power (0.21 MWh/tNH 3 more power than the system designed based on LP-SOEC), which is counted in the operation of the electrolyzer unit.Thus, the electrolyzer appears less efficient in Table 7.However, the needed power for the HBL feedstock pressurization is reduced by 0.29 MWh/tNH 3 compared to the LP-SOEC system.Thus, these two have "opposing" effects on the entire system's efficiency.Eventually, the system designed based on HP-SOEC has slightly higher efficiency than that designed based on LP-SOEC.
Results from the modeling reveal that in the systems designed based on LP-SOEC and HP-SOEC, around 57 % and 31 % of the required steam could be produced by heat integration with the HBL; 1.43 and 0.77 ton of steam at SOEC operating pressure per ton of produced ammonia, respectively.

Cost breakdown results
It is worth noting that the results associated with techno-economic Fig. 4. AEC-based green ammonia plant.
H. Nami et al. analysis only estimate green ammonia cost now and in the future under specific assumptions and cannot be easily generalized to different locations in the world with different LCOE and capacity factors.The systems should be assessed for specific sites using real wind and solar profile data to calculate a close-to-real green ammonia cost see e.g.Ref. [10].However, a sensitivity analysis is presented next which elucidates the effects of different LCOEs and capacity factors in the following.
In addition, the main aim of this work is to assess the economic impact of the possible technical developments under identical and conceivable conditions.As hydrogen storage is expensive at small scale, it is assumed in this analysis, that the produced green hydrogen will be directly converted to ammonia; no storage facilities are considered.
It is assumed that the systems operate 50 % of the year (4380 h) when green electricity from wind and sun is available.Therefore, the systems are intentionally oversized by a factor of 2 to meet the desired annual production of green ammonia.
The cost breakdown of green ammonia production based on AEC and SOEC technologies is illustrated in Fig. 6, considering the years 2020, 2030 and 2050, with a capacity factor of 50 % and a LCOE of 30 €/MWh.
In this figure, "other CAPEX" refers to the capital expenditures of the entire system except for the electrolyzer.Since HTP-AEC and HP-SOEC are not commercially available, the cost of ammonia produced based on these technologies is not included in the year 2020.
As can be seen, by 2020, the AEC-based system leads to the most costeffective ammonia production.However, by 2030 and 2050, the cost of ammonia produced using SOEC is expected to experience a considerable decrease due to operating efficiently and the anticipated CAPEX reduction for SOEC, mainly due to technology development and scaling up of production volumes.The cost benefits of incorporating SOEC in green ammonia systems are further discussed later by calculating the LCOEs and needed capacity factors to ensure economic competitiveness.
Comparing the cost of ammonia produced using low-and highpressure SOEC technologies, the latter is the more cost-efficient.The lower internal resistance at high pressure increases the current density at the thermo-neutral operation, resulting in the need for fewer cells and thus lower CAPEX.Besides, employing HP-SOEC reduces electricity expenses for the HBL feedstock compression.These advantages of using HP-SOEC almost overcome the effect of the assumed 20 % higher CAPEX for a pressurized system and reduced heat integration between the electrolyzer and the HBL.By 2050, the expenditures related to plant CAPEX will have a lower share than the OPEX in the cost breakdown.Therefore, the only significant advantage of utilizing HP-SOEC here is the reduced power required for feedstock compression.
In 2030, employing high-temperature alkaline electrolysis becomes less attractive because of the estimated higher CAPEX than lowtemperature alkaline.However, by 2050, employing HTP-AEC could become cost-effective under the projected cost developments compared to LTP-AEC.These estimates rely on the assumption of the rate of upscaling to mass manufacture, which depends on the willingness for investment.It is more accurate to state that, even with a higher CAPEX, the up-scaled HTP-AEC is expected to eventually outperform the LTP-AEC technology.
In 2050, the cheapest projected ammonia cost is 495 €/ton, which could be achieved with the system that utilizes HP-SOEC as the source of the green hydrogen.

