Producing Hydrogen Through Electrolysis and Other Processes

Producing hydrogen through electrolysis
and other processes 9

Sebastian Metz Tom Smolinka
Fraunhofer Institute for Solar Energy Systems ISE
Christian I. Bernäcker Stefan Loos Thomas Rauscher
Lars Röntzsch
Fraunhofer Institute for Manufacturing Technology and
Advanced Materials IFAM, Dresden institute branch
Michael Arnold Arno L. Görne Matthias Jahn
Mihails Kusnezoff
Fraunhofer Institute for Ceramic Technologies and Systems
IKTS
Gunther Kolb
Fraunhofer Institute for Microengineering and Microsystems
IMM
Ulf-Peter Apfel Christian Doetsch
Fraunhofer Institute for Environmental, Safety, and Energy
Technology UMSICHT
Abstract
In addition to expanding renewable energy sources and implementing electrical

energy storage systems, hydrogen will form the third building block of the en-
ergy transition. Hydrogen and hydrogen-based synthetic fuels are particularly
suitable for use in industrial processes, such as in the steel industry and the
chemical industry, as well as in long-distance mobility, heavy goods transport
and aviation, where battery storage meets its technological limits. Hydrogen ap-
plications will also become increasingly significant in reconversion processes,
© Springer Nature Switzerland AG 2022 203

R. Neugebauer (Ed.), Hydrogen Technologies,
https://doi.org/10.1007/978-3-031-22100-2_9
204 9 Producing hydrogen through electrolysis and other processes
as well as in the heating sector to a certain degree. It will only be possible to

couple hydrogen with regenerative energy sources (wind power, photovoltaics)
and integrate the energy system across sectors if hydrogen is produced by elec-
trolysis.
Below, the various technologies for hydrogen production that are either al-
ready available on the market or currently being launched on the market are pre-
sented and described in terms of their advantages and disadvantages. Whereas
alkaline electrolysis has been an established practice on the market for decades,
PEM electrolysis is still in the phase of being scaled-up in order to reduce
costs. Other promising technologies that are still under development or in the
demonstration phase include high-temperature electrolysis and alkaline mem-
brane electrolysis. These “green” processes for hydrogen production will be
compared to both existing processes (gray hydrogen: steam methane reform-
ing) and new processes based on fossil fuels (blue/turquoise hydrogen). In the
outlook, photo catalytic and biological processes are presented.
9.1 Hydrogen production processes
Today, the majority of hydrogen is produced by the steam reforming of natural gas,
a process that releases large amounts of carbon dioxide (CO2 ). The production of
one ton of hydrogen produces around ten tons of CO2 . Endothermic reforming
also requires energy, so additional natural gas has to be combusted to keep the
reaction going. A large proportion of “gray hydrogen” is required for processing
crude oil in refineries, and consequently, for mass producing fuels and lubricants.
It is also required for synthesizing ammonia (which is ultimately used to produce
fertilizers) and producing chemical raw materials such as methanol, higher alco-
hols and amines (Fig. 9.1). In addition, it is used as part of the Fischer-Tropsch
synthesis to produce synthetic fuels. Aside from natural gas, coal is also used as
a fossil fuel. Hydrogen, also known as black hydrogen, is produced in a hydrother-
mal gasification process using steam. The term blue hydrogen is used when the
CO2 released during steam reforming processes is separated and stored in subsoil.
This is known as carbon capture and storage (CCS). This way, the CO2 is not re-
leased into the atmosphere. However, the challenge is to store it safely in the long
term. Turquoise hydrogen is also produced from natural gas, with methane being
broken down into its components by the supply of thermal or electrical energy to
produce solid carbon and gaseous hydrogen. Therefore, the process itself does not
result in the release of CO2 , provided the endothermic process is operated with
9.1 Hydrogen production processes 205
Fig. 9.1 Hydrogen production and applications worldwide. (own diagram based on IEA)
CO2 -free energy. However, as with all production processes based on natural gas,
it is necessary to take into account the long-term sequestration of the carbon, as
well as emissions that emerge upstream in the supply chain and cannot be avoided
completely.
In both steam reforming processes and methane pyrolysis, it is also possible
to use biogas instead of natural gas. In terms of sustainability, there are particular
benefits to using biogenic residues. However, the entire process chain and the as-
sociated emissions must be considered in light of the materials used, for example
refuse materials or residue materials. One already-established method of produc-
ing hydrogen using electrical energy is chloralkali electrolysis, whereby hydrogen
is produced as a by-product. However, the share of total hydrogen production is
small.
In the long term, green hydrogen and its synthesis products will have to play the
most important role in building a sustainable hydrogen economy. The majority of
green hydrogen is produced via water electrolysis, a process that only uses electric-
ity from renewable energies and purified water (Fig. 9.2). This makes producing
hydrogen a largely emission-free process. However, there are currently a variety
of definitions for green hydrogen. For the German Federal Ministry for Economic
Affairs and Climate Action (BMWK) and the Federal Ministry of Education and
Research (BMBF), hydrogen is considered “green” if it is produced via electrol-
ysis using renewable energies. On the other hand, testing organizations such as
Reforming natural gas with steam

Gray - Associated with high CO2 emissions
- Production costs: approx. €1.6/kg,
Reducing greenhouse gas

Reforming natural gas with steam, coupled with
CCS
Blue
- Requires permanent storage of CO2
Thermal breakdown ofmethane (pyrolysis) with

renewable energies
Turquoi - Requires only the storage of solid carbon
Water electrolysis with electricity from

renewable energies
Green - Electricity prices heavily influence
Fig. 9.2 Comparison of different methods of producing hydrogen in terms of current deficits
and production costs. (own diagram)
the TÜV or the CertifHy European Guarantee of Origin have broader definitions
when it comes to possible production processes and what renewable energies can
be used. It is expected that these varying definitions will be coordinated in the near
future, at least at a European level.
As it stands, green hydrogen is on average twice as expensive as blue hydrogen,
and about three times more expensive than gray hydrogen. Due to the increase in
CO2 charges and the potential limits of low-cost CO2 storage facilities, it will be
difficult to further reduce the cost of gray and blue hydrogen in the future. However,
the cost of providing green hydrogen may nevertheless decrease the future if, first,
there is an improvement in the efficiency and long-term stability of the electrolysis
processes used, second, suitable regulatory framework conditions are established
and, third, a hydrogen infrastructure is constructed in line with requirements.
Various electrolysis processes can be used to produce green hydrogen. These
differ mainly in terms of operating temperature and their particular stage of de-
velopment. In order to select the appropriate technology, both environmental con-
ditions and operating conditions must be taken into account. This will be further
explained in Sect. 9.2. Further possibilities for producing green hydrogen can be
9.2 Hydrogen production by electrolysis 207
found in biotechnological processes and strategies for obtaining solar hydrogen

