Combining Wind Energy with Power-to-Gas. A Case Study on Profitability and Economic Viability

Table of contents

List of figures

List of tables

List of appendixes

List of abbreviations

List of units

Nomenclature (case study)

List of equations

1 Introduction

2 Energy market environment and developments

3 Review of power-to-gas technology
3.1 Power-to-hydrogen
3.1.1 Working principle of different electrolysis cell technologies
3.1.2 Comparison of the AEL and PEMEL technology
3.2 Power-to-methane
3.2.1 Working principle of the chemical methanation
3.2.2 Comparison of chemical methanation plant designs
3.2.3 Working principles of biological methanation
3.2.4 Overview of different methanation plant concepts
3.2.5 CO₂sources
3.2.6 Overview of different CO₂sources for methanation
3.3 Power-to-gas facilities
3.3.1 Gas grid connected electrolyzer system
3.3.2 Gas grid connected methanation plant
3.3.3 Comparison of overall process efficiencies
3.4 Gas grid and power-to-gas related regulations

4 Literature review on applications of energy storage and power-to-gas
4.1 An overview of energy storage applications
4.2 Power-to-gas specific applications and business models
4.3 Power-to-gas and wind energy

5 Literature review on economical assessment of energy storage
5.1 An overview of energy storage valuation
5.2 Methods and results on the economic viability of hybrid systems
5.2.1 General overview of relevant literature on hybrid systems
5.2.2 Detailed presentation of three pivotal studies on hybrid systems
5.3 Review on the economic assessment of power-to-gas

6 A case study on the economic viability of wind energy and power-to-gas
6.1 General description of the applied model
6.1.1 The case study’s framework and data
6.1.2 Economic valuation method and assessment criteria
6.1.3 Reference cases
6.1.4 Working principle of the Excel model
6.2 Assessment of the economic viability of a wind-power-to-hydrogen plant
6.2.1 Results scenario I – transport grid with very high continuous gas flow
6.2.2 Results scenario II – supra regional grid with seasonal gas flow
6.2.3 Results scenario III – regional grid with seasonal gas flow
6.2.4 Sensitivity analysis
6.3 Assessment of the economic viability of a wind-to-methane plant
6.3.1 Results of the methanation plant
6.3.2 Qualitative assessment of additional income streams

7 Conclusion
7.1 Overview of case study’s results
7.2 Comparison to other studies
7.3 Implications and outlook
7.4 Overall summary

Appendix

References

List of figures

Figure 1: Renewable energy mix in 2013 and in a 100%-renewable energy scenario in 2050

Figure 2: Total wind curtailment as a function of wind energy penetration

Figure 3 Power-to-gas concept – A general overview

Figure 4: Schematic of the operating principle of an AEL cell

Figure 5: Schematic of the operating principle of a PEMEL cell

Figure 6: Schematic of the operating principle of a HTEL cell

Figure 7: Process efficiencies of an electrolysis and methanation plant

Figure 8: Typical electricity and gas price characteristics for selected days

Figure 9: Gas flow characteristics of a supra regional and regional grid

Figure 10: Overview of the working principle of the case study's Excel model

Figure 11: Annual losses [k€] of an AEL electrolyzer in the base case for several sizes

Figure 12: Annual losses [k€] of a PEMEL electrolyzer in base case for several sizes

Figure 13: Composition of total annual costs of a 5 MW AEL and PEMEL electrolyzer

Figure 14: Comparison of annual losses [k€] of methanation & electrolyzer plants

Figure 15: Additional revenue streams for the chemical methanation

List of tables

Table 1: Costs of water electrolysis - state of the art and future developments

Table 2: Energy consumption, capture cost and potential of different CO₂sources

Table 3: Ancillary units of a water electrolysis: techno-economical parameters

Table 4: Taxonomy of energy storage applications based on Eyer and Corey (2010)

Table 5: Curtailment assumptions and corresponding values

Table 6: Simplified technical assumptions made in Excel model

Table 7: Electrolysis - model parameters applied in this thesis’s case study

Table 8: Results for a 5 MW electrolyzer plant at a high curtailment level

Table 9: Methanation - model parameters applied in this thesis's case study

Table 10: Results for a 5 MW methanation plant at a high curtailment level

List of appendixes

Appendix I: Characteristics of water electrolyzers

Appendix II: Advantages and disadvantages of the AEL and PEMEL technology

Appendix III: Schematic of the chemical methanation process – Operational challenges

Appendix IV: Advantages and disadvantages of methanation technologies

Appendix V: Overall components of a gas grid connected electrolysis system

Appendix VI: Overview of relevant regulations and laws concerning power-to-gas

Appendix VII: Gas grid characteristics and scenarios analyzed by Müller-Syring et al.

Appendix VIII: Overview of selected papers on power-to-gas economic valuation

Appendix IX: Excel model input mask

Appendix X: Detailed results for scenario I - Transport grid

Appendix XI: Detailed Excel output – Scenario I, AEL, 5 MW, high curtailment

Appendix XII: Detailed Excel output – Scenario I, PEMEL, 5 MW, high curtailment

Appendix XIII: Detailed results for scenario II – Supra regional grid

Appendix XIV: Detailed results for scenario III – Regional grid

Appendix XV: Support schemes sensitivity for a 1, 5, and 10 MW electrolyzer

Appendix XVI: Gas price sensitivity for a 1 MWel electrolyzer

Appendix XVII: Wind park size sensitivity for several electrolyzer sizes

Appendix XVIII: Detailed results for a 1, 5, and 10 MW methanation plant

Appendix XIX: Detailed Excel output – Biological methanation, 5 MW, high curtailment

Appendix XX: Detailed Excel output – Chemical methanation, 5 MW, high curtailment

Appendix XXI: Conversion chart

List of abbreviations

illustration not visible in this excerpt

List of units

illustration not visible in this excerpt

Nomenclature (case study)

illustration not visible in this excerpt

List of equations

Equation 1: Definition of annualized investment costs

Equation 2: Definition of annual profit (or project value)

Equation 3: Definition of power-to-gas generation costs

Equation 4: Power-to-gas plant’s full load hours

Equation 5: Revenue from wind energy assuming no market constraints

1 Introduction

In the last years, the share of energy generated by renewable energy sources has sharply increased, meanwhile amounting to 25% of total energy consumption in Germany^[1]. Until 2050, 80% of electricity production shall be stemmed by renewables, whereas wind energy is supposed to generate the largest amount^[2]. However, with an increasing share of renewable energy – and especially wind energy – new technical and economic challenges arise. For instance, in many power systems, the real-time renewable availability is often negatively correlated with energy prices^[3]. It is furthermore expected that the amount of excess energy will rise with increasing market penetration of renewables^[4]. In this context, energy storage has recently gained great momentum. Amongst others, storage appears to be one promising option to mitigate the variable and unpredictable generation, and at the same time, is supposed to increase the value of renewable production^[5]. Against this background, one technology called power-to-gas (PtG) has been put in the limelight, since it is assumed to take over a major role in energy systems of the future^[6]. One reason is that the technology provides completely new possibilities, as electricity is stored in form of gas. For instance, a material use of the ‘renewable gas’ in the mobility sector is conceivable. Moreover, gas grid and caverns have a huge storage capacity, which could be used to seasonally store electricity in form of gas. Hence, compared to other technologies, large amounts of energy could be stored over a long period of time^[7].

