6.4.5. Influence of CO2 Costs

Figure 20 shows the cost of the product as a function of that of the required CO2. According to Brynolf and Taljegard [37], the expected long-term costs for CO2 that are separated from the exhaust gases of a cement plant are between €30 and €50 per ton. Accordingly, for the sensitivity analysis, production costs in this range were assumed. For the sake of completeness, the cost of the product was also calculated to be between €70 and €90 per ton of CO2. The cost of the product was around €1.7/lDE and €1.9/lDE. It is clear that the CO2 price has an influence on the production costs, but that it is significantly lower compared to the other cost factors examined. This means that a favorable CO2 price in the range considered is advantageous for fuel synthesis, but not absolutely essential.

**Figure 20.** Product production costs as a function of the CO2 price.

6.4.6. Influence of the Cost Factors in Comparison

A tornado diagram, which can be seen in Figure 21, was used to compare the influence of the respective cost factors on the levelized cost of electricity (LCOE). This diagram compares the areas in which the levelized product costs are dependent on the respective cost factors, based on the base case of €1.85/lDE. For this comparison, the best case and the worst case assumptions from the previous sections were used and highlighted together with the base case assumptions in Figure 21. In Figure 21, it is once again clear that the electricity price had the greatest influence on the fuel production costs and could influence fuel production costs up or down by up to 40 ct/lDE for the price range examined. It is important to note that the tornado diagram only considered the cost parameters individually. In reality, it can be assumed that there will be a mixture of "best", "base", and "worst" cases presented. However, due to the sum of all possible savings when all of the best cases arrive, a lower limit for the fuel production costs was given at €0.94/lDE. In the same manner, with the arrival of all worst cases with €2.95/lDE, there was an upper limit for the cost of the product.

**Figure 21.** Tornado diagram of the sensitivity analysis of product generation costs.

#### *6.5. Comparison with Alternative Power-to-Liquid Processes*

In the following section, the calculated production costs are compared with those of alternative power-to-fuel processes. As noted in Section 6.4, the methanol and DME synthesis investigated by Schemme [54] and the Fischer–Tropsch synthesis in combination with a RWGS reactor were used for the comparison. The production costs of the three processes are compared in Figure 22 with the manufacturing costs for the worst, base, and best cases specified in Section 6.4.

**Figure 22.** Comparison of the production costs of the developed fuel synthesis with those of alternative power-to-liquid processes (costs of methanol, DME, and FT with the RWGS process according to Schemme [54]).

It must be taken into account that with respect to Schemme [54], different framework conditions were selected in some cases than in the economic analysis carried out in this study. In the three comparison processes, for example, the required hydrogen was not produced in the process, but was obtained externally and taken into account in the cost calculation at a price of €4.6/kg of H2. Therefore, the following comparison serves as a qualitative classification of the calculated manufacturing costs rather than an exact quantitative comparison of the production costs of the different power-to-liquid processes. In addition, the processes investigated by Schemme [54] were smaller than the fuel synthesis investigated herein.

For example, the Fischer–Tropsch reactor investigated by Schemme [54] was fed with a feed stream of approximately 245,000 Nm3/h, whereas in the economic analysis carried out in this study, approximately 350,000 Nm3/h flowed into the Fischer–Tropsch reactor. It is therefore possible that there is a potential for savings through scaling effects for the processes examined by Schemme [54]. However, Figure 22 clearly shows the great potential of the fuel synthesis that has been developed, also taking into account these potential savings. The production costs of €1.85/lDE are already competitive for the base case with the fuel production costs of the methanol and DME synthesis of €1.87/lDE and €1.82/lDE, respectively. In the best case, the costs can even be undercut with €0.94/lDE. An additional advantage of the developed fuel synthesis compared to methanol and DME synthesis is that the infrastructure for traditional fuels already exists and further costs can therefore be saved.