Fossil-based ammonia 4.4.1. Emissions
Ammonia, the second most produced chemical, is responsible for approximately 2 % of worldwide fossil fuel consumption and approximately 1.8 % of global CO 2 emissions [55].However, it is hard to determine the specific CO 2 emitted from fossil ammonia plants, since it highly depends on the utilized fuel in the plant (natural gas, coal, oil or naphtha).Besides, some ammonia plants operate more efficiently and/or less polluting than others.The average value of CO 2 emissions of ammonia production is estimated to be 2.4 tCO 2 /tNH 3 [1], while ranging between 1.6 and 2.7 for major regions [56].The modern NG-based sites in the Asia Pacific region are responsible for the lower value of this range; coal-based plants widespread in China can be considered the most CO 2 -intensive production route with almost triple emissions, i.e., 4.4 tCO 2 /tNH 3 [57].It should be highlighted that the impacts of any possible methane leaks associated with the process are not considered in these assessments.

Cost of grey ammonia and threshold CO 2 taxes
Ammonia produced from NG is here referred to as grey ammonia.It is responsible for considerable CO 2 emissions worldwide today.The cost of grey ammonia is mainly a function of fossil feedstock price.The steam methane reforming (SMR) process is the most mature conventional fossil hydrogen production technology.Based on Yadav et al. [58], the industrial SMR process emits 10.6 kg CO 2 per kg of hydrogen.Also, the power production from natural gas resources releases 443 gCO 2 /kWh emissions [59].
To estimate the cost of grey ammonia in this study, the cost of grey hydrogen is first obtained via the following correlation [58]: where C H2, SMR and C CH4 are the cost of fossil hydrogen and natural gas (NG) in terms of $/kg and $/Nm 3 , respectively.After calculating the cost of grey hydrogen, the cost of grey ammonia is estimated in the same way as for green ammonia.
During the last ten years before COVID 19, the NG price has been relatively stable in Europe, ranging from 10 to 30 €/MWh [60].However, when drafting this investigation, the NG price in Europe has experienced a wide fluctuation.In 2020, it dropped below 10 €/MWh.
Then, it increased by a factor of almost 7-8 compared with the

Table 7
Technical results of the systems thermodynamic modeling (LHV of 120 and 18.6 MJ/kg are considered for hydrogen and ammonia, respectively).A/cm 2 for AEC and thermo-neutral value for the SOEC (ca.0.96 A/cm 2 for LP-SOEC and 1.25 A/cm 2 for HP-SOEC)).
H. Nami et al. maximum price before COVID due to Russian-Ukraine conflict in 2022.
Although it is difficult to estimate the NG price for the future, this analysis assumes that it will remain in the pre-COVID situation range by 2030 and 2050.Table 8 represents the estimated cost of fossil hydrogen and ammonia without CO 2 tax, while LCOE is assumed to be 30 €/MWh.The cost of grey ammonia is estimated to be 275-450 €/tNH 3 , changing the natural gas price from 10 to 30 €/MWh.Implementing a CO 2 tax will increase the cost of fossil ammonia.The cost of grey ammonia is reported to be 385 €/tNH 3 by Pozo and Cloete [61] for the conventional plants at European energy prices (60 €/MWh electricity and 6.5 €/GJ ∼ 23.4 €/MWh natural gas).
For green ammonia to reach cost parity with grey ammonia, a CO 2 tax would need to be imposed.To estimate the threshold CO 2 tax, the life cycle greenhouse gas emission of e-ammonia is assumed to be 10 gCO 2 / MJ (0.186 tCO 2 /tNH 3 ).This assumption is based on reported figures of approximately 5 and 15 gCO 2 /MJ (only electricity-based emissions) for wind-and solar-driven e-ammonia plants, respectively, in the IEA report [62].Fig. 7 depicts the estimated threshold CO 2 taxes for different LCOE and natural gas prices for today and estimated green ammonia costs for the future, assuming a capacity factor of 50 % for the green ammonia plant.The obtained thresholds are thus the difference between the grey and green ammonia costs that need to be balanced by a CO 2 tax to incentivize the transition.Fig. 7 shows that since the cost of green ammonia is 2-4 times higher than grey ammonia for 2020, relatively high CO 2 taxes are required to reach parity.The lowest taxes relate to the LTP-AEC case.Assuming a LCOE of 30 €/MWh and a NG price of 30 €/MWh, a CO 2 tax of 115 €/tCO 2 (~275 €/tNH 3 ) is required for cost parity.
By 2030 and 2050, lower CO 2 taxes will lead to cost parity due to a reduction in the cost of green ammonia, mainly because of the lower CAPEX assumed for electrolyzers and technology improvement (hightemperature for AEC technology and high-pressure for SOEC technology).
By 2030 and LCOE values of higher than 20 €/MWh, green ammonia produced via SOEC-based system reaches cost parity with grey ammonia at lower CO 2 taxes compared with the AEC-based system.This is because the cost of green ammonia from SOEC-based systems will be lower than that of AEC-based systems for LCOEs higher than 20 €/MWh by 2030.