Even though the chemical foundations are known for a few decades, the practical implementation has only started recently. The number of plants in operation is quite rare, and there are several technological aspects that are still in a research, pilot or demonstration phase respectively (see also chapter 3)^[8]. A number of practical projects have been initiated in Germany fostering to gather first experiences with wind-connected PtG plants^[9]. One of the largest German PtG plant was put into operation by Audi in 2013. At an industrial scale, the PtG plant has a plant capacity of 6 MW and is able to produce about 3.3 MWh methane (300 m³/h) per hour. The methane – produced with excess wind energy – could be further used in Audi’s gas-powered, climate-friendly A3 g-tron^[10].

In general, the idea that the value of wind energy could be increased by producing ‘renewable gas’ is appealing as shown with the example of Audi. However, the PtG technology is associated with high investment costs, which could compromise the benefits of such a system set-up. Hence, it is worth having a closer look at the economics of wind-to-gas plants. This thesis comprises a case study that investigates, if investing in a PtG plant that is connected to a wind park, is adequately remunerated from an investor’s point of view. Therefore, this thesis is organized as follows.

At first, chapter 2 introduces the current market setting in greater detail, which is supposed to influence the PtG technology today and in the near future. The following technology review in chapter 3 lays down the fundamental principles of the PtG concept. Specifically, the working principles of water electrolysis, methanation, and their underlying costs are addressed. Thereby, the basis for the parameters used in this thesis’s case study is establishes. Chapter 3 is closed by a brief review on gas grid and power-to-gas related regulations. With the foundations laid out, chapter 4 subsequently reviews which applications and business models can be implemented for energy storage systems such as PtG. This serves as a basis to understand how PtG plants could be implemented to accomplish a viable operation. Thereby, it is introduced how PtG is supposed to generate economic value in this thesis’s case study. Chapter 5 outlines how wind-storage projects (hybrid systems) have been economically evaluated in recent research. Three studies are presented in greater detail that evaluated wind-based hybrid system, since they appeared to be pivotal to establish the methodology applied in this thesis. Subsequently, a few papers are discussed that set initial approaches to determine the economic viability of PtG. Chapter 5 includes criticism and differentiation of existing literature. The thesis’s methodological approach is ultimately introduced in chapter 6. Thereinafter, the results found in the conducted case study are presented in detail. Finally, the results are subsumed and discussed in chapter 7, whereby the most striking points of the paper at hand are outlined, but also some possible limitations. Chapter 7 closes with an overall summary of this thesis.

2 Energy market environment and developments

Before going into detail of the PtG technology, it is worth having a closer look at the current market environment and future developments of the energy sector that have the potential to influence a successful implementation of a wind-to-gas plant on the market.

The German government has set ambitious climate and environmental protection goals that promote a large-scale penetration of renewable energy sources (RES) in the power generation sector. These objectives were constituted in the Renewable Energy Act (EEG), stating that, until 2050, at least 80% of the electrical energy production has to be generated by RES. Thereby, intermediate targets are set promoting an incremental increase of renewable energy production to 40 – 50% in 2025 and to 55 – 60% in 2035.

In 2013, RES accounted for 24.7% (147 TWh) of the total German electricity consumption. Wind energy was the largest contributor to the renewable energy mix with a share of 34%, and nominal capacity of about 34.2 GW. Together with photovoltaic (PV), wind energy represents 83% of the currently installed nominal power of renewable energies. In 2013, wind power generated 8% – that is 49.8 TWh – of total energy in Germany^[11]. In order to reach the above stated goals, researchers assume that PV and especially wind energy will play a major role in the future energy supply^[12]. Figure 1 depicts the renewable energy mix from 2013 and for an ideal future scenario with a 100%-RES integration in all energy sectors, meaning electricity, transport and heating.

Figure 1: Renewable energy mix in 2013 and in an ideal 100%-renewable energy scenario in 2050^[13]

Abbildung in dieser Leseprobe nicht enthalten

Gerhardt et al. (2014) analyzed electricity requirements, minimal load fluctuations, as well as viable and costs effects in order to determine the optimal future energy mix of 2050 (as presented in figure 1). As illustrated, in this ideal scenario, 73% of the total energy would be generated by on- and offshore wind parks. Furthermore, Gerhardt et al. (2014) assume that PtG would cover the residual requirements for chemical energy of all energy sectors (heating, electricity, transport) in such a scenario^[14].

However, with an increasing share of RES supply, new challenges arise for the whole energy system. Electricity supply and demand have to be guaranteed and balanced at all times, which is challenging due to the highly volatile and hardly predictable production of RES^[15]. Therefore, a huge body of research has investigated, how energy systems can maintain their stability, flexibility and security of supply, while facilitating an economic and large-scale RES penetration. Currently under discussion are, for instance, network expansion and modification, load management by using smart grids and markets, or a more flexible operation of conventional power plants^[16]. However, in the limelight of all proposed measures, is the integration of energy storage. In chapter 4 this aspect will be discussed in more detail, thereby presenting an overview of possible energy storage applications in the energy market. Against this background, many researchers see a tremendous potential of the PtG technology^[17].

Another factor that has to be considered, is the warranted support scheme for RES by law. In Germany, wind generators receive a statutory feed-in remuneration according to §49 and §50 EEG. Under the terms of EEG 2014, electricity from onshore produced wind is remunerated with an initial rate of 8.7 ct/kWh for a period of at least five years, followed by a base remuneration of 4.7 ct/kWh. For offshore wind energy, an initial rate of 15.4 ct/kWh is paid for a period of 12 years, followed by a basis rate of 3.9 ct/kWh. In comparison, the average day-ahead spot market price for electricity was 4.27 ct/kWh in 2012 and 3.78 ct/kWh in 2013^[18]. Hence, for a wind park operator, the current situation is quite comfortable, as the guaranteed feed-in tariffs mostly beat the electricity market price. In case, these remunerations would be cancelled in future, investments in wind parks are exposed to a much higher market risk. Some scholars argue that RES could not be funded via whole sale and control energy markets^[19]. With this regard, the combination of wind parks with energy storage was discussed in literature, examining if a storage system can increase the value of wind, in case it is freely participating on the energy market (see also chapter 4 and 5). This wind-value problem becomes even more apparent with the following issue:

With an increasing penetration of RES, the phenomena of excess energy is observable in times, when RES are highly available (e.g. good wind conditions) and/or demand is low, leading to a ‘demand-driven’ curtailment. Besides, ‘grid-driven’ curtailment can occur, when the electricity grid is not capable of transporting RES energy in times of high energy output^[20]. Under critical conditions, that is when all measures by conventional power generators are exhausted (§11 EEG), TSOs can temporarily reduce the feed-in from RES to prevent an overloading of the grid. In 2012, this ‘downpowering’ mainly concerned wind power (93.2%). The energy output loss for the same year amounted to 385 GWh, which corresponds to 0.71% of total wind energy production in 2012. Until now, wind park operators do not face any drawbacks, since the TSO must provide compensation for the curtailment of energy according to §12 EEG and §13a EnWG^[21]. A number of researchers examined, how the ‘downpowering’ of wind energy will develop with increasing shares of RES^[22]. Denholm and Hand (2011) ran simulations for the Texas ERCOT market^[23]. The dependence of curtailment on the fraction of wind generation is depicted in figure 2 (p. 6). Thereby, the system flexibility refers to the ability of an aggregated set of conventional generators to accommodate their output to the variable generation (e.g. 100% means a reduction to 0 GW). Even though, their results cannot be directly transferred to the German market, they are alarming, as they show that a significant amount of energy is at risk to get lost with increasing RES shares. Similarly, for the Danish market, Jørgensen and Ropenus (2008) estimate that at 50% wind power generation, in 20% of the time (1,600h) excess wind energy occurs. Hence, if wind curtailment is not remunerated in future, another financial risk would arise for wind park operators.