through photocatalysis. Currently, the process of splitting water via photocataly-
sis is being tested on a laboratory scale. It promises future advantages in terms of
cost thanks to its low system complexity and the tried-and-tested use of large-scale
technologies from the photovoltaic industry. These electrolysis procedures, as well
as other innovative procedures, will be discussed in more detail in the following
sections.
9.2 Hydrogen production by electrolysis
9.2.1 Fundamentals of water electrolysis
Introduction
In general, electrochemically splitting water into its components, hydrogen and
oxygen, by means of electrical energy is referred to as water electrolysis. The pro-
cess of splitting water is endothermic. The following equation describes the basic
reaction:
1
H2 O H2 C O2 HR D 286 kJ=mol (9.1)
2
This is the opposite direction to a reaction in a fuel cell, whereby hydrogen and
oxygen react to form water while producing electrical energy. The water can be
supplied to the process either in liquid form or as steam; oxygen and hydrogen are
produced as gases.
At present, the technically relevant processes are alkaline electrolysis with a liq-
uid basic electrolyte (AEL), acidic electrolysis with a solid polymer electrolyte
(PEMEL) and high-temperature electrolysis with a solid oxide electrolyzer cell
(SOEL). However, there are other electrochemical processes that can be used for
water splitting which are currently under development (Fig. 9.3). These include
alkaline membrane electrolysis, which uses an alkaline electrolyte membrane that
is conductive to OH ions (AEMEL); proton conducting ceramic electrolysis (PC-
CEL); and co-electrolysis, which is also based on solid oxide cells (CoSOEL) and
produces synthesis gas (CO C H2 ) by directly reducing CO2 and splitting water
in a cell. The reverse water gas shift reaction (rWGS), which produces CO, runs
in parallel. The Hydrogen Evolution Reaction (HER) at the cathode and Oxygen
Evolution Reaction (OER) at the anode vary according to the electrolyte used and
are summarized in Table 9.1.
Fig. 9.3 Schematic representation of various water electrolysis procedures. Links: Low-
temperature electrolysis. Right: High-temperature electrolysis. (Fraunhofer ISE)
Table 9.1 Half-cell reactions, typical temperature ranges and charge carriers of the three
main types of water electrolysis
Procedure Tempera- Cathode reaction Charge Anode reaction
ture carrier
AEL 70–90 °C 2H2 O C 2e ! H2 C 2OH OH 2OH ! 12 O2 C H2 O C 2e
AEMEL 50–70 °C 2H2 O C 2e ! H2 C 2OH OH 2OH ! 12 O2 C H2 O C 2e
PEMEL 50–80 °C 2H+ C 2e ! H2 H+ H2 O ! 12 O2 C 2H+ C 2e
PCCEL 450–600 °C 2H+ C 2e ! H2 H+ H2 O ! 12 O2 C 2H+ C 2e
SOEL 650–850 °C H2 O C 2e ! H2 C O2 O2 O2 ! 12 O2 C 2e
CoSOEL 700–900 °C H2 O C 2e ! H2 C O2 O2 O2 ! 12 O2 C 2e

CO2 C 2e ! CO C O2
CO2 C H2 CO C H2 O
One important factor that distinguishes the various methods is the choice of
electrolyte. This determines the type of charge carrier (H+ , OH or O2 ), thus in-
directly influencing the operating temperature of the cell. This is because ensuring
that the ionic conductivity of the electrolytes is high enough requires establishing
a minimum temperature. Determining temperature and pH value also determines
what catalyst material should be used, since electrodes must have a sufficiently
high level of electrochemical activity in this operating window to develop hydro-
gen and oxygen and must also remain stable for a long time. The types of materials
and cell constructions, as well as other aspects of the system used for the individual
technologies, will be discussed in the following sections.
Under standard conditions (298.15 K and 101.325 kPa), the energy required to
split the water according to Eq. 9.1 is HR0 = 285.8 kJ/mol, thus corresponding to
the heating value of hydrogen. According to the Gibbs-Helmholtz equation, the

reaction enthalpy consists of two components:
HR D GR C TSR (9.2)
Free enthalpy of reaction (Gibbs free energy) GR describes the minimum re-
quired amount of energy that must be supplied in the reaction in the form of
electrical energy. It corresponds to the heating value of hydrogen under standard
conditions and has a value of GR0 = 237.2 kJ/mol. The product of the temperature T
and enthalpy of the reaction SR from Eq. 9.2 is the portion of enthalpy of the re-
action that can also be supplied to the reaction as thermal energy. However, unlike
high-temperature methods (Fig. 9.3), low-temperature electrolysis (alkaline elec-
trolysis and PEM electrolysis) cannot absorb (or can barely absorb) heat from the
environment. Instead, the missing heat energy must similarly be introduced to the
process in the form of electrical energy.
The above values apply to standard conditions at 298.15 K but vary according to
the process temperature. Fig. 9.4 is a diagram showing the temperature dependence
of the thermodynamic variables. The discontinuity at 100 °C (373.15 K) is due to
the phase transition of the water from liquid to steam. While the enthalpy reac-
tion HR at T > 373.15 K (100 °C) is largely temperature-independent, the Gibbs
free energy GR and the portion of entropy TSR show considerable variation
according to temperature, resulting from the temperature dependence of the heat
capacities of the substances involved. As the temperature increases, the Gibbs free
energy GR decreases; at the same time, the product of temperature and entropy
TSR increases. As the operating temperature increases, the minimum propor-
tion of the enthalpy of reaction that has to be supplied in the form of electrical
energy for the decomposition reaction decreases. This is the main advantage of
high-temperature electrolysis, provided that intense heat can be used in the pro-
cess.
Given that electrolysis is an electrochemical process, the Gibbs free energy can
be used to calculate the necessary voltages required for operating the electrolysis
cells. The reversible cell voltage is calculated from the free reaction enthalpy GR0
0
to Vrev = 1.23 V and is equal to the cell voltage of a fuel cell operated under ideal
conditions. However, when it comes to low-temperature electrolysis, no heat can
be fed in, meaning that the minimum energy required for splitting water under
standard conditions is equal to the reaction enthalpy HR0 , which can be used
to calculate the minimum (thermoneutral) decomposition voltage Vth0 = 1.48 V. At
temperatures of approx. 800 °C, the electricity required GR is reduced by approx.
Fig. 9.4 Correlation between the thermodynamic condition variables HR and GR and
the temperature for the decomposition of liquid and gaseous water at 101.325 kPa. (Fraun-
hofer ISE)
25 percent, with the theoretical decomposition voltage V rev dropping to values of

around 1.0 V.
However, during the actual operation of an electrolysis cell, these ideal cell
voltages cannot be achieved—here, it must be taken into account that the voltage
is dependent on the concentration, as described by the Nernst equation:
GR .p; T; H2 Ox/ RT xH2 O

