**1. Introduction**

The systematic conversion of the global energy industry to a renewable basis poses a challenge to national economies and is referred to in Germany as the 'energy transition' (Energiewende). In addition to the supply of electricity from wind, solar, biomass, and hydro- and geothermal sources, an affordable and demand-oriented energy supply is essential for a sustainable and functioning economy. Therefore, there is a need for flexible, efficient, and scalable energy storage for hydrogen, a chemical storage medium, as a product of electrolysis [1–3]. To this end, eroded salt domes with large volumes are suitable for underground compressed gas storage. In the future, adapted or repurposed pipeline networks will be available to transport and distribute hydrogen from storage sites to various points of end use. There will be a focus on the direct use of hydrogen as a fuel for fuel cell-electric passenger cars, vans, buses, trucks, and trams. Additional downstream possibilities for hydrogen will be offered by the steel industry as a reducing agent in the

**Citation:** Emonts, B.; Müller, M.; Hehemann, M.; Janßen, H.; Keller, R.; Stähler, M.; Stähler, A.; Hagenmeyer, V.; Dittmeyer, R.; Pfeifer, P.; et al. A Holistic Consideration of Megawatt Electrolysis as a Key Component of Sector Coupling. *Energies* **2022**, *15*, 3656. https://doi.org/10.3390/ en15103656

Academic Editors: Bahman Shabani and Mahesh Suryawanshi

Received: 17 March 2022 Accepted: 8 May 2022 Published: 16 May 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

production of pig iron, by future fuel producers via synthesis with carbon dioxide from bionic or industrial point sources [4], or even captured from air (direct air capture) [5] to produce liquid fuels [6] equivalent to petroleum, diesel, and kerosene for trucks, aircraft, and ships. The demand-oriented reconversion of hydrogen in appropriate gas turbines into electricity will round out the portfolio of hydrogen usage. The results of analyses by Jülich scientists have demonstrated that the use of power-to-X (PtX) technologies and the use of hydrogen for transport and industrial applications will lead to significant hydrogen demand in the future. The CO2-free generation of electricity is an essential prerequisite to this. A H2 infrastructure encompassing generation, storage, and transport will be required for its implementation [7]. Due to its diverse range of applications, hydrogen plays an especially important role in scenarios that assume complete carbon neutrality. Thus, hydrogen demand in 2050 will amount to more than 12 million tons. Over 50% of this demand will be covered by domestic production and the rest will be supplied by imports. This level of demand means that water electrolysis will be a key technology for a future sustainable energy industry.

#### **2. An Overview of the Electrolyzer State of the Art**

Work on the material and technological improvement of water electrolysis began at Forschungszentrum Jülich in the early 1980s. Initially, the focus was on increasing the efficiency of alkaline electrolysis through new, innovative approaches to cell composition and structure. As a result of the high foreseeable demand for highly efficient electrolyzers, research work at Jülich was subsequently expanded to include PEM electrolysis technology.

The decomposition of water, or water-splitting, by means of electrolysis involves two partial reactions separated by an ion-conducting electrolyte. Three technically-relevant water electrolysis processes result from the species of ion being used:


At present, reversible high-temperature electrolysis is only being pursued to a limited extent due to the ceramic materials used in research and industry, and there are currently very few commercial products with a performance range relevant to the energy industry. An experiment involving high-temperature electrolysis in the kW range is being conducted using the network of facilities at Forschungszentrum Jülich, which are described in further detail below [8]. However, as this article focuses on MW application, the technology itself is not described any further.

Low-temperature electrolysis processes based on alkaline and PEM technologies can be used to establish large, powerful systems (1–100 MW) [9]. In contrast to PEM electrolysis, a number of alkaline electrolysis designs have been used for several decades at various scales. Their operating temperature is typically around 80 ◦C and systems of up to 750 Nm3 H2/h are available. These systems use an aqueous solution of potassium hydroxide (KOH) as an electrolyte with a typical concentration of 20–40% and achieve current densities of 0.25–0.45 A/cm2. Rectangular and circular electrodes and cells with an active area of up to approximately 3 m<sup>2</sup> have also been used. The plant engineering firm Lurgi manufactures pressurized electrolyzers that supply hydrogen and oxygen below 30 bar. Lurgi's pressurized electrolyzer produces 760 Nm3/h of hydrogen, which corresponds to an electrical output of approximately 3.6 MW [10]. Lifetimes of up to 90,000 h are achieved by the stack, with the electrodes and diaphragms needing to be replaced after this period. According to the NOW study by the Fraunhofer Institute for Solar Energy Systems (ISE, Freiburg), the voltage efficiency of the stack amounts to 62–82% in relation to the higher heating value (HHV) [11]. At less than 20%, the lower partial load operation is especially critical for its flexible application in combination with renewable energy sources. This is a result of the diaphragms that are used, which facilitate the mixture of hydrogen and oxygen due to diffusion, which in turn leads to safety-related switch offs. Another disadvantage of

alkaline electrolyzers as compared with PEM electrolysis is the costly gas treatment of the product gases, for which expensive noble metals must be used.

