The production costs of hydrogen, e-methane and e-methanol were calculated with eqs. (1) and (2) based on Thunman et al. [37].

The operating period and interest rate are considered within the capital recovery factor, eq. (3). Electrolyzer and methanation or methanol synthesis reactor investment costs are considered in the overall production costs of the respective product:

$c_{{\text{H}}_2}=\frac{CRF\times I{C}_{{\text{H}}_2}+C_{\text{om}}}{FLH}+\frac{c_{\text{ele}}}{\eta }+ c_{{\text{H}}_2\text{O}}$ (1)

$c_{\text{e}-\text{meth}}=\frac{CRF\times I{C}_{\text{e}-\text{meth}}+C_{\text{om}}}{FLH}+\frac{c_{{\text{H}}_2}}{\eta }+ c_{{\text{CO}}_2 }$ (2)

$CRF=\frac{{\left(1+r\right)}^{n}r}{{\left(1+r\right)}^n-1}$ (3)

c_(H2) = production costs of hydrogen [EUR/MWh], c_e−meth = production costs of e-methane or e-methanol [EUR/MWh], CRF = capital recovery factor, IC_(H2) = electrolyzer investment costs [EUR/MW], FLH = full-load hours, c_(H2O) = costs for water [EUR/MWh],IC_e−meth = investment costs for a methanation or methanol synthesis reactor [EUR/kW],C_om = fixed operating and maintenance costs [EUR/kW], c_ele = electricity costs, η = energy efficiency of the process, c_var = variable costs [EUR/MWh], c_CO2 = costs for CO₂ [EUR/MWh], n = plant lifetime, r = depreciation rate.

Table 2 shows the parameters for the economic analysis. The investment costs of the electrolyzers include all system components and installation. These costs decrease in regards to the scale [27]. The investments used in this study for 2 MW_el alkaline and PEM electrolyzers are 1400 and 1800 EUR/kW_el and for 20 MW_el 900 and 1400 EUR/kW_el respectively [38]. Fees for the connection to the grid were added to these costs. The stacks need to be replaced after ten years, accounting for 50% and 60% of the system costs of an alkaline and PEM electrolyzer. Carbon capture costs as EUR/t CO₂ were included in the analysis as variable costs [39]. The assumption was that the e-methane and e-methanol producers would cover the costs of implementing carbon capture, for example, at a pulp and paper mill.

The investment costs of the methanation and methanol reactor represent the total costs, including plant installation [25]. The system costs were adapted to 2022 price levels using the chemical engineering plant index. It is important to notice that the investment costs for electrolyzers are stated for the electricity input in kW_el, whereas for the methanation and methanol synthesis in product output kW_out.

Technological learning can lead to lower costs per kW installed. The learning rate gives the percentual cost reduction per doubling of unit output. The range of learning rates in literature varies tremendously. Schoots et al. [51] analyzed the historical learning rates of electrolyzer equipment and defined a value of 18 ± 13%. Detz et al. [38] used a low rate of 12 % and a high learning rate of 20% for their analysis. Reksten et al. [52] estimated high learning rates of 25-30% for alkaline and PEM electrolyzers. The main argument for these high rates is that the scaling up has not been considered in previous publications. However, the highest learning rate for a technology in history has been seen for PV panels with 22% [53].

The applied learning rates are derived from Böhm et al., who assessed different learning rates of the main electrolyzer components. The difference in this work is that a learning rate of 18% was only used for the electrolyzer stacks. Other parts such as power electronics were assumed to be the conventional share and account for 50% and 40% of alkaline and PEM electrolyzer investment costs. Technological learning was not applied for the conventional components [53]. Stacks also need to be produced more often than the other components because the current stack lifetime is approximately ten years [27].

For the methanation section, a learning rate of 10% was found in the literature [50]. The same value was used for the methanol synthesis. The equipment typically accounts for approximately 20% of the system cost and technological learning is rather derived from the plant engineering and project-specific costs [37].