VNernst;H2 D ln 1=2
(9.3)
2F 2F xH2 xO2
In addition, additional loss mechanisms occur when a current flow is applied,

which in actual operation leads to an increase in the operating voltage, and thus
to a reduction in efficiency:
Ohmic losses: The electron flow and the ion flow run in opposite directions
in an electrolysis cell. They are mainly caused by the ionic resistance of the
electrolyte, the internal electrical resistance of the electrodes and by contact
resistance.
Kinetic losses: These occur at the anode and cathode respectively, due to the
speed-limited transition of the electrons to the boundary surface between the
electrode and the electrolyte. These losses create overvoltages at the electrode,
which work in the opposite direction of the reaction. The overvoltages on the
hydrogen side (cathode) are significantly lower than the overvoltages on the
oxygen side (anode).
Losses due to mass transport limitations: These losses are caused by obsta-
cles to transporting gases and liquids to and from the electrode. Less conversion
takes place at the electrode given the insufficient supply of the electrochem-
ical reaction to the reactants. In order to uphold the conversion process, it is
necessary to increase the overvoltage as a driving force at the electrode. A well-
known example of this mechanism can be found in the overvoltage caused by
gas bubbles in gas-producing electrodes in liquid electrolytes, which are also
exploited in alkaline water electrolysis.
Although the effects of the unavoidable losses that occur when a current flow is
applied to the cell can vary wildly, they nevertheless lead to an increase in the ac-
tual cell voltage compared to the ideal cell voltage in all electrolysis technologies.
As a result, more energy has to be supplied as current density increases and the ef-
ficiency of the electrolysis cell is reduced. Since the rate of hydrogen production is
proportional to the current supplied (Faraday’s law), developments the area of new
materials and components aim to achieve the highest possible current densities,
while keeping cell voltages low. In other words, it is necessary to find a compro-
mise between high efficiency (low operating costs) and high current density (low
acquisition costs).
The losses outlined above appear to different degrees in the various electroly-
sis technologies. Fig. 9.5 shows typical voltage-current density characteristics for
the three most important technologies: AEL, PEMEL and SOEL. The figure also
estimates future development up to around 2030.
Alkaline electrolysis uses nickel-based electrodes. It is carried out at tempera-
tures of around 80 °C and current densities between 0.2 and 0.6 A/cm2 , while cell
voltages are below 1.9 V. Recent developments with more complex electrodes are
also suitable for current densities up to 1.0 A/cm2 at the same cell voltages. In
the future, current levels of more than 1.0 A/cm2 should also be possible at cell
voltages of around 1.8 V.
PEM electrolysis can achieve high current densities of approx. 2.0 A/cm2 at
approx. 60 °C thanks to its electrodes, which contain precious metal, and its very
compact design. The cell voltages are between 1.8 and 1.9 V, making them compa-
rable to those found in alkaline electrolysis. The current density will be increased to
Fig. 9.5 Comparison of the different voltage-current characteristics for the three essential
procedures, AEL, PEMEL and SOEL, and an estimation of typical operating points for the
present day and 2030. (Fraunhofer ISE)
well over 3.0 A/cm2 by 2030, and the cell voltage will be reduced to approx. 1.7 V.
High-temperature electrolysis, a comparatively new technology, also offers great
potential in terms of development. Due to its high operating temperatures, the cell
voltage is typically only 1.3 V. This will not change significantly in the next few
years, but the current density will be increased considerably.
A significant improvement in the life span of the cell stack is also expected as
the technological maturity of the cells increases. Based on a literature study on
the service life forecasts in Herz et al. [1], Fig. 9.6 shows an adapted representa-
tion for the three technologies AEL, PEMEL and SOEL. Although there are still
significant differences in service life, over the coming decades, it is expected that
individual cells or complete stacks will not have to be replaced in any of the three
processes until over 80,000 hours of operation have passed. However, when com-
ing up with a strategy for use, it should be considered that the mode of operation
of the electrolysis system can have a considerable influence on the service life of
the cells.
A brief history
The principle of electrochemically decomposing water in an electrolysis cell has
been known for more than 230 years. The first electrochemical production of
hydrogen by means of electricity was carried out as early as in 1789 by van
Troostwijk and Deiman, using an electrostatic generator as a DC power source [2].
Fig. 9.6 Service life forecasts for the various electrolysis technologies, adapted according
to Herz et al. (Fraunhofer IKTS according to [1])
Shortly after Volta had developed the voltaic pile in 1800, Carlisle and Nicholson
used such a device to break water down into hydrogen and oxygen [3]. In the same
year, Ritter carried out similar experiments in Jena. In addition, at the beginning
of the 19th century, Cruickshank used a photovoltaic battery for the electrochem-
ical decomposition of NaCl solutions into hydrogen and chlorine. Nevertheless,
it took decades for these procedures to be used in technical applications. Around
1890, Charles Renard constructed a water electrolysis system for the production
of hydrogen for French military airships. The world’s first electrolyzer to be built
with cell stack (filter press) architecture was patented in 1899 by Oscar Schmidt
from the company Oerlikon (R.P. 111131) and presented to the general assembly
of the German Society of Electrochemistry in Zurich in August 1900 [4]. It is
estimated that around the year 1900, more than 400 industrial alkaline water elec-
trolyzers were in operation worldwide [5]. In addition, the large-scale technical
application of the chloralkali process began, pioneered by the Griesheim-Elektron
company in Bitterfeld, at the largest site of its kind at the time. Later, in the first
half of the 20th century, various types of commercial alkaline water electrolyzers
were developed in order to produce the hydrogen needed for producing ammo-
nia fertilizer by making use of low-cost hydroelectric power. In Trail in Canada,
Rjukan and Glomfjord in Norway, at the Aswan Dam in Egypt and elsewhere,
large-scale plants with atmospheric electrolyzers and connection capacities of
over 100 MW were built [6]. The production of heavy water also contributed to the
commercialization of water electrolysis during this period. In the second half of
the 20th century, the more cost-effective method of producing hydrogen through
steam methane reforming increasingly replaced water electrolysis, and toward the
end of the 20th century, the process was only used in niche applications.
Work on PEM electrolysis began in the 1960s with NASA’s Project Gemini.
General Electric then made the decisive breakthrough in the early 1970s [7] by
using DuPont’s Nafion® membrane, which Walter G. Groth had developed some
years before [8]. In the first twenty years, due to the high material costs, devel-
opment work was almost exclusively focused on laboratory, military and space
applications, although General Electric also developed concepts for large-scale
use [9]. BBC then took its first steps to open up new markets with the 100 kW
MEMBREL PEM system in the 1980s [10]. Also in the late 1960s, General Elec-
tric and the Brookhaven National Laboratory began developing a high-temperature
electrolysis system with solid oxide cells [11]. In Germany, Dornier pursued the
development of tubular HTEL cells between 1975 and 1987 as part of the Federal
Ministry of Education and Research (BMBF) project HOT ELLY (High Operating
Temperature ELectroLYsis) [12]. However, despite all these technical advances,
these processes did not become widely commercially established, as they were not
able to compete with the advantages offered by steam reforming. Water electrol-
ysis has been attracting attention again since the mid-1980s, when hydrogen was
used as a green energy carrier in conjunction with renewable energy sources such
as wind and solar energy. The coupling of water electrolysis with renewable en-
ergies was successfully demonstrated in projects such as the DLR’s HySolar or
Solar Wasserstoff Bayern (Solar Hydrogen Bavaria) in Neunburg vorm Wald [13].
However, it is only in the last ten years that the worldwide interest in water elec-
trolysis has increased significantly with the adoption of ambitious national climate
protection programs. It is now regarded as a key technology for sector coupling.
9.2.2 Alkaline water electrolysis
In addition to the peripheral systems (gas drying, compressors, pumps, rectifiers,

etc.), an alkaline electrolyzer consists mainly of a stack of several electrolysis cells,
in which water is separated into H2 (cathode) and O2 (anode) at the two electrodes.
The structure of a single alkaline electrolysis cell in principle is shown in Fig. 9.7.
The cathode and anode chambers (known as half-cells) are separated by a gas-
impermeable membrane or a diaphragm. According to the general state of the art,
Fig. 9.7 Basic structure of an alkaline electrolysis cell. (Fraunhofer ISE)
an aqueous solution, usually of 30% KOH, is used as the electrolyte at typical