Another market factor, that is particularly interesting for the PtG concept, is the development of natural gas and hydrogen prices due predicted demand growth of gas and hydrogen in the related mobility sector. According to Nitsch et al. (2012) who established long-term scenarios under Germany’s climate goals (see above), PtG produced hydrogen and methane will play a key role in future transportation^[24]. In this context, the resource depletion and the increasing prices for conventional fuels are supposed to support the implementation of electrical and fuel cell-driven mobility, thereby levering the PtG technology, too^[25].

Figure 2: Total wind curtailment as a function of wind energy penetration^[26]

Abbildung in dieser Leseprobe nicht enthalten

3 Review of power-to-gas technology

In the following, the PtG concept is described in greater detail. Figure 3 provides an introductory overview of its fundamental process.

Figure 3 Power-to-gas concept – A general overview^[27]

Abbildung in dieser Leseprobe nicht enthalten

The PtG concept is categorized as a chemical energy storage system, since it converts electrical energy via chemical process – the electrolysis – to an energy-rich gas, namely hydrogen (H2). The different types of electrolysis are described and discussed in subchapter 3.1. In a second transformational process – the methanation – H2, together with carbon dioxide (CO₂), can be further converted to methane (CH4). Thereby, different plant concepts exist that operate with different reactor types, which will be introduced in section 3.2.1 and 3.2.3. In section 3.2.4, different possibilities of obtaining CO₂required for the methanation process are discussed. As figure 3 reveals, the final product of the PtG process can either be H2 or CH4, which can be stored, or respectively fed into the natural gas grid. Besides the cost for electrolysis and methanation plant components, additional investments for auxiliaries are necessary. A cost estimation for these components and a comparison of the overall system efficiencies is presented in chapter 3.3. Feeding H2 into the natural gas grid is limited to a certain amount. Related regulations are discussed in chapter 3.4. In contrast, CH4 can be fed into the grid almost unlimited. However, the second transformational process (methanation) induces further efficiency losses (see also section 3.3.3).

The thesis investigates how the decision for, or against the methanation step is influencing economics of a PtG plant by comparing two mutually exclusive alternatives: (1), feeding in H2, or (2), CH4 into the natural gas grid. The third option of storing H2 is not included in this thesis’s research issue. In the following subchapters, the economic parameters for the respective technology used in this thesis’s model are derived from current research.

3.1 Power-to-hydrogen

3.1.1 Working principle of different electrolysis cell technologies

The fundamental basis of the PtG concept is water electrolysis, which is described by the following basic endothermic chemical reaction:

Abbildung in dieser Leseprobe nicht enthalten ^[28]

Currently, three types of water electrolysis exist. The alkaline water electrolysis (AEL) and the proton exchange membrane electrolysis (PEMEL) are commercially available, and are therefore included in the analysis of this study. The high-temperature water electrolysis (HTEL) is a young technology that might provide a promising alternative in future. Therefore, its basic working principle is briefly described, too. The basic functioning of AEL and PEMEL, as described in the following, is complemented by a detailed comparison in section 3.1.2. Thereby, technical aspects of the two technologies are compared, and dis-/advantages concerning a combination with wind energy are contrasted.

Alkaline water electrolysis

Compared to its two alternatives (PEMEL and HTEL), the AEL represents the most mature and commercialized technology, which is established in practice for a few decades^[29]. As illustrated in figure 4, an AEL cell – housed in a compartment of steel – consists of two electrodes (anode and cathode) which are completely immersed in an aqueous potassium hydroxide (KOH) electrolyte, and separated by a microporous diaphragm separator^[30]. The alkaline KOH-fluidity is stored in two separated vessels which additionally serve as liquid-gas-separators^[31]. The electrolysis takes place when a direct current (DC) is applied to the AEL cell. Water splits into H2 and hydroxide-ions (OH–) at the cathode. The OH– -ions are migrating through the separator and are oxidized at the anode. As described by the equation (figure 4), the KOH-electrolyte is not consumed in the reaction, but has to be replaced over time due to system losses. To form an electrolysis module (stack), the cells can either be connected in parallel (unipolar) or in series (bipolar). An electrolysis stack is commonly composed of 30-200 single cells. Nowadays, the bipolar configuration is preferred over the unipolar design due to lower ohmic losses and a more compact design^[32].

Figure 4: Schematic of the operating principle of an AEL cell^[33]

Abbildung in dieser Leseprobe nicht enthalten

Proton exchange membrane electrolysis

Compared to AEL, the PEMEL is less developed and mainly used in niche applications. The first commercial PEMEL electrolyzer was developed by General Electric in 1966^[34]. Recently, the technology has gained a lot of attention due to overcoming some restrictions – as discussed in the next subchapter – of the AEL technology (see also subchapter 3.1.2)^[35].

The main difference is the acidic proton conducting membrane, which is used as a solid polymer electrolyte. Figure 5 shows the working principle of a PEMEL cell. The electro-catalyst layers are directly attached to the membrane, which is commonly referred to as membrane electrode assembly. Noble metals, such as platinum or iridium, are used as electro-catalyst materials due to the high applied voltages at high current densities. The membrane consists of Nafion, which shows good mechanical and electrochemical stabilities. Water is fed into the anode compartment and is oxidized electrochemically to form oxygen, H2-ions and electrons. The H2-ions are migrating through the membrane towards the cathode, where they recombine with the electrons to form hydrogen gas^[36]. The stack is exclusively constructed by connecting single cells in series (bipolar electrolyzer) and is commonly composed of up to 60 single cells. These are at least 5-10 times smaller as compared to the ones applied in AEL technology. Hence, a PEMEL electrolyzer is typically quite compact due to the lack of a liquid electrolyte and other necessary equipment, such as pumps or gas separators^[37].

Figure 5: Schematic of the operating principle of a PEMEL cell^[38]

Abbildung in dieser Leseprobe nicht enthalten

High-Temperature Water Electrolysis

HTEL represents the least mature water electrolysis technology, being researched since the early 1970’s. In Germany, the technology was supported by a government-encouraged project called HOT ELLY (High Operating Temperature Electrolysis) from 1975 to 1987^[39]. However, the technology is still in the phase of basic research and not commercially available yet^[40]. Therefore, HTEL is not part of this thesis’s analysis.

In the so-called solid oxid electrolysis cell (SOEC)^[41], a dense solid ceramic material is used as electrolyte, which becomes conductive for oxygen ions at high temperatures. The temperature ranges from 700 – 1000 °C and is therefore beneficial for thermodynamic and kinetic reasons^[42]. Thus, it supports the endothermic splitting of water and reduces the electrical energy demand about 25%^[43]. However, at such high temperatures, the fast degradation of the cell components represents a severe problem^[44]. Moreover, the HTEL system has a reduced load dynamic, which makes it less attractive for the integration with renewable energies^[45]. This is another reason, why the technology is not incorporated in this thesis’s case study. Figure 6 shows the operating principle of an HTEL cell. Water vapor is fed at the cathode side, where the reduction of water takes place^[46]. The generated oxygen ions move towards the anode, where they discharge electrons and form oxygen gas^[47].