Cost of blue ammonia
Blue ammonia is produced the same way as grey ammonia, while 85-95 % of the produced CO 2 is assumed to be captured and stored downstream.Blue ammonia is often promoted as an alternative to green ammonia, as much of the CO 2 from fossil fuel utilization can be captured.Most of the CO 2 is already captured after the SMR to separate it from the hydrogen in the current ammonia plants, it is however vented into the atmosphere.Only the part CO 2 from the NG used for heating the SMR is currently not captured.The potential for significant attention towards blue ammonia hinges on two factors: rapid scalability and the pace of cost reduction in renewable electricity sources.Although not sustainable, this route can be considered a relatively environmentally friendly alternative to grey ammonia.The lack of industry standards for proven permanence of carbon capture and storage (CCS) and mitigation of upstream methane emissions are serious challenges impeding blue ammonia.
The cost of blue ammonia mainly depends on NG price and the cost of CCS.The CO 2 already captured is more than 60 % of the CO 2 from a NG-based ammonia plant, but handling and storage will still add costs.However, the reformers are also powered by fossils (typically NG as well), resulting in CO 2 emissions [63].The CO 2 coming from the reformers of Yara's ammonia plant in Porsgrunn, Norway, is around 415 ktCO 2 out of the overall CO 2 emission of 1 Mt per year at full capacity [64].In this work, it is assumed that 40 % of produced CO 2 stems from the dilute flue gas, which requires adding a CCS unit for capturing.It is possible to supply the heat for the steam methane reforming by electricity rather than by burning natural gas (referred to as "e-SMR"), whereby only half the amount of CO2 needs to be captured.This will reduce the overall cost of the CCS plant but add cost compared to NG-fired reforming.This approach, albeit promising, is not analysed further here, since detailed process costs are not available in literature.
Table 9 outlines the estimated cost of blue ammonia for 2020, 2030 and 2050 without imposing a CO 2 tax.It is assumed that the NG price will remain 10-30 €/MWh until 2050.The cost of blue ammonia is the cost of grey ammonia, outlined in Table 8, plus costs related to CCS.Also, it is supposed that from 2020 to 2050, CCS cost will reduce from 133 to 87 €/tCO 2 [65].As can be seen, currently, CCS cost constitutes 22-32 % of blue ammonia cost depending on NG price.
Assuming a NG price of 30 €/MWh, the following points can be seen from the obtained results.
• By 2020, only AEC-based green ammonia can compete with blue ammonia when considering a "low" value of 15 €/MWh as LCOE.
• By 2030, the cost of green ammonia (based on both AEC and SOEC technologies) will be lower than that of blue ammonia when considering LCOEs below 15 €/MWh.

Sensitivity analysis 4.5.1. Effects of levelized cost of electricity (LCOE)
In 2020, despite the disruptions imposed by COVID, the trend in cost decrease continued for wind and solar power.Compared to 2019, the global weighted-average LCOE from new capacity additions of concentrating solar power, offshore wind, onshore wind and utility-scale PVs declined by 16, 9, 13 and 7 %, respectively [66].
At the time of writing this work, the LCOE for wind farms across the world ranges from 29 (Denmark) to 896 €/MWh (Italy), while for solar plants varies from 34 (France) to 275 €/MWh (Italy) [67].However, for the plants with a capacity of higher than 100 MW, LCOE from wind energy span from 31 (Norway) to 200 €/MWh (Japan) and for solar plants ranges between 35 and 142 €/MWh in the United States of America [67].
The trend shows that these costs must be expected to keep decreasing onwards.Norway and Algeria are among the countries with the lowest LCOE for industries; ~43 and ~19 €/MWh, respectively [68,69].
However, estimating future cost levels of electricity is not trivial as this highly depends on technological and societal developments and adopted policies.This factor could be the primary source of uncertainty in projecting future ammonia prices.Some attempts to predict the expected electricity costs can be found in the literature [70,71].From the analyses presented by Campion et al. [70], it seems that LCOE levels around 20-30 €/MWh could likely be achieved in several places in the world with good wind and sun potential, even when counting in some purchases from the grid to enable around 7000 h of operation per year [70].