process temperatures between 70 and 90 °C.
The nature of the electrodes depends on two main factors: first, low overvoltage
and second, a lengthy service life of the electrocatalyst material and the electrode
structure. The overvoltage ˜ on an electrode is defined as the difference between
the reversible potential (thermodynamic ideal) for the corresponding half-cell reac-
tion (Table 9.1) and the potential that actually has to be applied for the production
of H2 or O2 . The industry-standard cell voltage values in alkaline electrolyzers
range from 1.65 to 2.0 V, which corresponds to a specific energy consumption of
4.0 to 4.8 kWh per standard cubic meter of H2 produced (electrolysis efficiency of
73 to 88 percent relative to the heating value (HHV) of H2 ). In general, the elec-
trode overvoltage rises as current density increases, meaning that current densities
of 0.2 to 0.6 A/cm2 are typically used on a commercial scale.
Platinum results in a very low overvoltage, making it the most suitable catalyst
material for the cathode, which is where H2 is produced (hydrogen evolution re-
action, HER). However, like other precious metals, it is not economically viable
due to the high material prices. Therefore, several cheaper materials, which also
have high HER activity (low overvoltage), were developed. Raney Ni and Ni-Mo
compounds have shown the best properties in this regard [14–23]. The electrode
catalyst materials most widely used today are layers of Raney Ni, which are ap-
plied to metallic carrier plates. Compared with smooth surfaces, Raney layers have
a significantly higher effective surface area for the reaction and an increased struc-
tural defect density, making it possible to achieve a low HER overvoltage with the
same effective current density. On the anode side, where the O2 formation reaction
(OER) occurs, developments have mainly focused on nickel compounds [24, 25].
Raney Ni and Ni-X compounds (X = Co, Fe) play a leading role in this process
and their activity can be further increased through the targeted addition of other
transition metals [18, 26–28].
In addition to the electrochemical and mechanical properties of the electrode
material, the way that the electrode is integrated into the individual cell is also
crucial. The reaction chambers for HER and OER are separated by a gas-tight
diaphragm (or membrane) to avoid cross-contamination of the gases. There are
numerous common cell architectures which have different arrangements and dis-
tances between the individual components. For example, zero gap arrangements
[24] are used, whereby the electrodes are directly pressed onto the diaphragm to
reduce the voltage drops within the electrolysis cell by lowering the ohmic resis-
tance of the electrolyte solution. In the classic design, however, a distance of a few
millimeters is left between the electrodes and the diaphragm (or membrane) [25].
The advantages of this architecture lie in the simple and robust construction. The
removal of the gas bubbles has always been problematic in all previous designs,
however, because the resulting bubbles significantly increase the cell voltage. This
is because they both temporarily block the active electrode surface and increase
the ohmic resistance of the electrolyte solution. This is why the electrodes in con-
ventional AEL cells usually consist of perforated sheets with as rough a surface as
possible, which are positioned as “pre-electrodes” close to the diaphragm. By per-
forating the pre-electrodes, the gas bubbles, which are formed on the side facing
the diaphragm, can be discharged into the space between the pre-electrode and the
end plate. However, the fact that a significant part of the surface area cannot be used
due to the perforation (up to 30 percent of the pre-electrode area) is a considerable
disadvantage. This in turn limits the space-time yield of the entire electrolysis cell.
Recent electrode developments are based on porous metallic carrier structures
(porosity of 60 to 90 percent) with electrocatalytically active layers deposited on
the surface. The production of such porous, current-carrying 3D substrate materials
with an electrocatalytically active alloying system and the associated investigation
of the structure-property relationship has been a core competence of Fraunhofer
IFAM for several years and goes hand in hand with the development of technolo-
gies for production on an industrial scale [29, 30]. Metallic foams, fiber struc-
tures, webbing and mesh are used as substrate materials. These are generally made
a b
d
c
Fig. 9.8 Examples of macroporous metallic substrate materials for live pre-electrodes:
(a) metallic foams, (b) metallic fiber structures, (c) porous metal films, (d) additively
manufactured metal structures, and (e) metallic nets and mesh which can also be used three-
dimensionally as a layered stack (optionally even with a porosity gradient). (Fraunhofer
IFAM)
of nickel and are also widely available in sizes of one or more m2 . Moreover,
they have a good corrosion resistance in the KOH solution at higher temperatures
(Fig. 9.8).
In general, the efficiency of the electrolysis process decreases as current density
increases. Alongside the reaction kinetics (Butler-Volmer behavior), the increas-
ing ohmic resistance of the cells should be considered the main factor for the
increase in the cell voltage. The resistance is caused by gas bubbles in the elec-
trolyte between the electrode and the separator. Since the gases produced lead to
a significant increase in electrolyte resistance, it is possible to noticeably improve
how effectively these gas bubbles are handled by employing three-dimensional
cellular metallic structures. Due to the porous three-dimensional structure of the
electrode, gas bubbles flow from the entire surface, i.e., along the electrolyte space
between the electrodes and the separator (Fig. 9.9). As a result, the overall cell re-
sistance falls and the efficiency of the process increases. These porous electrodes
are particularly suitable for the zero-gap arrangement, which can therefore work at
lower cell voltages or can work at higher current densities, for example.
Fig. 9.9 Comparison of classical and zero gap cell architecture in alkaline electrolysis.
(Fraunhofer IFAM)
The coating of 3D substrate structures with electrocatalytically active layers