Figure 6: Schematic of the operating principle of a HTEL cell^[48]

Abbildung in dieser Leseprobe nicht enthalten

3.1.2 Comparison of the AEL and PEMEL technology

To answer the question, which electrolysis is the most favorable alternative for the combination with wind power, one needs to understand their fundamental differences.

Appendix I provides an overview of the technologies’ current state of the art and expected developments. Hereinafter, a few aspects are discussed by means of expected future developments. As described above, an AEL cell (~ 4 m²) requires a lot more space than a PEMEL cell (~ 0.5 m²). This might be relevant, when choosing a location for an electrolyzer plant. However, a PEMEL module yields lower production rates (500 Nm³/h) than an AEL module (1,500 Nm³/h). Thus, it creates the need of various interconnected modules for larger electrolyzer plants. Currently, larger electrolyzer plants are only realizable with the AEL technology.

Especially interesting for the operation with an intermittent RES, such as wind, is the lower limit for the partial load of an electrolysis system. Considering this aspect, an AEL is inferior to a PEMEL system, as an AEL system requires a continuous load of roughly ten percent of the nominal H2 production, whereas no partial load is necessary for a PEMEL system. Furthermore, a PEMEL system is supposed to reach higher efficiency (~73.2 %) as compared to an AEL system (~68.8 %) in the long-term. Generally, since the technologies are investigated heavily in current research, further technology developments are expected, which will most likely lever the PEMEL technology. Besides these technical aspects, the costs of the electrolysis technologies have to be considered as well. As the electrolysis is the central element of the PtG concept, its cost determine the present and future economic feasibility. Table 1 depicts the investment, fixed and variable cost of the respective technologies, depending on the size of the electrolyzer. The cost information is adopted from Stolzenburg et al. (2014), who based their assumptions on the research of Smolinka et al. (2011). As can be seen in table 1, the investment costs of AEL are much lower than of PEMEL technology. In contrast to other researchers, Smolinka et al.’s (2011) assumptions regarding the cost reduction potentials with increasing marketability of PtG are rather conservative. For instance, Bajohr et al. (2011) and Wenske (2011) state that investment costs can potentially drop to 500 €/kW for the AEL (> 1 MW), and to 1000 €/kW for the PEMEL technology (> 500 kW) in future^[49].

Table 1: Costs of water electrolysis - state of the art and future developments^[50]

Abbildung in dieser Leseprobe nicht enthalten

In case the PtG concept is operated in an intermittent manner, the PEMEL technology is technically favorable. However, its expected capital costs are much higher as compared to AEL technology. This thesis’s model will assume an intermittent operation of the PtG plant. Thereby, the question shall be answered which technology is economically favorable for such an operation mode. Hence, the analysis in chapter 6 outlines, whether cost disadvantages of PEMEL can be compensated by its technological advantages, in case it is combined wind energy. A detailed qualitative review of all dis-/advantages is provided in appendix II.

3.2 Power-to-methane

As presented in figure 3 (p. 6), the PtG concept may also include a methanation process, in which H2 and CO₂, or carbon monoxide respectively is converted to methane (CH4). This so-called Sabartier process is described in the following section 3.2.1. The chemical methanation, as compared to the electrolysis, is a rather mature technology, since its principles are known for more than one century and the chemical systems are state of the art for several decades^[51]. However, when the chemical methanation process is part of the PtG concept, specific challenges arise. These are also discussed in the section 3.2.1. Moreover, four different plant configurations, which differ along the dimension of their reactor type, are presented in section 3.2.2 and evaluated for their best fit into the intermittent operation of a RES. A rather young idea is biological methanation, which is presented in section 3.2.3. Chemical and biological methanation are then contrasted in section 3.2.4. The required CO₂can be obtained from four different sources, which are presented in section 3.2.5.

3.2.1 Working principle of the chemical methanation

Methanation comprises three steps: (1), a catalytic hydrogenation of carbon dioxide to methane (eq. I), combined with (2), a reversed endothermic water shift reaction (eq. II), and (3), an exothermic methanation of carbon monoxide (eq. III)^[52].

Abbildung in dieser Leseprobe nicht enthalten

Reaction (I) and (II) are strongly exothermic, whereby the following reaction enthalpies (H0R) are given for a temperature of 25° C. Although the reaction mechanism is underexplored, equation (I) is often interpreted as the sum of reaction (II) and (III), meaning that CO₂methanation is achieved by an intermediate conversion to carbon monoxide (CO)^[53].

Even though the methanation of CO₂is an exergonic reaction, a catalyst is required to achieve full reduction of the oxidized carbon atom. Nickel- and ruthenium-based catalysts have been found to be most effective for methanation^[54]. Depending on the catalyst type, the reaction temperature typically ranges between 200° and 500° C, whereby pressure ranges between 1 and 100 bar^[55]. Lower temperatures and higher pressure result in better conversion rates, whereas lower temperature might cause unfavorable reaction kinetics (reaction speed)^[56].

3.2.2 Comparison of chemical methanation plant designs

As mentioned before, the process of methanation reached industrial maturity after the oil crisis in the late 1970s. Primarily, this development is attributable to intensified research activities, which were initiated in order to find a natural gas substitute^[57]. At this time, coal gasification was the state of the art technology to produce CH4, which was also referred to as substitute natural gas (SNG)^[58]. Nevertheless, due to falling energy prices in the following years, the SNG production slowed down, but returned around the year 2000, when biomass gasification became of interest^[59]. However, the previously developed plant concepts for coal gasification were neither technically, nor economically applicable to the biomass-to-gas concept. Hence, new process developments were initiated. The chemical methanation plant and its system processes, which were developed since the 1970s, can be classified as follows^[60]:

- 2-phase systems (gaseous educts, solid catalyst)

- Fixed bed
- Fluidized bed
- Coated honeycombs

- 3-phase systems (gaseous educts, liquid heat carrier, solid catalyst)
- Slurry bubble column

Nowadays, the above listed methanation processes are intensively researched. Primarily, their suitability for a RES-operated PtG system is investigated. Moreover, biological methanation, which is still in a laboratory and pilot stage, has gained a lot of attention in recent years, since it seems superior to chemical processes in some regards.

Against this background, the figure in appendix III shows the chemical methanation process and its challenges. When methanation is coupled with a RES, H2 is unsteadily generated by the electrolysis, due to the fluctuating input of the RES. Such an intermittent operation mode is also investigated in this thesis’s case study. However, the chemical methanation has to be operated at elevated temperatures and pressures, leading to a reduced start-up and shut-down flexibility of the plant. Depending on the reactor concept, the sensitivity to load changes differs, generating the need for a tank to store the intermittently produced H2. Its size depends on the load flexibility of the respective reactor, and the fluctuating feed stream of the electrolysis. For similar reasons, the intermittent storage of CO₂is necessary, too. Besides, both gases have to be compressed to the operating pressure of the methanation reactor^[61]. Due to the strong exothermic reaction, the most critical factor is the underlying heat management of chemical reactors^[62].