Fig. 7.
The threshold CO 2 tax that leads to the cost parity of grey and green ammonia (capacity factor of 50 % for green ammonia plant and current density of 0.5 A/ cm 2 for AEC and thermo-neutral value for the SOEC (ca.0.96 A/cm 2 for LP-SOEC and 1.25 A/cm 2 for HP-SOEC)).

H. Nami et al.
To generalize the obtained results, a sensitivity analysis is conducted to see the effects of changing LCOE on the estimated cost of green ammonia.Changing LCOE will change the share of electricity cost in the total cost of green ammonia, while the other costs remain constant.
Fig. 8 shows the costs related to the consumed power by the entire system based on different electrolysis technologies for LCOE values of 10-60 €/MWh.The two vertical lines show the current LCOE for industries in Algeria and Norway -representing locations with the lowest LCOE for the industry in the world and Europe, respectively.Referring to this figure, the cost associated with the consumed electricity is almost the same for the systems that employ LP-SOEC and HP-SOEC.Both highand low-pressure SOECs operate under thermo-neutral operating conditions.So, although the HP-SOEC uses less power for feed gas compression, additional heat must be added to the HP-SOEC system elsewhere to make the SOEC unit run thermo-neutral.Electric heaters will supply this additional heat.
As seen in Fig. 9(a), the system designed based on AEC will produce cheaper ammonia than the SOEC-based system, by 2020, due to its lower CAPEX.Also, it should be noted that maintenance costs in these estimates highly depend on the estimated CAPEX for the entire system.
By 2030, the CAPEX of the SOEC-based system is expected to be considerably lower, which here results in the cheapest green hydrogen route for ammonia synthesis.Although similar cost is assumed for the HTP-AEC and the LP-SOEC, this is still not economically competitive compared with LTP-AEC for LCOE below 50 €/MWh, as the efficiency is still lower than the SOEC.So, although HTP-AEC has better efficiency than LTP-AEC, the CAPEX is still too high.
By 2050, the CAPEX of all the electrolysis technologies is estimated to have dropped significantly, so OPEX will have a higher impact on the cost of the produced ammonia.Therefore, the HTP-AEC will be more competitive than the LTP-AEC, while the SOEC-based plants are superior to both.
Interestingly, the HP-SOEC only yields slightly cheaper ammonia than the LP-SOEC with the assumed 20 % higher CAPEX.Referring to Fig. 8, the cost related to the consumed power by the entire system is approximately the same for both systems operating with LP-and HP-SOEC.However, the electrolyzer CAPEX is lower for HP-SOEC (per  produced amount of hydrogen).Consequently, the system equipped with HP-SOEC has slightly better economic performance.Fig. 9(c) also compares the estimated cost of grey and blue ammonia, assuming modest price development of these fuels.By 2050, assuming a NG price of 30 €/MWh, the cost of green ammonia will be lower than that of blue ammonia if a LCOE of 35-40 €/MWh can be achieved.Also, assuming a NG price of 30 €/MWh, green ammonia will be costcompetitive to grey ammonia for LCOE lower than 20-25 €/MWh.