can be carried out via electroplating, sintering or plasma processes. The powder
metallurgical method has been shown to be particularly favorable as a continuous
production process. Fig. 9.10 shows an example of a Raney Ni coating applied
to a nickel substrate via a powder metallurgical process. The following process
steps are required for production: First, the Ni foams are coated with aluminum
in a powder metallurgy process route. Subsequently, a heat treatment is carried
out to produce Ni-Al phases close to the surface, from which aluminum is then
selectively extracted, resulting in skeletal nickel, known as Raney nickel.
Fig. 9.10 Schematic representation of the manufacturing steps of the Raney Ni coating of
electrode substrate materials. (Fraunhofer IFAM)
a After heat treatment b After leaching
Fig. 9.11 Cross-sections of the Ni foam (450 µm) (a) following heat treatment and (b) fol-
lowing leaching (I: leached Ni2 Al3 phase, II: leached NiAl3 phase). Inserted images: view
from above. (Fraunhofer IFAM)
Fig. 9.11 shows cross-sectional views of the Ni foams following heat treatment
and leaching. After the heat treatment, a homogeneous layer can be seen on the Ni
foam joints. After the leaching of the Al-rich phases, a porous layer (Ni skeleton)
with channels arranged perpendicular to the foam joints, resulting in a significant
increase in surface area compared with uncoated foam (increased by a factor of
1000 to 10,000). It is also evident that the channel-like layers consist of several
Ni-Al phases. As shown in the Ni-Al phase diagram, these are Ni2 Al3 or NiAl3 .
The effect of the resulting Raney Ni layer on the activity of the hydrogen evo-
lution reaction (HER) can be explicitly demonstrated by means of electrostatic
measurements and steady-state current density potential curves (Fig. 9.12).
Regardless of pore size, uncoated Ni foams exhibit an overvoltage of approx-
imately 390 mV at a current density of 0.3 A/cm2 . As a result of the Raney Ni
coating, the overvoltage is reduced to approx. 70 mV. This corresponds to an im-
provement of approx. 320 mV (approx. 85 percent). The Ni foam structures coated
with Raney Ni therefore represent highly active electrodes for the development of
hydrogen in highly concentrated alkaline solutions. The development of the over-
voltage over time is close to constant. The steady-state current density potential
curves confirm the high activity of the Raney Ni-coated Ni foams. At the same
time, no transport limit is evident, even with high currents, meaning that an un-
hindered gas transport through the porous 3D Raney Ni foam structure can be
assumed.
Of all the electrolysis technologies, alkaline electrolysis (AEL) has so far
proven the longest service life of 90,000 h, which is due, among other things, to
Fig. 9.12 Electrochemical evaluation of Ni foams coated with Raney Ni with different pore
sizes (450 µm, 580 µm) compared with uncoated foams. Links: Galvanostatic measurements
at a geometric current density of 0.3 A/cm2 for 5 h; right: Steady-state current density po-
tential curves (Tafel plots) after 5 h at 0.3 A/cm2 for different electrode materials at 29.9%
KOH by mass at 60 °C. (Fraunhofer IFAM)
the robustness of the materials and components used. With regard to current den-
sity, however, alkaline electrolysis is lagging behind the other methods. Currently,
studies are being conducted regarding the operation of AEL cells up to 1 A/cm2 ,
and in some cases beyond that, which would significantly increase the space-time
yield of alkaline electrolyzers in the foreseeable future. At the same time, compo-
nent and plant manufacturers are currently making extensive efforts to automate
production in order to achieve annual production capacities in the gigawatt range
by the late 2020s.
9.2.3 PEM electrolysis
The general structure of a PEM electrolysis cell is shown in Fig. 9.13. Both half-
cells are separated by a proton conducting membrane that serves as a solid elec-
trolyte in the cell. A thin catalyst layer (CL) is applied as an electrode on both sides
of the membrane. Oxygen is produced at the anode and hydrogen is produced at
the cathode; see the reaction equations in Table 9.1. This structure is known as
membrane electrode assembly (MEA). Porous transport layers (PTLs) are pressed
against the electrodes, which draw the electrical current to or from the electrodes.
Fig. 9.13 Schematic diagram of a PEM electrolysis cell; PTL: porous transport layer.
(Fraunhofer ISE)
The porous structure also means that the anode can be supplied with educt water
and the gases produced can be dissipated at both electrodes. The PTL is connected
to a flow field plate, which ensures that the educt water is distributed over the entire
surface area of the cell and the gases produced are discharged from the back side
of the PTL. In a cell stack, the flow field is often part of the bipolar plate. Alterna-
tively, multi-layered expanded metals can be used as flow fields, as they work as
electrically conducting spacers.
The proton conducting membrane is usually a perfluorinated sulfonic acid
membrane (PFSA) such as Chermour’s Nafion® or FuMA-Tech’s fumapem® with
a thickness of 50 to 180 µm. This membrane is characterized by its very high
proton conductivity, low gas permeability and excellent mechanical and chemical
stability. During operation, however, it is not possible to completely suppress the
permeation of oxygen and hydrogen through the membrane. Typical H2 purity
levels at cell output range from 2.8 to 4.0 (based on dry hydrogen). The degree
of purity can be significantly increased by using an internal, catalytically active
recombination layer. As it absorbs water, the PFSA membrane swells. Undesirable
wrinkles can occur, especially at large surface dimensions, and the membrane
in the pressed cell can be mechanically damaged as a result. In order to prevent
this, mesh structures inside the membrane provide reinforcement. These coun-
teract the swelling, but also reduce conductivity. For these reasons, alternative
ionomers are also being developed for PEM electrolysis, for example hydrocarbon
membranes. However, for technical reasons, these materials have not yet become
widely established in practice.
Proton-conducting membranes have highly acidic properties. Combined with
the high potential in an electrolysis cell, it is necessary to use precious metals as
electrode materials. The catalyst layers are only a few micrometers thick on the an-
ode and cathode alike. Efficient hydrogen production requires platinum as a HER
catalyst. In most cases, platinum is placed on a carbon support for more efficient
material use. The precious metals iridium (the preferred option), ruthenium and
their oxides are used for the oxygen production. Modern MEAs have a catalyst
load of approx. 1.5 to 2.5 mg/cm2 on the anode side and approx. 0.8 to 1 mg/cm2
on the cathode side [50, 51]. Typical power characteristics are shown as voltage-
current density characteristics in Fig. 9.5. There are significant efforts underway to
reduce these loads by approx. 40 to 60 percent in newer MEA generations. Tita-
nium or titanium dioxide-based carrier materials are also being developed to this
end for the anode side. The primary reason for reducing precious metal loads is
not because it is necessary to reduce material costs, but rather because iridium is
a critical material [31]. The annual production volume of iridium is only about 6 to
8 tons worldwide. At current power densities and loads, however, the production
of PEM electrolyzers requires approx. 650 to 700 kg of iridium per 1 GW output.
In order to produce PEM electrolyzers capable of producing gigawatts of energy
on a large scale in the future, the specific iridium consumption should therefore
be reduced to approx. 50 kg/GWel . Alternatively, there are ongoing efforts to re-
place precious metals in PEM electrolysis. Transition metal oxides and sulfides, in
particular, are, alongside alloys, subjects of extensive discussions in scientific liter-
ature as possible materials for anodes and cathodes. Significant progress has been
seen in recent years in terms of stability and performance characteristics. However,
most of these materials are still a long way from an industrial application due to the
lack of studies conducted on upscaling these materials and the fact that they are not
sufficiently incorporated into the necessary membrane electrode assemblies [52].
The porous transport layers (PTL) in PEM electrolysis are only a few hun-
dred micrometers thick. They ensure an even distribution of the electrical current
between the bipolar plate and the electrodes, and allow high gas and water perme-
ability. On the hydrogen side, the electrode potential is close to 0.0 VRHE , making
the use of carbon paper or carbon-based non-woven material possible, just as with
PEM fuel cells. On the oxygen side, titanium is almost exclusively used as a ma-
terial for its high corrosion resistance, on account of the high potential. However,
when titanium comes into contact with oxygen, it forms a passivating and electri-
cally insulating layer of titanium oxide on the surface. For this reason, protective
Fig. 9.14 Typical PTL and spacer materials for PEM electrolysis cells: (a) sintered Ti pow-
der, (b) sintered Ti-based non-woven fabric, (c) Ti expanded metal, and (d) carbon paper.
(Fraunhofer ISE)
layers are sometimes also applied to ensure better electrical contact. Fig. 9.14
shows porous transport layers of titanium and carbon for PEM electrolysis. Sin-
tered Ti-based non-woven fabrics have become established as the best material for
a PTL, but this also greatly depends on the particular cell design and level of poros-
ity [53]. Sintered Ti particles have also produced good results but are significantly
more expensive than expanded metals and non-woven materials.
The flow field or bipolar plates in a stack must also be made of corrosion-
resistant materials such as titanium or coated steel. However, the latter approach is
not used in commercial products today. On the hydrogen side, cheaper carbon com-
posite materials could also be used, but the use of a single-layer titanium bipolar
plate is more cost-effective. The plate is made from thin sheets (of a few hun-
dred micrometers to approx. 1 mm) through milling, punching, deep drawing or