#### **7. Conclusions**

Power-to-fuel technology represents a promising possibility for making the transport sector CO2-neutral in the future. An especially interesting power-to-fuel concept is the coupling of high-temperature co-electrolysis with Fischer–Tropsch synthesis, as this carries some thermodynamic advantages. The aim of this study was to develop such a power-tofuel process, model the developed process in a process simulation program, and then carry out a techno-economic analysis of the overall process. In the developed fuel synthesis, the entire process chain was considered, starting with water and CO2 and ending with the fuel according to specifications. First, water and CO2 were converted into synthesis gas, consisting of hydrogen and carbon monoxide, by means of high-temperature co-electrolysis. In the next step, the synthesis gas was converted into hydrocarbons through a Fischer– Tropsch synthesis, and then processed into synthetic diesel according to EN 15940 and synthetic kerosene type FT-SPK according to ASTM 7566. The fuel preparation consisted of a hydrocracker, reformer, and carrier steam distillation. An additional high-temperature water electrolysis system was used to provide the hydrogen required for the hydrocracker. The process simulation was implemented in the simulation program Aspen Plus, whereby the model was designed for the calculation of any mass flows and so any system sizes. In addition, an energy integration analysis was conducted. The results of the process simulation provide information regarding the material and energetic balance of the process. In the developed fuel synthesis, 1 L of diesel equivalent (35.9 MJ) of synthetic fuels was produced, which was then broken down energetically into 38.9% kerosene and 61.1% diesel. An examination of the fuels produced indicated that both synthetic diesel and synthetic kerosene meet the requirements of the above standards. To produce one liter of diesel equivalent, 2.54 kg of CO2, 3.99 kg of water, and 0.34 kg of oxygen are required. The energetic analysis of the process shows that the energy requirement of the high-temperature co-electrolysis was reduced by the energy integration from about 75 MJ/lDE over 20% to about 59 MJ/lDE. This makes it clear that the coupling of Fischer–Tropsch synthesis with high-temperature electrolysis represents an attractive power-to-fuel concept. In addition, it was found that the energy requirement of the process and so the power-to-liquid efficiency depends heavily on the efficiency of the electrolysis. The power-to-liquid efficiency for an electrolysis efficiency of 70% was approximately 46%, and with an electrolysis efficiency of 100%, the PtL efficiency was almost 67%. The assumed base case electrolysis efficiency of 80% resulted in a PtL efficiency of 52%, whereby the electrical energy for the co-electrolysis, with about 59 MJ/lDE, made up more than 85% of the total energy requirement of about 69 MJ/lDE. Accordingly, the co-electrolysis represents the critical element of the developed fuel synthesis and presents itself as a topic for further research in order to develop a better understanding of the technology as well as to identify possible energy-saving potentials. The energetic analysis also showed that the developed power-to-fuel process generated an excess heat of around 1.005 MWh per ton of CO2 consumed. This heat can be used for CO2 capture technologies. This study showed that the excess CO2 could cover around 67% of the thermal energy required to separate a corresponding amount of CO2 from the ambient air and around 97% of the thermal energy requirement for separating CO2 from industrial waste gases (cement works). The thermal energy requirement of CO2 separation from biogas can be fully covered. This option is very attractive since it offers a biogenic CO2 source, resulting in a completely sustainable route. In the long-term, the further technical development of the direct separation from ambient air will surely enable a broad application possibility for the developed technology. The thermal coupling of the power-to-fuel process with CO2 capture technologies therefore represents a good opportunity to improve the efficiency of the entire process chain, from CO2 to synthetic fuel. If fuel synthesis is coupled with CO2 separation from biogases, the overall efficiency can be increased from 48.1% to 52.1% (i.e., by four percentage points). In the case of CO2 separation from the ambient air, the thermal coupling can increase the efficiency by 4.8 percentage points, from 41.3% to 46.1%. The largest increase in efficiency was found when the fuel synthesis was coupled with the separation of CO2 from industrial exhaust

gases. These values show the importance of the CO2 capture technology and its relevance for the overall process. In order to maximize the efficiency of the future demonstration projects, it is therefore important to select the optimal site concerning the CO2 potential. The process model developed in this study can be used to analyze the economic viability of the examined locations given that, as already described, it is suitable for calculating any system parameters. The further development of the outlined process model can also be the subject of future research. For example, the models developed for calculating the co-electrolysis or Fischer–Tropsch reactor could be enhanced by kinetic models.

In order to realize the potential of the developed fuel synthesis, two important considerations are necessary. On one hand, it is important to invest in the research and development of SOEC technology and to increase the low TRL of the SOEC and bring high-temperature electrolysis to a megawatt scale and market maturity. On the other hand, the choice of location plays a decisive role. In this way, the great potential of the developed fuel synthesis can, above all, be realized in locations where cheap electricity and high-temperature heat are available for the operation of the SOEC.

**Author Contributions:** Conceptualization: R.P., R.C.S. and N.W.; Methodology: R.P., F.S. and R.C.S.; Literature review and writing: N.W.; Process analysis ASPEN PLUS: N.W. and F.S.; Techno-economic analyses: F.S., N.W. and D.S.; Technical writing—original draft preparation: N.W.; Writing and visualization—review and editing: All authors; Supervision: R.P., J.R., M.G. and D.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study received no external funding.

**Acknowledgments:** The authors would like to thank all members of the Institute for Electrochemical Process Engineering (IEK-14) for fruitful discussions on all of the technologies of the energy transition and their role in the future energy system. Special thanks to the research groups of Ro. Peters and Q. Fang on SOCs, systems engineering, and electrochemistry for valuable discussions related to technical data on SOECs. The exchange of information with D. Schäfer deserves special mention. Additionally, we thank C. Wood for the language editing.

**Conflicts of Interest:** The authors have no conflict of interest to disclose.

#### **Abbreviations**


### **Appendix A. Effect of Chain Growth Probability on the Product Distribution**

In order to illustrate the influence of the chain growth probability on the product distribution, Figure A1 compares the product distribution of a Fischer–Tropsch synthesis for α = 0.88 and α = 0.92 for C1 to C60 according to the Anderson–Schulz–Flory distribution (Equation (12)). The course of the mass fractions of the various hydrocarbons was similar in both curves. First, the mass fractions increased with increasing chain length until a maximum was reached and the curve fell again. However, the curve for α = 0.88 rose much more rapidly in the area of short hydrocarbon chains and already reached the maximum at a chain length of *n* = 8. In addition, it flattened out very steeply, which means that the proportion of long-chain hydrocarbons in the product was very low. In comparison, the curve for α = 0.92 rose much more slowly and only reached the maximum at a chain length of *n* = 12, and then flattened out more slowly. Accordingly, the proportion of long-chain hydrocarbons in the product of the Fischer–Tropsch synthesis was significantly higher for α = 0.92, or for high chain growth probabilities in general. It follows that to maximize kerosene and diesel production in a power-to-fuel process, the chain growth probability of the Fischer–Tropsch synthesis should also be maximized.

**Figure A1.** Product distribution of the Fischer–Tropsch synthesis for α = 0.88 and α = 0.92.

#### **Appendix B. Key Data from Process Simulations**

**Table A1.** Energy-specific material balance of the developed fuel synthesis.


**Table A2.** Resources used.


**Table A3.** Energy-specific resource balance of the developed fuel synthesis. Negative values in the sum mean that excess steam is available from the process. See Table A2 for details on the resources used.



**Table A4.** Capacities of the Fischer–Tropsch reactor, the hydrocracker and the reformer.