Effects of capacity factor
All results presented and discussed up to this point assume a capacity factor of 50 %; thus, the green electricity is available at the LCOE for 50 % of the year.Under the base case conditions, all the systems were scaled to produce 430 ktNH 3 per year with a capacity factor of 50 %.It is assumed that operating with different capacity factors does not affect the stack replacement time and system economic life.Fig. 10 presents the effects of the availability of low-price renewable electricity on the cost of green ammonia.Here, the capacity factor is the ratio of system equivalent full-load operating hours to the maximum possible operating hours in a year (8760).For instance, a capacity factor of 50 % means that the system should be scaled up by a factor of 2 to produce 430 ktNH 3 per year.
Operating with lower capacity factors resulting from a lower amount of available renewable electricity will make the green ammonia more costly due to the depreciation of the capital investmentthe plant has now to be oversized to reach the same amount of product annually.The CAPEX refers to a significant share of the green ammonia cost.Thus, as can be seen in Fig. 10(a), the cost of green ammonia is extremely high for lower capacity factors, especially for SOEC-based system.The system OPEX per produced amount of ammonia remains constant, and the CAPEX per produced amount of ammonia increases.
Referring to Fig. 10 (b), by 2030, cost of green ammonia produced from the system based on HTP-AEC is the same as that produced from the system based on LTP-AEC.Although HTP-AEC performs efficiently compared with LTP-AEC, CAPEX depreciation significantly affects the cost of green ammonia.For SOEC-based systems, efficiency improvement due to heat integration between electrolyzer and HBL compensates for CAPEX depreciation.
By 2050, the share of electrolyzer CAPEX will be low and system efficiency (OPEX) has a higher impact on the cost of green ammonia.Therefore, employing SOEC to produce green ammonia is the best choice, followed by HTP-AEC.
The method here applied assuming a certain LCOE and capacity factors of higher than 50 % will underestimate the ammonia costs, as green electricity is only available at these capacity factors in few but highly relevant places in the world.If based on solar or wind sources, the capacity factors would range from 25 to 50 % and hence running the plant more than 50 % of the time would require that the electricity from the grid is also purchased at times of no wind and sun.This electricity would generally not be green.
Accounting for different capacity factors provides the results in the plots in Fig. 10.However, this overestimates the economical penalty of reducing the annual number of operation hours, as possibilities of storing H 2 or electricity from times of low electricity costs to periods with little production are neglected.

Effects of the estimated CAPEX
Since SOEC technology is currently in the initial commercialization phase, its future performance, scalability and consequent economic assessment are uncertain.Besides, the electrolyzer CAPEX is estimated to be decreased significantly in the long term.Consequently, a ±50 % uncertainty is considered for the estimated CAPEX of the electrolyzers in the long term, as shown in Fig. 11.
Referring to this figure, by 2050, for a capacity factor of 50 %, the costs associated with the consumed electricity will be so high that a ±50 % change in the electrolyzer CAPEX does not considerably affect the ammonia cost and this is even more so at higher capacity factors.This is rather interesting, as it highlights that CAPEX will be a parameter for competition, but the ability to heat integration (to reduce the OPEX) Fig. 10.Cost of green ammonia produced with different electrolysis technology vs. capacity factor (LCOE = 30 €/MWh and current density of 0.5 A/cm2 for AEC and thermo-neutral value for the SOEC (ca.0.6 A/cm2 for LP-SOEC and 0.86 A/cm2 for HP-SOEC)).
H. Nami et al. even more so.