hydroforming processes and can also be coated to minimize contact resistance by
passivation.
The most common design for PEM electrolysis is the filter press design with
approx. 50 to the current maximum of 220 cells per stack, which are electrically
connected in series, with the fluids flowing through them in parallel. This means
that two adjacent cells are separated by a bipolar plate (BPP), which simultane-
ously acts as the anode of one cell and the cathode of the other. These days,
typical cell sizes have geometric dimensions of 300 to 1500 cm2 (active surface
area). Tests are also underway using the first prototypes with a cell area of up to
5000 cm2 . The cell thickness ranges from 2 to 5 mm. This makes PEM electrolysis
stacks significantly more compact than alkaline electrolysis stacks in terms of the
number of cells, but above all also the cell area and volume. In addition, they have
a higher power density. The electrical connection power of a single stack ranges
from several hundred kilowatts to the current upper limit of approx. 1.5 MW [54].
In the meantime, the rectangular design has become the norm for PEM electrol-
ysis, despite having to withstand pressures of up to almost 50 bar. Due to the cell
design, a PEMEL stack can also be operated at differential pressure with several
MPa of differential pressure between the anode and cathode. While the cathode
produces hydrogen under pressure, the oxygen side works at close to atmospheric
conditions. The advantage of this is that the hydrogen is already electrochemically
compressed, meaning that the significantly less efficient process of mechanical
compression at a later stage is no longer necessary. In addition, low-cost compo-
nents can be used for the peripherals on the oxygen side, as these do not have to be
pressure-resistant, and ultimately, the uncompressed oxygen represents a signifi-
cantly lower safety risk. A stack of this kind, operated at differential pressure, can
be seen in Fig. 9.15. To examine the long-term behavior, all cell voltages in this
stack are tapped individually. The single voltage tap can be seen on the right side
of the image.
The electrolysis stack is the key component of any electrolysis system. In order
to achieve higher production capacities, several stacks are usually interconnected in
a single system. A number of additional components are required for water splitting
in order to be able to operate the stacks as desired and in a stable condition. The
basic structure of a PEMEL system is comparable to that of an alkaline system.
The DC current required for water splitting is specified via a rectifier. After the
water is treated to achieve DI water quality, the water is circulated on the anode
side in order to continuously feed educt water to the PEMEL cell and to cool the
cell via a heat exchanger. No such circulation is required on the cathode side. On
both sides, gas-water separators and demisters are placed behind the stack exits in
Fig. 9.15 Measurement of a NEL Hydrogen 250 kW PEMEL stack with single cell voltage
tap for monitoring the cells in continuous operation. (Fraunhofer ISE)
order to retain the liquid water. This also accumulates on the cathode side through
the electro-osmotic water transport via the membrane and has to be returned to the
cell. The hydrogen, which is still wet, is then dried; residual oxygen is catalytically
removed in a deoxo stage. Pressure-retaining valves regulate the pressure on the
anode and the cathode.
9.2.4 AEM electrolysis
Two technologies are currently used in the classical production of hydrogen

through electrolysis of liquid water: Alkaline electrolysis (AEL), in which base
metals such as nickel alloys and non-ferrous compounds are used as electric cat-
alysts, bipolar plates (BPP) are made of cost-effective nickel-plated steel and the
alkaline liquid electrolyte used is KOH. The disadvantages of the method are the
use of potassium hydroxide as a circulating electrolyte and the (still) low current
densities. The more recent PEMEL technology uses an acidic membrane elec-
trolyte, DI water, precious metals (iridium, platinum) as catalysts and titanium (Ti)
as the material for bi-polar plates and the porous transport layers (PTL). Due to
the compact design of the membrane electrode assembly (MEA) and the use of
precious metal catalysts, PEMEL cells are characterized by cell voltages below
2.0 V and current densities greater than 2 A/cm2 . There are also disadvantages with
this method, in particular the high material costs for the catalyst-coated membrane
(CCM) and the titanium-based bipolar plates and the porous transport layer (PTL).
In particular, the availability of iridium as an anode electrocatalyst is critical [31].
In order to minimize the disadvantages of the two abovementioned methods,
while nevertheless exploiting their respective advantages, widespread efforts are
underway to combine both technologies in AEM electrolysis by using an alkaline
anion exchange membrane (AEM) as a solid electrolyte (Fig. 9.2). The system
design of an AEM electrolyzer is similar to that of a PEM electrolyzer but uses
cost-effective materials that have been proven to be effective in AEL. For the
construction of an efficient AEMEL system, stable, conductive and, above all,
cost-effective anion exchange membranes must be developed, while stable, elec-
trochemically active catalysts need to be found.
An anion exchange membrane installed in an electrolyzer must have a level of
conductivity above 0.1 S cm1 and a thickness of between 50 and 80 µm. In ad-
dition, the membrane should be stable in the long term. The basic approach for
constructing an anion conductive membrane is to produce a self-conducting ho-
mogeneous membrane. In this case, cations are bound to a stable non-conductive
polymer framework, for example to polyarylethers or fluorinated polymers. Qua-
ternary ammonium salts or analog phosphonium and sulphonium salts are used as
cations [32–34]. Since the ion mobility of the hydroxide ion is significantly lower
than that of the protons, the anion exchange membrane usually has a lower level
of conductivity than the proton exchange membrane (PEM), e.g., Nafion. Various
strategies are being pursued to improve the conductivity of the alkaline membrane.
On the one hand, researchers are attempting to increase the ion exchange capac-
ity (IEC) within the membrane. However, this leads to a strong swelling of the
membrane due to water absorption, which in turn has negative consequences for
the stability of the membrane and the integrity of the membrane electrode assem-
bly. On the other hand, the development of “phase-segregated” AEMs is favored.
By using hydrophobic and hydrophilic phases, the aim is to create specific “ionic
highways” where fast ion transport is possible. Pan et al. were able to produce an
anion exchange membrane with the conductivity of Nafion using a functionalized
polysulfone [35]. The main disadvantage of homogeneous membranes is reduced
stability. This is due to the fact that the quaternary salts, a good leaving group for
the purposes of Hofmann elimination or a nucleophilic substitution reaction, can be
quickly separated in the alkaline medium [36]. Various concepts for avoiding this
degradation pathway have been discussed, including the use of sterically demand-
ing quaternary amines, bypassing hydrogen atoms in “ position to the quaternary
amine and using electrolytes containing carbonate [32, 37–39].
In addition to the homogeneous membranes, heterogeneous membranes are also
a subject of discussion. These are characterized by the fact that a mostly inor-
ganic ion exchanger material is embedded in an inert polymer matrix. For example,
polyethylene oxides (PEO or PEG) or polyvinyl alcohols (PVA) are used, in which
an inorganic salt, such as KOH, can be dissolved. The conductivity of the mem-
brane is only made possible by the salt. These are known as ion solvent membranes.
Conductivity of up to 103 S cm1 can be achieved here. One particularly notewor-
thy variant is the use of polybenzimidazole (PBI), which is characterized by good
chemical stability and has conductivity of up to 101 S cm1 [40–42].
Another current task in the area of AEMEL R&D is the development of suitable
electric catalysts. Stable, PGM-free and electrochemically active catalysts are re-
quired for this [43]. When it comes to the OER reaction, not only precious metals,
but also non-stoichiometric transition metal oxides appear to be the primary poten-
tial candidates [44]. Perovskites (ABO3-• ) and layered double hydroxides (LDH)
have been shown to be particularly catalytically active here. Most studies, however,
do not carry over well to a real-world application environment. From a technical
point of view, mainly Ni-Fe and Ni-Co compounds have become established for
the OER side. For example, at 80 °C and 1 M, NaOH overvoltages of 265 mV with
a current density of 0.5 A/cm2 have been achieved [45]. A study from 2015 presents
a series of transition metal (oxy)hydroxides as trifunctional catalysts (OER, HER,
ORR). Although the study only looked at low current densities, these are highly
interesting for reversible AEMEL cells [46].
With regard to HER, several studies have been carried out to attempt to improve
the catalytic properties of cathode materials [47, 48]. In addition to the catalysts
containing precious metals (Pt, Pd), it is Ni-based compounds, especially Ni-Mo,
that have become established. In principle, the preferred solution would be to use
DI water, as in PEM electrolysis. Under these conditions, however, most AEM
electrolyzers show high cell voltages. The use of an alkaline electrolyte, on the
other hand, seems to be more feasible on a technical basis, as the HER/OER cata-
lysts are more stable and active here due to the alkaline environment, which poses
less of a risk of corrosion. Cell voltages of less than 1.9 V could be achieved with
current densities up to 1 A/cm2 . For example, Liu et al. showed that an AEM elec-
trolyzer with an optimized AEM and NiFeCo catalyst (HER) and NiFe2 O4 delivers
a stable (2000 h) current density of 1 A/cm2 at approx. 1.9 V on the anode side in
1 M KOH [49].
Despite the material and technical challenges, AEM electrolysis has the poten-
tial to become an important technology in the field of water electrolysis. Assuming
a technological revolution similar to that of membrane-based PEM electrolysis, it
can be expected that AEMEL systems will be established on a megawatt scale by
the late 2020s.
9.2.5 High-temperature steam electrolysis
High-temperature electrolysis (HTEL) is generally understood to mean electroly-