In contrast to alkaline water electrolysis, PEM electrolysis with proton-conducting membranes uses platinum group metals for the electrodes. Due to the use of dense membranes as an electrolyte and the possibility of integration with recombination catalysts, the systems can be operated at 0–100% power; however, in technical facilities the lower threshold is limited to approximately 5% of nominal power due to the internal consumption of peripheral components [10].

In particular, the high overvoltage at the oxygen electrode is a challenge for the development of materials and is partially responsible for the energy loss of the systems. Studies have shown that RuO2 and IrO2 are especially suitable catalysts for oxygen electrodes [12–14]. These metal oxides exhibit a high level of activity, adequate long-term stability, and low performance losses [15–24]. For this reason, IrO2 is often used as an anode catalyst for PEM electrolyzers [25]. Despite its low specific activity compared to RuO2, IrO2 is especially suitable due to its low overvoltages [22,26] and excellent electrochemical stability compared to RuO2 [22].

In current commercial systems, approximately 2 mg/cm2 iridium is required for the anode and approximately 1 mg/cm2 of platinum is required for the cathode. At the given operating conditions, these systems run at voltages of about 2 V, current densities of approximately 2 A/cm2, and operating pressures of up to 30 bar [27]. This is equivalent to the voltage efficiency of alkaline electrolysis (approximately 67–82%), but with much higher current densities (0.6–2.0 A/cm2) [28].

Other studies have investigated the level of iridium loading that is advantageous for function and stability, and identified 1–2 mg/cm2 as the ideal range [29]. However, other studies have investigated much lower iridium loading and found that this also results in reasonable operating times [30]. When using noble metals which are limited in terms of availability, this automatically raises the question as to whether they will be available in sufficient quantities. In [31], it was investigated whether the extraction of iridium could sufficiently cover demand for the increasing expansion of PEM electrolysis. It was found that noble metal loading would need to be significantly reduced in the next 15 years to ensure a sufficient supply of noble metals. The target loading should reach 50 mg/kW in 2035. Based on a power density of 6 W/cm2, this corresponds to a catalyst loading of 0.3 mg/cm2. However, it was found that recycling of the catalyst is an essential prerequisite for this. Carmo et al. describe how this can be implemented in a simple and environmentally-friendly manner, despite the use of solvents [32].

All large PEM electrolysis manufacturers are working on the development of MW systems with various stack concepts. Hydrogenics, for instance, recently developed and built a 1 MW PEM electrolyzer with a single stack and a nominal power of 1 MW [33]. For the most part, the commercially-available stacks only operate with current densities < 2.0 A/cm2. In these systems, approximately 6 mg/cm<sup>2</sup> of iridium or ruthenium is required for the anode, and approximately 2 mg/cm<sup>2</sup> of platinum is required for the cathode [34]. In contrast to alkaline electrolysis, the lifetime of PEM electrolysis stacks is estimated at <20,000 h. However, Proton Onsite has already achieved a lifetime of more than 50,000 h for stacks used in the PEM electrolyzers of the HOGEN C series [35].

In terms of thermodynamic considerations, the hydrogen-producing electrode should ideally be operated under pressure [36]; the oxygen-producing electrode can be operated under atmospheric conditions, as the oxygen being produced is not typically stored under pressure. Tjarks et al. [37] showed that an electrolyzer operating pressure of up to 20 bar can improve the system's efficiency.

One of the technical limitations of this differential pressure operation is the mechanical stability of the membrane electrode assembly (MEA) and the sintered body. In addition, the design of modern stacks must account for the fact that in the future, thinner membranes will be used that significantly increase the hydrogen production of an electrolyzer. The influence of membrane thickness on cell performance has been simulated many times and analyzed in the literature [36,38–40]. Stähler et al. [41] were able to achieve current densities of 11 A/cm<sup>2</sup> at 2 V in a single cell with a membrane thickness of approximately 20 μm. If the membrane thickness is reduced, H2 permeation through the membrane and the mechanical stress of the membrane, particularly during pressure operation, must be taken into account [36,39].

Another important aspect is the adjustment of compression and contact pressure for the porous transport layers (PTLs) being used. Stähler et al. [42] showed that by increasing compression, the performance of the electrolysis cell can be enhanced, although excess compression of the gas diffusion layers (GDLs) leads to an increase in hydrogen permeation through the membrane. When designing the stack, it is important to avoid increased compressive and shear stresses [43]. Borgardt et al. conducted analyses on membrane and stack component mechanics in terms of their relation to contact pressure, efficiency, and creep behavior [44,45].

In addition to the noble metals required for the electrodes, another challenge of system development is posed by titanium-based separator plates and current collectors, to which coatings must be applied due to hydrogen embrittlement, the formation of oxide layers, and the associated increase in contact resistance. According to publications by Ayers et al. (Proton OnSite), the separator plates (including the current collectors) represent around 48% of the stack's costs [46].