Conclusion
In this work, various green ammonia producing systems based on electrolytic hydrogen from AEC or SOEC have been modeled under steady-state operating conditions.To investigate the effect of different potential technological improvements for both AEC and SOEC, pressurized AEC is modeled in both low-temperature (80 • C) and hightemperature (200 • C) operating modes and SOEC is also modeled considering both low-pressure (atmospheric) and high-pressure (30 bar) modes.
Both the systems are modeled at the components level, and it was found that.
The possibility of supplying cheaper renewable electricity is studied, and the cost of green ammonia is estimated considering the current and currently projected electrolyzer CAPEX.It was found that.
• Although SOEC performs efficiently compared with AEC, employing this technology to produce green ammonia is not economically favorable before the cost of SOEC is decreased to 1500 €/kW (assuming a capacity factor of 50 %, LCOE of 30 €/MWh and AEC CAPEX of 830 €/kW) through, e.g., upscaled manufacturing.
• By 2030 and 2050, upscaled manufacturing and technology improvements are expected to decrease the cost of the SOEC.Therefore, employing SOEC in a green ammonia producing system appears to be economically favorable to the AEC in this analysis.
• By 2050, assuming a LCOE of 30 €/MWh and a capacity factor of 50 %, a green ammonia cost of 495 €/ton is projected, produced with a pressurized SOEC-based system.The cost of green ammonia is compared with grey and blue ammonia.The threshold CO 2 tax that would bring green ammonia on par with grey ammonia is estimated for different years, assuming NG prices of 10-30 €/MWh and a wide range of LCOE values.The main findings of this are.
• With a LCOE of 30 €/MWh, a capacity factor of 50 % and a NG price of 30 €/MWh, a CO 2 tax of 105 €/tCO 2 (~250 €/tNH3) is required for green ammonia to be cost comparable with grey ammonia.
• By 2050, with 30 €/MWh as the projected NG price, implementing a CO 2 tax of 100 €/tCO 2 (~240 €/tNH 3 ) would make SOEC-based green ammonia cost-competitive even for a LCOE value as high as 60 €/MWh.• By 2050, assuming a projected NG price of 30 €/MWh and a capacity factor of 50 % for green ammonia plant, the cost of green ammonia produced with the SOEC-based system will be competitive with grey and blue ammonia for LCOE values lower than 25 and 40 €/MWh, respectively.
A sensitivity analysis is carried out to assess the effects of LCOE, capacity factor and estimated CAPEX on the cost of ammonia.It was found that.
• By 2050, increasing the capacity factor from 30 to 90 % decreases the cost of green ammonia by almost 80 %.
Ammonia serves as a hydrogen carrier and is likely to be traded internationally to leverage global collaboration for cost minimization.The practical implications of this study for industry and policymakers involve utilizing emerging electrolysis technologies for green ammonia production, projecting future ammonia prices, and determining the necessary CO2 tax for fossil ammonia to achieve cost parity between green and fossil ammonia.These implications address efficiency improvement, cost considerations, and climate impact, respectively.This research is conducted with the understanding that a more detailed economic analysis should be undertaken at specific sites with distinct weather profiles, as these factors may influence the economic conclusions presented herein.Therefore, these economic assessments should be regarded only as guidelines.The potential error in this investigation might arise from the projected 2030 and 2050 CAPEX values for emerging electrolysis technologies.However, it is noteworthy that since the cost of consumed electricity largely determines the green ammonia price, this factor may not significantly alter the economic conclusions presented in this research.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
here, f ob is the objective function, n p and n j are numbers of the operating pressures and load current densities, respectively, and abbreviations exp and sim denote the experimental and simulation data, respectively.Operating pressures of 0.4, 3, and 10 bar are considered [72], and 28 load current densities between − 0.5 and 0.85 A/cm 2 with a fixed step size of 0.05 A/cm 2 are used.It should be noted that the experimental data are interpolated over these current densities so that the operating points are the same as those used in the simulation for calculating the error between the experimental and simulation data, Eq. (A.1).
Table A.2 lists the optimum values of the fitting parameters evaluated by the genetic algorithm for the old and new cells produced and used for the experiments in DTU-Energy.The label "old cell" is used for the cells used over a decade ago, and their experimental data are used due to their available high-pressure operation data [72].Nonetheless, the fitting parameters are adjusted for updated cells with improved performance, especially of the air electrodes, which have been used by Sun et al. [73].Unfortunately, the high-pressure operation experiments have not been performed on the new cells; thus, the fitting is done just for their atmospheric operations.All the operating conditions, except the operating pressures, are the same for the experimental data from both old and new cells used for the fitting here.A.2.A good match is seen between the polarization curves for all operating pressures.One can see better performance if the material parameter identified for the newer cells are used to simulate the high-pressure operation.The latter is used in the onwards simulations.The higher open circuit voltage (OCV) and lower ASR for higher operating pressures and higher pressure effects for lower operating pressures are consistent with the trends reported in the literature, e.g.Refs.[74,75].H. Nami et al.

Fig. 6 .
Fig. 6.Cost breakdown of green ammonia production via different electrolysis technology (LCOE = 30 €/MWh, capacity factor of 50 % and a current density of 0.5

Fig. 9 .
Fig. 9.Cost of green ammonia produced with different electrolysis technology vs. LCOE (capacity factor of 50 % for green ammonia plant and current density of 0.5 A/cm2 for AEC and thermo-neutral value for the SOEC (ca.0.96 A/cm2 for LP-SOEC and 1.25 A/cm2 for HP-SOEC)).