Fixed bed methanation

Within the fixed bed reactor (FBR), the catalyst exists in form of solid particles in millimeter-range, thereby building a static catalyst bed^[63]. The preheated reactant gas, with temperatures of 250° to 300° C, flows through the catalyst bed and rises significantly in temperature due to the strong exothermic reaction. To avoid local temperature peaks (hot spots) within the solid catalyst bed, which may lead to catalyst destruction, the reaction heat has to be controlled carefully. Moreover, due to thermodynamic reasons, conversion rates and selectivity decreases with temperatures higher than 400° to 500° C. To avoid these effects, several reactors are connected in a row, whereby gas cooling, gas recycling and reaction heat recovery takes place between each of the reactor steps^[64]. As part of the fixed bed reactor technology, various processes exist, which operate with a different number of reactors, resulting in a complex system interconnection^[65]. Yet, the Lurgi, TREMP and Linde processes are commercially available, whereas some processes such as HYGAS or HICOM are still in a pilot phase.

Fluidized bed methanation

Within the fluidized bed methanation (FLBR), the solid catalyst particles are much smaller than within the fixed bed reactor, and are fluidized by the flow gas^[66]. Due to the fluidization of these particles, the reactors show a good heat release profile and a high specific surface area of the catalyst. Hence, reactor cascades, as required in FBR, are not necessary. This leads to a simplified system set-up. However, the gas flow is limited to a certain range to guarantee the whirling of the catalyst material, resulting in limitations of unsteady operation. Another drawback is the high mechanical stress on the catalyst, resulting in abrasion of the catalyst material and reactor internals. For the FLBR, different processes are available. Nevertheless, all of them, such as the Comflux or Bi-Gas technology, are still in a pilot phase^[67].

Slurry bubble columns

Within the slurry bubble column reactor (SBCR), the methanation process is split into a three-phase system, comprising of (1), gaseous educts, (2), solid catalysts and (3), a liquid carrier medium^[68]. The catalyst material is bloated in mineral oil and fluidized by the ascending gas bubbles^[69]. Compared to the fluidized bed reactor, the catalyst abrasion is reduced. However, the mineral oil, which is used as a liquid heat carrier medium, is prone to degradation as a result of the oil’s reduced temperature stability^[70]. The SBCR was developed by Chem System Inc. in 1972, but production and research was stopped in the 1980s. Primarily, the investigation phased out due to the aforementioned problems regarding the mineral oil’s inappropriate temperature characteristics. However, the concept was recently revitalized by Forschungszentrum Karlsruhe and Deutscher Verein des Gas- und Wasserfaches e.V. (DVGW) in Germany. Thus, current research activities concentrate on finding suitable ionic liquids to overcome the above mentioned challenge^[71]. In general, the liquid heat carrier medium enables effective and accurate heat control. Hence, a three-phase reactor is less prone to quick temperature changes in case of fluctuating feed streams, compared to the above presented reactor technologies^[72].

3.2.3 Working principles of biological methanation

The aforementioned reactor types belong to the cluster of chemical methanation processes, which are characterized by a metallic catalyst. Within the biological methanation, the chemical catalysts are substituted by enzymes, which serve as bio-catalysts. Hence, the methanation of H2 and CO₂is carried out in a biological system. The enzymes are produced by methanogenic bacteria, which belong to the domain of Archaea – a thermophilic organism, which uses H2 and CO₂to produce CH4. Biological methanation follows two main reaction paths^[73]:

Abbildung in dieser Leseprobe nicht enthalten

Currently, there are two different types of biological reactors, namely pure and mixed culture reactors. In a pure culture bioreactor, the Archaea organisms exist in pure form. Besides the educt gases, CO₂and H2, the Archaea organisms only need minerals being dissolved in water to grow. The methanation reaction is described by equation (IV). Besides the supply of the Archaea with trace elements, it is necessary to eliminate water, as well as fermentation waste products, which emerge during the methanation process. The sewage can be used as a fertilizer in biogas plants, or needs to be purified in wastewater treatment plants respectively. Under optimal conditions, the reactor can be brought from stand-by mode to full-load operation in several seconds. Moreover, load changes seem to have no negative effects on the bacteria^[74]. A linkage to biogas processes is theoretically possible^[75]. The second design of biological methanation is using a reactor, which operates with a mixed microbe population. This bacterial culture consists of various methanogenic bacteria strains, which can be found in already existing processes, such as in biogas or sewage treatment plants^[76]. Therefore, the process is also called integrative methanation. Equation (V) shows the typical process route for the decomposition of biomass^[77]. In case the mixed culture methanation is linked to a biomass plant, the supply of minerals is not necessary, since they are already part of the fermentation substrate. The fact that already existing reactors, power connections and the gas infrastructure of biomass plants can be used for integrative biological methanation, serves as its biggest advantage. Hence, by using this technology, investment costs of a PtG plant could be reduced^[78].

In general, the biological methanation takes place at moderate temperatures (30-60° C) and atmospheric pressures. Moreover, the process shows a higher tolerance against pollutant substances in the feed gas. Against this background, the biological methanation has recently gained increasing attention, as neither heat control, nor stress on the catalyst is an issue as compared to chemical methanation^[79].

3.2.4 Overview of different methanation plant concepts

Although the chemical methanation process is a rather mature technology, its integration within the PtG concept raises some challenges, as well as the need to develop new reactor technologies and plant designs. In the preceding section 3.2.2, such existing designs were introduced. Yet, this chapter provides an overview of the current status and potential developments in the field of PtG connected methanation plants.

The most important criteria for integrating wind-energy with a PtG concept, is a flexible operation of the underlying methanation plant^[80]. Although the heat removal is quite effective and the mass transfer very good, FLBR is not suitable for a PtG concept with fluctuating H2 supplies^[81]. Hence, the FLBR is considered to play a minor role in the PtG plant design^[82]. A similar concept is the honeycomb-coated methanation, where the catalyst is put up the metallic honeycomb structures. Since this methanation neither shows a good load flexibility, it is not further discussed^[83]. Regarding the concept of chemical methanation, the FBR and the SBCR seem to be rather promising processes for a PtG concept. So far, only FBR methanation is available on an industrial scale, whereas the SBCR is still in a pilot and development phase^[84]. In contrast to FLB methanation, the flexible operation of reactors is possible. However, recent studies assess the feasible load flexibility different. Therefore, several research teams, such as from ZSW Stuttgart, Engler-Bunte-Institut Karlsruhe, Max-Planck-Institut, TU München and FAU Nürnberg-Erlangen, are still investigating the ability of FBR to handle load changes, start-up and shut-down, as well as reaction behaviors^[85]. Hence, future research will need to finally clarify, if the reactors can be coupled with an intermittent energy source.

As discussed above, the integration of biological methanation within the PtG concept is a rather new idea. Current research focuses also on the question, whether an intermittent operation bears any negative consequences for the methanogenic bacteria^[86]. Even though laboratory tests did not reveal any drawbacks in the productivity of bacteria during flexible operation, long-term investigations need to validate the stability of these results^[87]. In general, due to its expected flexibility and comparatively low investment costs, the biological methanation shows a very high potential for integration within the PtG concept.

Another criterion to compare different methanation processes is the handling of reaction heat. Especially for FB reactors, heat control and release is critical, since temperature hot spots destroy the catalyst and thus derogate the whole methanation process. Therefore, several FB reactors build a cascade. In contrast, in a single SBRC, the exothermic reaction can be handled very effectively. For the biological methanation, heat control is not an issue, since the reaction takes place at significantly lower temperatures. Appendix IV provides a detailed overview over the advantages and disadvantages of the introduced methanation technologies.