On the anode side, porous transport layers made of titanium are typically used to enable the transport of water to the electrode and electrical contact with the flow field. Liu et al. [47] showed that an iridium coating is important for ensuring long-term stability and low contact resistance of PTLs. A crucial means of limiting costs is the increasing integration of stainless steel materials instead of titanium [48], which requires a coating without any defects [49]. In order to reduce costs, the use of grade 316 stainless steel is being investigated with various coatings [50,51]. A potentially inexpensive and stable coating is comprised of Ti/TiN, and was investigated in greater detail by Rojas et al. [52].

In the long term, PEM electrolysis can play a significant role in providing operating reserve, as its better dynamics in relation to alkaline electrolysis make it interesting for larger applications with systems > 1 MW. However, the electrode or cell area must be scaled up to 600–2000 cm2 in order to reduce the high level of investment costs (>2000 €/kW [10]). This is confirmed in the Plan-DelyKaD study, which demonstrated that PEM electrolysis has a moderate cost advantage over alkaline electrolysis in terms of investment costs [53]. As part of a study by Bertuccioli et al. [48], the manufacturers of such systems were asked about anticipated reductions in costs. The study came to the conclusion that in the long term, similar costs to alkaline electrolysis are expected. A current list of the most significant manufacturers of PEM electrolysis systems is given in Table 1. PEM electrolysis is currently in a phase of development in which the absolute power of the systems is to be further increased. The most powerful systems are currently offered by Siemens and ITM.


**Table 1.** Commercially-available electrolysis systems.


**Table 1.** *Cont.*

In a study that focused on optimizing the technology and reducing the costs of electrolyzers, the factors influencing the levelized cost of hydrogen (electricity costs, operating costs, and control and investment costs) were investigated for the various electrolysis technologies [64]. The study showed that for optimized PEM electrolyzers in the MW range in the year 2030, investment costs of € 585/kW (roughly four times lower than the current costs) can be expected. To this end, PEM electrolysis will need to be further scaled up to ensure that it is economically viable in a higher MW range. A reduction in noble metal loading in the electrodes can help reduce costs, whereas additional stack components such as bipolar plates and sintered bodies also offer further potential for reducing costs.

In addition to systems development work, the integration of technology is vitally important. From a technological standpoint, it is crucial for the required system dynamics; from an economic perspective, it is important for evaluating value creation. Electrolyzers can be used to serve the various operating reserve requirements of the energy market [65]. In a study conducted at Energiepark Mainz, the operation of an electrolyzer for an energy system participating in the energy market was presented as an example. The associated pressure operation requirements of an electrolyzer were also shown [66]. There are numerous other possible approaches to producing hydrogen on a large scale under as realistic conditions as possible [67–69]. However, all of these projects share an emphasis on the use of commercially-available systems and the scaling of performance; the focus is not on making a leap in technological development or the possibility of realizing new approaches in a dimension that enables all energy and mass flows to be evaluated.

With the development described in the following sections, we therefore address this vital link in terms of energy supply and aim to show how a 400 kW next-generation electrolysis stack can be integrated into a highly dynamic, intelligent energy system of the future and what results can be expected. We have a unique infrastructure at our disposal that, for the first time, enables validation of the systems that have been almost exclusively modeled to date.

#### **3. Research and Experimental Work for Next-Generation PEM Electrolysis**

Electrolysis research generally focuses on understanding the mechanisms that are essential to improving the components required for water electrolysis. Starting with the catalyst material and emphasizing suitable support and distribution structures, the various approaches and specialized methods involve the research, development, production, and assembly of the electrolyte, MEA, bipolar plates, media, and power supplies at scales ranging from micrometers to meters. The following sections will describe this work in further detail and focus on the development of a 400 kW electrolysis stack.

The focus of this investigation is the electrolysis stack, which was planned, designed, produced, and assembled within the scope of electrolysis research and is described in detail in the following sections. The MW electrolysis system is an instrument of research infrastructure that enables the industrial-scale use of electrolysis for technological research and development. The close-to-realistic conditions, in combination with the testing facilities, were aimed at fundamental and application-oriented areas for improving electrolysis technology and its integration into a sustainable energy supply pathway:


In the following sections, three fields of work will be presented in which the MW electrolysis system is the key component for various research and development tasks (see Figure 1). Electrolysis research employs innovative approaches to tackle fundamental issues with the aim of improving cell components for PEM water electrolysis and developing efficient, long-lasting, and economical electrolysis stacks. The Energy Lab 2.0 and Living Lab Energy Campus (LLEC) projects are focused on integrating PEM water electrolysis into application-oriented energy pathways for methanation, use in a micro gas turbine, and the synthesis of liquid electrofuels, as well as temporary storage as a compressed gas or in the form of liquid organic hydrogen carriers (LOHCs). For the various uses of stored hydrogen, the LLEC intends to utilize an alkaline fuel cell in which the oxygen from water electrolysis can also be used to generate electricity. In addition, hydrogen can be added to natural gas for combined heat and power internal combustion engines.

**Figure 1.** The MW electrolysis system as the focus of research.