Fig. 11 (
Fig. 11(b) compares the cost of green ammonia produced based on HP-SOEC and HTP-AEC as emerging technologies for green hydrogen.

Fig. 11 .
Fig. 11.Effect of a change in LCOE and estimated electrolyzers CAPEX on the cost of green ammonia by 2050 (capacity factor of 50 % for green ammonia plant and current density of 0.5 A/cm2 for AEC and thermo-neutral value for the SOEC (ca.0.6 A/cm2 for LP-SOEC and 0.86 A/cm2 for HP-SOEC)).

Parameter Value -old cell Value -new cell
Constant of the prefactor used for the anode exchange current density (γ 0,a ) 1.2236 × 10 8 A m − 2 K − 1 2.4341 × 10 8 A m − 2 K − 1 Constant of the prefactor used for the cathode exchange current density (γ 0,c ) 5.6601 × 10 6 A m − 2 K − 1 8.9977 × 10 6 A m − 2 K − 1 Activation energy used for the anode exchange current density (Eact,a) 1.4916 × 10 5 J mol − 1 1.4814 × 10 5 J mol − 1 Activation energy used for the cathode exchange current density (Eact,c) 1.2491 × 10 5 J mol − 1 1.2055 × 10 5 J mol − 1 Charge transfer coefficient of the anode reaction (αa) 0.6573 0.8303 Charge transfer coefficient of the cathode reaction (αc) 0.5000 0.7725 Power of the oxygen partial pressure used for the anode exchange current density (m) 0. 1853 0.1825 Power of the hydrogen partial pressure used for the cathode exchange current density (a) − 0.2154 − 0.0583 Power of the steam partial pressure used for the cathode exchange current density (b) 0. 3143 0.4881 Correction factor for the diffusion coefficients of the species in the anode porous media (k a D,eff ) 0.0995 0.1996 Correction factor for the diffusion coefficients of the species in the cathode porous media (k c D,eff ) 0.0201 0.1695 Material-specific constant used for the ohmic overpotential (B ohm ) 8.2598 × 1012 S K m − 2 7.8247 × 1011 S K m − 2 Activation energy used for the ohmic overpotential (E act,ohm ) 1.2055 × 105 J mol − 1 8.0022 × 104 J mol − 1

Fig. A. 1
Fig. A.1 compares the polarization curves from the experimental data and the developed model for the optimum values of the fitting parameters given in TableA.2.A good match is seen between the polarization curves for all operating pressures.One can see better performance if the material parameter identified for the newer cells are used to simulate the high-pressure operation.The latter is used in the onwards simulations.The higher open circuit voltage (OCV) and lower ASR for higher operating pressures and higher pressure effects for lower operating pressures are consistent with the trends reported in the literature, e.g.Refs.[74,75].

Fig. A. 1 .
Fig. A.1.Comparisons of the polarization curves from the experimental data and the model for different operating pressures.The abbreviations exp and sim denote the experimental and simulation data, respectively.

Table 1
Literature review of green ammonia production.
c photoelectrochemical water splitting.dThisis solar energy to ammonia efficiency.H.Nami et al.

Table 2
Different electrolysis systems modeled in this study.
• C Hydrogen is produced at 30 bar HTP-AEC Stack is operating at 200 • C Hydrogen is produced at 30 bar SOEC LP-SOEC Hydrogen is produced at 1 bar Stack is operating at 700-800 • C HP-SOEC Hydrogen is produced at 30 bar Stack is operating at 700-800 • C H. Nami et al.

Table 4
[47]t data and assumptions considered to model the HBL[47].

Table 6
Input data for the economic analysis.
H.Nami et al.
0.70 and 0.91 MWh/tNH 3 are the consumed electricity by ELH1 in the systems designed based on LP-SOEC and HP-SOEC, respectively.6.07 MWh/tNH 3 is the consumed electricity by the stacks and is the same for both cases.** 0.29 MWh/tNH 3 is the extra consumed power for the extra pressurization of the hydrogen after the LP-SOEC.Consumed power for feedstock pressurization for the HBL is 0.29 MWh/tNH 3 higher for the system designed based on LP-SOEC. *

Table 8
Cost of grey hydrogen and ammonia produced from natural gas as a function of natural gas price (LCOE = 30 €/MWh).