In contrast to electrolysis, just a few papers investigated the cost differences between respective reactor designs. One possible explanation might be that only fixed bed methanation is commercially available, whereas all other technologies are still in a laboratory or pilot stage. One paper published by the Dutch company DNV KEMA compares investment cost of chemical and biological methanation. Grond et al. (2013) estimated the investment cost for chemical methanation, depending on the respective plant capacity. Currently, the lowest estimate of investment cost amounts to ~800 €/kWCH4 at a plant capacity of 7000 kWCH4, but is assumed to drop to ~300-500 €/kWCH4 (~ 160-280 €/kWel) with an ongoing standardization^[88]. For biological methanation, investment cost were estimated to be comparatively low, ranging from 100 €/kWCH4 at a capacity of 2000 kWCH4, to maximal 800 €/kWCH4 at a capacity smaller than 100 kWCH4. Researchers point out that the most challenging aspect of biological methanation is the development of a fully operational controllable system, as well as the up-scaling of the technology to larger MW-ranges. However, the technology is expected to quickly enter the market, because of being an inexpensive and flexible alternative to produce renewable CH4^[89]. A more detailed insight into current investment cost of different biological methanation plant concepts is provided by Graf et al. (2014a). By comparing various plant concepts offered by two distinct vendors (MicrobEnergy Gmbh and Krajete GmbH), they found that the integration of a fermentation plant yields the lowest investment cost, amountin to 349 €/kWCH4 (for a 5 MW plant)^[90]. For the simple reason of better comparability to other figures presented in this thesis, a conversion between €/kWCH4 and €/kWel can be achieved by multiplication with the efficiencies of the respective electrolysis and methanation (e.g. 70% and 80%)^[91].

3.2.5 CO2 sources

Besides the methanation process and the respective reactor types, the CO₂supply of methanation is another critical parameter for the overall PtG concept. Various sources of CO₂are theoretically possible, which can be differentiated among their origin (e.g. fossil or biogenic), their processing techniques, their costs, and achieved gas purity. In general, CO₂ is often an unavoidable waste product that evolves in fossil combustion, industrial processes, as well as in biomass gasification and fermentation. Usually, no market value is assigned to CO₂^[92]. In the following subchapters, the different possibilities of obtaining CO₂ are briefly described, closing with an economical comparison in section 3.2.6.

CO₂recovery from air

One obvious approach to obtain CO₂ is represented by the recovery from atmosphere (air capture). To extract CO₂from air, various technologies exist, which can be grouped into adsorption on solid sorbents or in liquid solutions, condensation in cryogenic distillation processes, as well as membrane separation. However, the latter two are far too energy-intensive to be economically feasible for CO₂recovery within the PtG concept. The standard technology for cleaning and conditioning is the adsorption process. To extract 1 kg CO₂out of the atmosphere, demanded energy amounts to 8.2 MJ (~2.3 kWh). To do so, 2,300 m³ of air have to be processed. In general, the recovery of CO₂from atmosphere reduces the renewable-power-to-methane efficiency by about 15%^[93]. Compared to all other CO₂capture technologies, air capture is the most energy-intensive and technically complex process, making it the least attractive alternative. Only if CO₂-capturing from air becomes necessary due to climate protection reasons, the technology may provide a suitable option in future^[94].

CO₂obtained from biomass

As biomass contains CO₂ that was originally captured via photosynthesis from air; obtaining CO₂from biogenic sources is the most environmental friendly alternative. The PtG concept can be combined with two main bioenergy conversion lines: gasification and fermentation^[95]. Especially biogas plants represent a promising CO₂source. The following two system set-ups are available:

a) Biogas plants with gas grid connection (CO₂separation)

b) Direct methanation of biogas (direct feed in of CO₂ / CH₄)

Biogas plants (biomethane plants) constitute a technical and economical favorable CO₂source, since within the biogas upgrading process, highly-concentrated CO₂emerges (up to 99 vol.-%), which is already separated (alternative (a)). Moreover, the proximity to the existing gas grid connection of biogas plants represents another interesting aspect. Besides these benefits, a combination of biogas, electrolyzer and methanation process bears heat recovery potentials that may lead to further synergies^[96]. Another alternative of integrating biogas plants in the PtG concept, is that the methanation is directly fed with a raw biogas stream (alternative (b)). In contrast to alternative (a), CO₂ is not separated via gas-conditioning, but the CO₂/CH₄ gas mixture is upgraded to highly-concentrated methane via the methanation process (with > 90 mol- % CH₄). Besides these two alternatives, a few more possibilities exist to gain CO₂ from biogenic sources^[97]. In this respect, bioethanol plants and sewage plants are worth to be mentioned due to their relatively high CO₂potentials in Germany^[98].

CO₂via carbon capture and storage

In addition to biogenic sources, CO₂is often a byproduct of conventional electricity production, which can be captured in fossil power plants via carbon capture and storage (CCS). Within the CCS technology, three main options exist to separate CO₂from the flue gas: (1), the post-combustion (separation after the combustion process), (2), the pre-combustion (separation in the gasification process), and (3), the oxy-fuel technology (combustion with oxygen). The captured CO₂can be compressed, liquefied, and transported by trucks, trains, pipelines and tankers to the methanation plant site^[99]. Yet, the technology is still emerging and has not been introduced on a large scale^[100]. Criticism includes that capturing CO₂in fossil plants, combined with its subsequent usage in methanation, is not ‘climate neutral’. In other words, the original idea of CCS, where obtained CO₂is stored, is not met, since CO₂enters the atmosphere at a later point of time. Against the background of the ‘energy system transformation’ (Energiewende) towards a climate friendly and renewable energy supply, a reasonable integration of carbon-intensive fossil power plants within the PtG concept is questionable. Besides this social-ecological aspect, cost issues need to be considered as well. As the storage of CO₂is just temporarily, emission certificates might have to be purchased. Moreover, one has to keep in mind that additional energy is necessary to capture CO₂in fossil power plants, which leads to a decreased overall plant efficiency, and at the same time, to an increased primary energy demand of about 20% to 44%^[101].

CO₂ from other industrial processes

In principle, CO₂emerges as a waste gas in various industrial processes, such as in cement production, steel manufacturing, chemical industry, plastic production and refineries^[102]. Even though the obtained CO₂is not ‘climate neutral’ (see also p.20), an integration with the PtG concept seems reasonable. Primarily, this is because emissions in industry are inevitable in the near future, and at the same time, also independent of the energy production sector^[103]. Using CO₂from industrial process is generally a very cost-effective option, since the gas is an unavoidable byproduct not having any market value. However, the gas purity of CO₂varies across above mentioned industries, sometimes making it complex and costly to attain the necessary purity level. For instance, the CO₂concentration in waste gas of cement and steel industry is rather low. In contrast, the CO₂waste stream of chemical processes shows a very high CO₂purity (up to 100 mol- %), making it attractive for PtG operation^[104]. The techniques to capture CO₂from flue gas are commonly the ones also used in CCS. Yet, the specific energy need to obtain CO₂depends on the gas composition and capturing method^[105].