The total investment costs, eq.(4), consist of the conventional and new components. The cost reductions are calculated with the eqs. (5) and (6). The learning rates are only applied to the share of new components:

$I{C}_{\text{total}}=I{C}_{\text{new}}+I{C}_{\text{conv}}$ (4)

$I{C}_{\text{new}}\left(t_1\right)=I{C}_{\text{new}}\left(t_0\right)\times {\left(\frac{Y_{t_1}}{Y_{t_0}}\right)}^{-b}$ (5)

$LR=1-2^{-b}$ (6)

IC_new = investment costs of new components, IC_conv = investment costs of conventional components, IC(t)= investment costs of a unit at time t, Y(t)= installed capacity at time t, LR = learning rate, b= parameter for the extent of learning measured.

Table 3 shows the installed capacities for electrolyzers, methanation and methanol production. At the end of 2022, approximately 690 MW electrolyzer capacities were installed globally. This number shall increase to 2200 MW at the end of 2023 [54]. For the 1.5 °C climate target an increase to 550 GW in 2030 is proposed by the IEA. However, there can be some restrictions for this tremendous growth. For example, the current iridium mining makes only 3-5 GW of PEM electrolyzers installations per year possible [27].

Therefore, two different scenarios were investigated for capacity installations of hydrogen. One is the business as usual (BAU) scenario with historic growth rates of 2015-2022 [55], extrapolated for 2030 and 2050. The growth scenario is based on the Global Hydrogen Review 2023, where 175 GW in 2030 are feasible to reach and 3670 GW is the target for 2050. Currently, 2/3 of the total installed electrolyzers are alkaline, PEM accounting for approximately 1/3 and high-temperature solid oxide electrolyzers (SOE) for less than 1%. This ratio is supposed to change in the investigated scenarios to 40:40:20 for alkaline, PEM and SOE in 2050 [27].

Methanation and methanol capacities are already established more widely than water electrolysis but based on fossil fuels as resources. For methanation and methanol production 25 GW and 71 GW were installed globally in 2020 and this number is expected to grow to 450 GW [56] and 345 GW (500 Mt) in 2050 [31].

The growth and BAU scenarios results are shown for the investment cost developments derived from technological learning. In the following sections, only the growth scenario will serve as the basis for the e-methane and e-methanol production costs. The BAU case will not provide enough hydrogen capacities to fulfill the green e-methane and e-methanol production targets in 2050.

From the literature analysis, the individual variables are expected to show different effects on the overall production costs of e-methane, e-methanol and hydrogen. Therefore, the following sensitivity analyses were conducted and the variables were modified in a range of ± 50%:

• Reductions in electrolyzer, methanation and methanol synthesis reactor investment costs depended on the learning rates. The base cases for learning rates are 18% for electrolyzers and 10% for methanation/ methanol synthesis.

• Hydrogen production costs in relation to the full-load hours or electricity price

• Investigation of different variables on e-methane production costs: The initial value of operating hours was changed to 5000 hours, the capture costs to 60 EUR/t CO₂ and a scale of 5 MW was chosen for the sensitivity analysis to depict a broad range of the variables. Other input data was not modified and used as stated in Tables 1 and 2.

The effect of carbon taxes on current market prices of fossil natural gas and methanol was also investigated as part of the sensitivity analysis. Emission factors of 297 g CO₂/kg natural gas [57] and 197 g CO₂/kg fossil methanol [58] cover the direct and indirect emissions resulting from sourcing, handling and the consumption. The externalities of fossil fuel use were not reflected adequately in previous times. Therefore, the European Union adopted an emission trading system to set a price on carbon emissions of businesses. Carbon prices are very effective measures to restrict the use of fossil fuels and support alternatives [59].

The SWOT analysis can describe the status quo and strategies for the future can be derived. This analysis aims to compare the technical aspects found in literature and results from the economic assessment for the three different fuels since hydrogen can also be directly used as fuel [60].

Table 4 and Figure 3 displays the results of the investment and overall system costs for the expansion scenarios. The investment costs in 2050 for the electrolyzers depend on the scenario and decrease for alkaline electrolyzers by 31-36% and for PEM electrolyzers by 41-46%. However, the difference between the growth and the BAU scenario is smaller than expected. Although the installed capacity is almost eight times higher, the investment costs differ only by 7-9%. This is mainly due to the conventional part becoming the dominant share of costs. However, the capacities of electrolyzers in the BAU scenario are insufficient for the expected capacities of e-methane and e-methanol in 2050. A substantial increase in capacity additions is needed to reach the defossilization targets. The cost reductions for the total system in the growth scenario are 32-38% for e-methane and 24-30% for e-methanol. The values of the growth scenario are the input used for the calculation of the overall e-methane and e-methanol production costs in the following sections. The initial system costs for e-methane and e-methanol production are almost similar on the 1 MW scale, but as the capacity additions of methanation reactors increase stronger compared to methanol synthesis, so will the costs be approximately 20% lower in 2050.