sis based on solid oxide electrolytes, known as solid oxide electrolysis (SOEL).
High-temperature electrolysis converts steam into hydrogen and oxygen, which
significantly reduces the need for electrical energy in the conversion process com-
pared with water electrolysis. The reason for this is the lower free enthalpy (Gibbs
free energy) of the steam compared with the liquid water, which is reduced by the
heat of evaporation. This relationship is illustrated in Fig. 9.4. In addition, a declin-
ing trend in Gibbs free energy is also evident as the temperature rises. This results
in a lower decomposition voltage and means that less electrical power is required
for electrolysis. Thermodynamically speaking, electrolysis is an endothermic pro-
cess that can only be carried out given a continuous supply of heat. The internal
resistance of the cell serves as a source of heat. In the case of high-temperature
electrolysis (HTEL), the Joule heat generation from the internal resistance of the
cell compensates for the other losses of the steam splitting once a certain operating
voltage is reached. If the amounts of thermal energy required for electrolysis and
heat provided by voltage losses are the same, the operating voltage is referred to as
the thermoneutral voltage. Due to the self-sustaining nature of the reaction, high-
temperature electrolysis is only viable when operated close to the thermoneutral
voltage. The absolute value of the thermoneutral voltage depends on the operating
temperature of the cell and the gas composition. At temperatures between 700 and
850 °C, it is in the range of 1.28 to 1.30 V. Since a higher local current density
generates a greater local steam conversion, and thus a higher local cooling capac-
ity, the current density and temperature distribution over the cell surface in SOEL
is much more uniform than in fuel cell operations. There, a higher local current
density leads to hotspots due to the exothermic reaction.
Despite the thermodynamic advantages, the high operating temperature poses
major challenges to the materials and seals of an HTEL stack. The first high-
temperature electrolyzers were constructed with tubular cells and had relatively
high internal resistances. The progress in the development of the planar solid oxide
Fig. 9.16 Schematic repre-

sentation of a solid oxide
cell according to [56].
(Fraunhofer IKTS)
cell for fuel cell applications, as well as the considerable increase in power density
while simultaneously reducing manufacturing costs, enabled the development of
a new generation of high-performance, planar HTEL cells.
Current cell structures of high-temperature electrolyzers (Fig. 9.16) are based
on Y2 O3 (YSZ) or Sc2 O3 (ScSZ) stabilized ZrO2 as an electrolyte, a cermet cathode
(Ni/G8VDC or Ni/8YSZ) and a perovskite (La0,6 Sr0,4 Co0,8 Fe0,2 O3 ) or composite
anode (perovskite/electrolyte). The steam is fed to the cathode side of the cell. As
nickel is a component of the steam electrode and oxidizes into NiO in pure steam,
a small amount of hydrogen is added to the steam. In electrolysis, the oxygen is
removed from the steam electrochemically and then is directed to the anode side
via an oxygen-conducting electrolyte. The hydrogen concentration on the cath-
ode side increases continuously from the gas inlet to the gas outlet. Especially in
concentration ranges of more than 90 percent hydrogen in the steam, the Nernst
voltage rises considerably [55], meaning that there is a considerable increase in
the operating voltage and the electrical energy supplied during electrolysis. This
does not make sense for an economical operation. In a concentration range of 20
to 85 percent hydrogen, the Nernst voltage only slowly changes as hydrogen con-
tent increases, meaning that at a hydrogen concentration of 10 to 20 percent in the
steam, steam utilization rates of 80 percent are not only reasonable to expect, but
entirely possible.
Fig. 9.17 Components and structure of an SOC stack. (Fraunhofer IKTS)
In the last decade, planar stack construction has become established, with two
different cell concepts widely used by various companies around the world for
stack building:
Electrolyte supported cells (ESCs) based on a thin electrolyte substrate (60 to

150 m) as the supporting element
Cathode supported cells (CSCs) consisting of a Ni-YSZ cermet substrate with
a sintered electrolyte layer (5 to 10 m)
Although the SOEL cell itself is the heart of an HTEL electrolyzer, it is only
one part of the stack, which also contains gas manifolds, seals and current collec-
tors. The structure and essential components of a planar SOC stack are shown in
Fig. 9.17. The stack consists of the coated metallic interconnector (bipolar plate),
the ceramic cell itself, glass seals and contact elements on the steam (Ni mesh) and
air side (oxide contact layers).
The SOEL cells are manufactured by means of tape castings and screen-printing
technology. The basic sequence of cell production is shown in Fig. 9.18. It is
a “sheet-to-sheet” production process that can be implemented with high cycle
times. A tunnel furnace can be used in mass production for electrode baking.
Fig. 9.18 Production steps for an electrolyte-supported cell. (Fraunhofer IKTS)
Fig. 9.19 Production steps for an SOEL stack. (Fraunhofer IKTS)
The production of an HTEL stack is shown schematically in Fig. 9.19. The