3.2.6 Overview of different CO2 sources for methanation

The previous subchapters briefly described the theoretically possible CO₂sources, which are now compared from an economic point of view. As introduced above, capturing CO₂from the atmosphere is an appealing idea, since the gas could be gained independently from location and third parties. However, this option is the most expensive one due to its high energy demand, and therefore high costs. As can be seen in table 2 (p. 22), the technology is not supposed to be cost competitive in near future^[106]. Another option to gain CO₂is its separation in fossil power plants via CCS. Finkenrath (2012) calculated the cost to capture CO₂, thereby comparing different CCS technologies (post-/pre-combustion, oxyfuel) and fuels (coal vs. gas). Capturing CO₂from coal fired plants is less expensive (43-58 USD/tCO₂) than from gas fired plants (80 USD/tCO₂), whereby pre-combustion is associated with the least cost^[107]. Biomass and industrial processes, in which CO₂emerges as waste product, are more promising sources for the PtG concept. The costs for obtaining CO₂from industrial processes depend on its capturing method and composition of the waste gas stream. Against this background, the chemical industry, which produces about 16 million tons of CO₂per year, is the most interesting source, since CO₂is already highly-concentrated in the flue gas. At best, the cost to capture and transport CO₂account for 25 €/tCO₂. However, a possible consideration could be, to cooperate with certain industries in a way that their emissions certificates are reduced. Such a cooperation could lower the purchasing costs of CO₂, or even turn CO₂in another income stream^[108]. In biogas plants connected to the gas grid, CO₂has to be separated from produced methane anyway. Thus, the cost to obtain CO₂, which amount to maximal 90 €/tCO₂(see table 2), could be also ascribed to the biogas plant operator.

Abbildung in dieser Leseprobe nicht enthalten

Table 2: Energy consumption, capture cost and potential of different CO₂sources^[109]

3.3 Power-to-gas facilities

So far, key components of the electrolysis and methanation were presented. The cell stack (AEL or PEM cells) appears to the most important element within the power-to-hydrogen conversion step, whereas the reactor concept, as well as the supply with CO₂, is crucial for methanation. For the operation of both, auxiliary systems are necessary that are shortly presented in the following subchapters. In respect of the model presented in chapter 6, electrolysis is first of all considered as a stand-alone system with direct feed-in of H2 into the natural gas grid (see section 3.3.1). The second option is the realization of a power-to-methane system, in which the electrolyzer provides the required H2 (see section 3.3.2).

3.3.1 Gas grid connected electrolyzer system

One possible PtG design is that H2 produced via electrolysis is directly fed into the natural gas grid. Water electrolyzers are normally offered as an all-in-one system, meaning that most ancillary components – as illustrated in appendix V – are already included in the calculation of investment cost^[114].

As outlined in section 3.1.2, estimated costs for the electrolysis are adapted from Stolzenburg et al. (2014) (see table 1, p. 11). However, these figures exclude the cost for the connection to the power grid, as well as the compressor and buffer storage^[115]. To run a wind power-to-gas grid system, these components have to be installed and included in a cost evaluation. Therefore, the necessity and costs of these parts are subsequently presented. The electrolyzer has to be connected to an electricity grid, whereby a rectifier converts the alternating current (AC) to direct current (DC). Apart from the rectifier, transformers and a cooling system are necessary for the power grid connection. Today, these technologies are already fully developed and no significant improvements are expected in the near future^[116]. The produced H2 needs to be stored temporarily in a buffer storage tank, before it enters the gas injection unit. In order to feed H2 into the grid, H2 has to pass a gas injection system that also implies a compressor^[117]. Table 3 provides a cost estimation for the aforementioned components. Stolzenburg et al. (2014) aligned these ancillary units to be suitable for an electrolyzer with a nominal power of 500 MWel^[118].

Table 3: Ancillary units of a water electrolysis: techno-economical parameters^[119]

(I) Aboveground steel-cylinder storage system

(II) Data based on a compressor suitable for feeding a cavern storage facility

Abbildung in dieser Leseprobe nicht enthalten

3.3.2 Gas grid connected methanation plant

An overview of the methanation concept is provided in appendix III (or section 3.2.2). The auxiliaries necessary to run the process of methanation are storage tanks, a compressor, gas purification systems and a gas grid injection unit^[120]. Hence, similar components as for an electrolysis system are necessary. As discussed in section 3.2.2, the educts H2 and CO₂are temporarily stored in a tank before entering the reactor. Both gases need to be compressed to the operational pressure of the methanation system^[121]. Moreover, ancillary units are required to purify H2 and CO₂feed streams^[122]. The effort to purify CO₂depends on the origin of the CO₂source, whereby eliminating tar and ammonia particles, as well as hydrogen sulfide, is especially challenging^[123]. Contrary, the main impurity expected to occur for H2 produced by electrolysis is oxygen, which is much easier to eliminate. The produced CH4 has already a high purity level, but needs to be dried and compressed before it can be fed into the gas grid^[124].

Sterner (2009) estimates the costs of a whole system (5-10 MWel) - comprising an electrolyzer, methanation, compression, power electronics, piping, civil construction and control systems - to amount to 2,000 €/kWel. With upscaling of the PtG concept to 20 – 200 MWel, the costs are expected to drop to 1,000 €/kWel by 2020^[125]. Still, the electrolysis is expected to amount for the greatest share of the PtG concept, and was estimated to account for 86% of total cost of a power-to-methane plant^[126].

3.3.3 Comparison of overall process efficiencies

Besides the total system costs, it is important to consider overall process efficiency. Thereby, the efficiencies of all components have to be multiplied in order to get an estimation for the overall system efficiency. Müller-Syring et al. (2013) depicted the necessary components (as also described in the prior sections) in case that H2 or CH4 are fed into the natural gas grid (see figure 7). For a stand-alone electrolyzer system, the efficiency is about 64%. Adding methanation to the process, the total efficiency decreases to about 51%.

Abbildung in dieser Leseprobe nicht enthalten

With regard to the efficiencies of transformers and rectifiers, Stolzenburg et al’s (2014) assumptions would yield a combined component efficiency of 97.5%. Unless these minor deviations, the estimations are mostly in line with other research results. For instance, Sterner, Jentsch, and Holzhammer (2011), who outlined the overall efficiencies for several process chains, derived similar system efficiencies. For a gas grid connected electrolyzer, they estimate the system efficiency to range between 57 – 73%, and for a gas grid connected methanation plant to range between 50 – 64%. With regard to their data, it gets foremost obvious, that a reconversion of gas to electricity shows very low round trip efficiencies, e.g. electricity-to-H2-to-electricity accounts for about 34 – 44% only^[128]. This is one reason why the latter option is not further exploited in this thesis.

[...]

---

^[1] See Berkhout et al. (2014), p. 9.

^[2] For instance, see Saint-Drenan et al. (2009), p. 4.

^[3] Sioshansi (2010), p. 11.

^[4] For instance, see Rundel et al. (2013), p. 20.

^[5] For instance, see Sioshansi (2010), p. 11.

^[6] For instance, see Agora Energiewende (2014).

^[7] For instance, see Agora Energiewende (2014), p. 41.

^[8] For a detailed review of PtG plants that were installed worldwide until 2013, the interested reader shall be directed to Gahleitner et al. (2013).

^[9] For an interactive project chart in Germany see the website of Deutsche Energie-Agentur (2015).

^[10] Nerreter (2015).

^[11] See Berkhout et al. (2014), p. 9.

^[12] For instance, see Gerhardt, Sandau, and Pape (2014), p. 67.

^[13] Source: own illustration, based on Berkhout et al. (2014), p. 9; Gerhardt et al. (2014), p. 69.

^[14] See Gerhardt et al. (2014), p. 68.

^[15] For instance see Rundel et al. (2013), pp. 7-9.

^[16] For instance, see Agora Energiewende (2014), pp. 33-50.