Figure 3.

Total system cost and investment cost reductions for e-methane, e-methanol and electrolyzers up to 2050

The sensitivity analysis of the learning rates is shown in this chapter to have an overview on the results regarding the investment costs. Figure 4 and Figure 5 show the respective investment costs in relation to high or low learning rates of electrolyzers, methanation and methanol synthesis.

Figure 4.

Investment cost reductions of alkaline (AEL) and proton exchange (PEM) electrolyzers up to 2050 for 9% (low) or 27% (high) learning rates

Figure 5.

Investment cost reductions of methanation (CH₄) and methanol synthesis (CH3OH) up to 2050 for 5% (low) or 15% (high) learning rates

The production costs for hydrogen significantly depend on the full-load hours among other variables. Figure 6 shows current hydrogen production costs in the growth scenario for 20 MW_el electrolyzers in relation to the full-load hours, which are technology specific.

Figure 6.

Hydrogen production costs in relation to the full load hours at 90 EUR/MWh electricity price

In the case of 1000 FLH, which is typical for PV in Central Europe, the costs are 102%-150% higher compared to 8000 FLH. At typical values of 2600-2900 FLH for onshore wind, the production cost increases by 26-44%. However, this evaluation did not consider the trade-off of faster degradation of the electrolyzer stacks with more FLH.

The electricity price was modified in Figure 7 to visualize the impact of this essential parameter on the hydrogen production costs.

Figure 7.

Hydrogen production costs in relation to the electricity price at 8000 FLH

The effect of the scale on the production costs can be seen in Figure 8 and is lower than expected in this work although the total system costs for e-methane and e-methanol are 3135 EUR/kW and 3890 EUR/kW for a 10 MW plant compared to 5590 EUR/kW and 5565 EUR/kW for a 1 MW plant. The used system costs were conservative assumptions based on recent publications which showed higher values than what was estimated in previous papers.

The total production costs will be reduced by 5-10% in 2023 from a 1 MW plant to the 10 MW scale. In 2050 the difference between the sizes accounts for 5-7%. The reason is that other variables affect the production costs more than the scale, which will be shown in the sensitivity analysis results.

The current production costs decrease by 17-22% in 2050. The production costs in 2050 are still too high to be cost competitive under these assumptions. Production costs can be reduced with accounting for revenues for off-heat and oxygen. Oxygen sales can be a valuable revenue stream. However, it is questionable if all the oxygen from electrolyzers can be sold on the market.

Figure 8.

Current production costs of e-methane and e-methanol compared to 2050

Figure 9 displays the overall production costs of e-methane, e-methanol and hydrogen. The lowest values are seen for hydrogen among the three fuels. Currently the grid-connected electrolyzer leads to the lowest production costs. However, wind-based hydrogen production will already show better economic performance in 2030, assuming that the grid electricity price is 90 EUR/MWh, as mentioned in the methodology. The wind-based hydrogen production costs are currently 6 EUR/kg H₂ and will fall to 4 EUR/kg H₂ in the growth scenario.

Figure 9.

Development of e-methane, e-methanol and hydrogen production costs using either electricity generated by wind turbines or from the grid

The costs for wind-based e-methane and e-methanol are on the contrary 15-47% (2030) and 6-45% (2050) higher than for the grid-connected electrolyzer.

Production costs for PV-based production costs were also calculated, however, not shown here, because they are the highest still at the scale of 10 MW with 470-510 EUR/MWh for e-methane and 570-610 EUR/MWh for e-methanol in 2050. PV plants as the only electricity source in the process chain are unfavorable in Central Europe mainly due to the low FLH, thereby increasing the capital costs significantly.

An alkaline electrolyzer can also be coupled with a PV plant as in the work of Ince et al. [29]. However, the efficiency of the electrolyzer was less than 60%, which is not favorable from an economic point of view.