^[17] For instance, see Agora Energiewende (2014); Rundel et al. (2013), p. 9; Gerhardt et al. (2014), p. 68.

^[18] See Berkhout et al. (2014), p. 12.

^[19] See Leprich and Klann (2014), p. 73.

^[20] For instance, see Loisel et al. (2010), p. 7329.

^[21] See Berkhout et al. (2014), p. 26.

^[22] For instance, see Jørgensen and Ropenus (2008); or Denholm and Hand (2011).

^[23] See Denholm and Hand (2011), p. 1821.

^[24] See Nitsch et al. (2012), p. 6.

^[25] See Joest et al. (2009), pp. 9-11.

^[26] Source: own illustration adapted from Denholm and Hand (2011), p. 1822.

^[27] Source: own illustration based on Schiebahn et al. (2013), p. 814.

^[28] See Schiebahn et al. (2013), p. 814; or Müller-Syring et al. (2013), p. 107.

^[29] See Schiebahn et al. (2013), p. 814; or Müller-Syring (2013), p. 115.

^[30] See Lehner et al. (2014), p. 24.

^[31] See Smolinka, Günther, and Garche (2011), p. 10.

^[32] See Lehner et al. (2014), p. 25.

^[33] Source: own illustration adapted from Lehner et al. (2014), p. 25.

^[34] See Schiebahn et al. (2013), p. 817.

^[35] See Lehner et al. (2014), p. 29.

^[36] See Schiebahn et al. (2013), p. 817; Lehner et al. (2014), p. 29.

^[37] See Lehner et al. (2014), p. 29.

^[38] Source: own illustration adapted from Lehner et al. (2014), p. 30.

^[39] See Smolinka et al. (2011), p.15.

^[40] See Schiebahn et al. (2013), pp. 818-819.

^[41] This is why the technology is also called solid oxide electrolyte electrolysis.

^[42] See Lehner et al. (2014), pp. 33-34.

^[43] See Smolinka et al. (2011), p. 16.

^[44] See Lehner et al. (2014), p. 34.

^[45] See Smolinka et al. (2011), p. 17.

^[46] See Lehner et al. (2014), p. 34.

^[47] See Schiebahn et al. (2013), p. 819.

^[48] Source: own illustration adapted from Lehner et al. (2014), p. 30.

^[49] Bajohr et al. (2011), p. 205; Wenske (2011), p. 10.

^[50] Source: own illustration based on Stolzenburg et al. (2014), pp. 71 – 75.

^[51] See Lehner et al. (2014), p. 41.

^[52] See Schiebahn et al. (2013), p. 820.

^[53] See Lehner et al. (2014), p. 42.

^[54] See Schiebahn et al. (2013), p. 820.

^[55] See Lefebvre et al. (2015), p. 83.

^[56] See Lehner et al. (2014), p. 42.

^[57] See Bajohr et al. (2011), p. 205.

^[58] See Lehner et al. (2014), pp. 42-43.

^[59] See Lehner et al. (2014), pp. 42-43.

^[60] See Bajohr et al. (2011), pp. 205-206.

^[61] See Lehner et al. (2014), p. 51.

^[62] See Bajohr et al. (2011), p. 84.

^[63] See Götz (2014), p. 8.

^[64] See Lehner et al. (2014), p.43.

^[65] See Bajohr et al. (2011), p. 206.

^[66] See Bajohr et al. (2014), p. 206.

^[67] See Lehner et al. (2014), p. 47.

^[68] See Lehner et al. (2014), p. 47.

^[69] See Bajohr et al. (2011). p. 207.

^[70] See Lehner et al. (2014), pp. 47-48.

^[71] See Bajohr et al. (2011), p. 207; Lehner et al. (2014), p. 48.

^[72] See Lefebvre et al. (2015), p. 84.

^[73] See Lehner et al. (2014), pp. 48-49.

^[74] See Sterner and Stadler (2014), pp. 350-351.

^[75] See Lehner et al. (2014), p. 50.

^[76] See Sterner and Stadler (2014), pp. 352-353.

^[77] See Lehner et al. (2014), pp. 49-50.

^[78] See Sterner and Stadler (2014), pp. 353-354.

^[79] See Lehner et al. (2014), p. 50.

^[80] See Sterner and Stadler (2014), p. 342.

^[81] See Lefebvre et al. (2015), p. 85.

^[82] See Sterner and Stadler (2014), p. 343.

^[83] See Bajohr et al. (2011), p. 207.

^[84] See Lehner et al. (2014), p.51; Sterner & Stadler (2014), p. 345.

^[85] See Sterner and Stadler (2014), pp. 344-345.

^[86] See Lehner et al. (2014), p. 50.

^[87] See Sterner and Stadler (2014), p. 351.

^[88] See Grond, Schulze, and Holstein (2013), p. 29.

^[89] See Grond et al. (2013), pp. 29-31.

^[90] See Graf et al. (2014a), pp. 41-42.

^[91] See Lehner et al. (2014), p. 59.

^[92] See Trost et al. (2012), p. 179.

^[93] See Sterner (2009), pp. 110-114.

^[94] See Trost et al. (2012), p. 182.

^[95] See Sterner (2009), p. 117.

^[96] See Trost et al. (2012), p. 179.

^[97] Interested readers shall be directed to Schiebahn et al. (2013), pp. 824-825; Trost et al. (2012), pp. 181-182.

^[98] See Trost et al. (2012), pp. 181-182.

^[99] See Sterner (2009), p. 120.

^[100] See Finkenrath (2012), p. 482.

^[101] See Trost et al. (2012), p. 184.

^[102] See Sterner (2009), p. 119.

^[103] See Trost et al. (2012), p. 182.

^[104] See Trost et al. (2012), pp. 182-183.

^[105] See Schiebahn et al. (2013), p. 825.

^[106] See Trost et al. (2012), p. 182.

^[107] See Finkenrath (2012), p. 485.

^[108] See Lehner et al. (2014), p. 76.

^[109] Source: own illustration based on the following literature (see row ‘ Reference ’)

^[110] See Sterner (2009), p. 110; Trost et al. (2012), p. 182; Schiebahn et al. (2013), p. 827.

^[111] See Schiebahn et al. (2013), p. 827; Finkenrath (2012), p. 487; Umweltbundesamt (2014).

^[112] See Schiebahn et al. (2013), p. 827; Trost et al. (2012), p. 183.

^[113] See Trost et al. (2012), p. 180.

^[114] See Stolzenburg et al. (2014), p. 73.

^[115] See Stolzenburg et al. (2014), p. 156; or Smolinka et al. (2011), p. 32.

^[116] See Stolzenburg et al. (2014), p. 63.

^[117] See Sterner and Stadler (2014), p. 415.

^[118] See Stolzenburg et al. (2014), p. 61.

^[119] Source: own illustration based on Stolzenburg et al. (2014), pp. 67-84.

^[120] See Müller-Syring et al. (2013), p. 132; or Lehner et al. (2014), p. 51.

^[121] See Lehner et al. (2014), p. 52.

^[122] See Müller-Syring et al. (2013), pp. 125-132.

^[123] See Lehner et al. (2014), p. 54.

^[124] See Müller-Syring et al. (2013), pp. 130-132.

^[125] See Sterner (2009), p. 111.

^[126] See Lehner et al. (2014), p. 59.

^[127] Source: own illustration based on Müller-Syring et al. (2013), p. 134.

^[128] See Sterner et al. (2011), p. 18.