process is similar to the manufacturing process of NT electrolysis stacks, with the
exception of the joining process. The joining process includes not only the serial
contacting of the cells, but also a check of the functionality and tightness of the
stack. The forecasts of the cost of production of SOC stacks are thus very close
to the targets for PEMEL and AEL stack production. Due to the high operating
temperature and the thermo-mechanical stresses in ceramic components that de-
velop when the stack cools down, the lateral dimensions of the SOC cells that can
currently be implemented are limited to approx. 25 25 cm2 [57]. For high plant
performance, smaller stacks (with cell sizes 10 10 cm2 to 15 15 cm2 ) are there-
fore integrated into larger modules [58–60]. The individual stack modules are then
installed in larger plants.
Since the solid oxide cell works in electrolysis, fuel cell and bidirectional op-
eration (known as a reversible solid oxide cell, rSOC), it seems appropriate to use
the same modules for different applications. There are also considerable differ-
ences in the electrode structure as well as in the thermal management in the stack
for stable SOFC and SOEL operations in the long term, resulting in application-
specific cells and stack designs [61]. While the stable electrolysis operation of ESC
at 800 °C has already been demonstrated by the development of robust electrodes
at Fraunhofer IKTS, long-term operation of CSC at lower operating temperatures
(650 to 750 °C) remains a challenge. Although the data of CSC stacks show a low
rate of degradation after an initialization phase, this is often not taken into account
when calculating the overall degradation [62].
The plants for high-temperature electrolysis have not yet reached the level of
technological maturity required for series production as seen in alkaline and PEM
electrolysis. Nevertheless, there are still some impressive demonstrations for rSOC
and SOEL pilot plants, as exemplified by a company from Dresden, SunFire GmbH
(formerly Staxera GmbH), which was established on the basis of stack technology
development and technology transfers from Fraunhofer IKTS. Since 2014, the fo-
cus of the company has been on the development of synfuels via high-temperature
electrolysis. SunFire has demonstrated the first industrial plants for SOEL oper-
ation (GreenHy, Salzgitter, 750 kW) [63] and for reversible electrolysis/fuel cell
operation (rSOC) (with Boeing, 150 kW/30 kW) [64].
Although the PEM stack, for example, is very well suited for pure hydrogen
production and use, rSOC technology has advantages in efficiency for stationary
applications, in particular thanks to its compact integration and efficient use of
heat. A 62.7 percent round-trip efficiency was recently demonstrated in this area by
the Forschungszentrum Jülich research center for an rSOC plant with an electrical
output of 5 kW [65].
9.2.6 Co-electrolysis of water and carbon dioxide
Due to the high temperatures, CO2 electrolysis may take place in an SOC high-
temperature cell similar to steam electrolysis. However, the free enthalpy of carbon
dioxide reduction is greater than that of splitting water (Fig. 9.20). For this rea-
son, higher electrical voltages are required for direct electrolysis of CO2 . The
production of carbon monoxide (CO) from CO2 in HTEL cells has already been
commercially implemented by the company Haldor Topsoe [66]. A CSC stack with
a modified cermet electrode is used for this purpose. The special feature of pure
CO2 decomposition is the avoidance of carbon formation from the Boudouard reac-
tion. Thermodynamic calculations show that the carbon-forming threshold in the
CO2 /CO gas mixture shifts towards higher CO concentrations as operating tem-
perature increases. Above 750 °C, carbon monoxide is thermodynamically stable.
Carbon formation is possible below this temperature. In this case, the nickel in
the cermet electrode acts as a catalyst for carbon formation. Due to the temper-
ature gradients in the stack, there are always certain temperature ranges that are
prone to involve carbon formation. The electrochemical processes in the cell are
similar to those of steam electrolysis, with the overvoltage for CO2 reduction on
the cathode being significantly higher than the overvoltage of steam decomposi-
tion (Fig. 9.20). The addition of steam to CO2 fundamentally changes the reaction
Fig. 9.20 Thermodynamic comparison of steam electrolysis and CO2 electrolysis. (Fraun-
hofer IKTS)
mechanism. Under the operating conditions of the SOEC, the rWGS reaction es-
tablishes a thermodynamic equilibrium state on the side of the products. Thus, by
using the smaller overvoltages for steam decomposition, carbon monoxide can be
produced in addition to steam. Since the overvoltage of hydrogen reduction is sig-
nificantly lower than that of CO2 reduction, the electrochemical steam reduction
(with hydrogen production at the three-phase boundary) and the reverse water gas
shift reaction (rWGS) with CO production on the surface of the catalyst (Ni) take
place simultaneously in the high-temperature cell (see the reaction equations in
Table 9.1). Nevertheless, the production of H2 and CO from the H2 O/CO2 mixture
is generally referred to as co-electrolysis.
This fact is also reflected in small differences in electrical energy consumption
for H2 O and H2 O/CO2 electrolysis. These have been detected for the first time
by Fraunhofer IKTS in measurements on cells and stacks at 800 °C. The higher
electrical energy consumption during co-electrolysis can be explained thermody-
namically by the slightly higher open circuit voltage when CO2 is added to steam.
The no-load voltage results from the higher free enthalpy (GR ) of CO2 at tem-
peratures below 823 °C (Fig. 9.20). This also changes the thermoneutral voltage,
which influences thermal management and the energy balance. Even if a gas mix-
ture of steam and carbon dioxide is used, H2 must be introduced at the gas inlet in
order to avoid Ni oxidation at the cathode. In addition, the utilization of steam and
CO2 is limited not only by the concentration-related increase in the Nernst voltage,
but also by the limit of carbon formation.
Co-electrolysis offers new possibilities for the single-stage production of syn-
thesis gas and its coupling with chemical synthesis processes. In particular, the use
of heat from exothermic reactions, as is the case in Fischer-Tropsch synthesis (e-
kerosene production), the Haber-Bosch process (e-ammonia production) and the
Sabatier reaction (e-methane production). The use of the heat of reaction to gen-
erate steam as an educt for electrolysis reduces the electrical energy required and
thus increases the overall electrical efficiency of the synthesis process. Initial fea-
sibility and demonstration projects at Fraunhofer IKTS show that, by combining
co-electrolysis with Fischer-Tropsch synthesis, highly efficient processes for the
production of wax and diesel are possible [67].
9.2.7 High-temperature ceramic-based proton-conducting

electrolysis
A promising option for hydrogen production is steam electrolysis with proton con-
ducting ceramic electrolytes (PCCEL). The motivation for the development of
a proton-conducting ceramic cell for higher operating temperatures comes from
fuel cell development and is based on the fundamental advantages of proton con-
duction over oxide ion conduction at lower operating temperatures (500 to 600 °C).
This results in a longer service life and lower costs for passive stack and BoP com-
ponents, as well as simpler system integration. At the same time, other advantages
of the SOEL, such as fast electrochemical kinetics and electrodes free from pre-
cious metals, come into play. Compared with SOEL technology, the lower steam
content on the cathode side constitutes an additional advantage, as the steam has
a far lesser influence on the Nernst voltage. However, the proportion of steam in
the gas also has a decisive influence on the proton conduction in the electrolyte,
meaning that sufficient steam still has to be fed into the cell on both sides. Similar
to PEM electrolysis, the steam is fed to the oxygen side in the proton conducting
cell through humidification. In the anodic reaction step on the air side, the pro-
tons are separated from the steam molecule and transported via proton conductors
to the cathode side, where they are reduced to molecular hydrogen (Table 9.1).
Thermodynamically speaking, this process has the same energetic advantages over
liquid water electrolysis as the SOEL process. Due to the enthalpy of evaporation
through the addition of heat, the Nernst voltage at which steam electrolysis starts
is considerably lower than in the case of electrolysis of liquid water.
There are a few different options for electrolyte composition from the
last ten years, with most cells having a proton conductor based on BZY
(Ba(Zr,Y,Yb)O(Zr,Y,Yb)O3 ). It is assumed that the proton conduction in this
material, due to the incorporation of proton defects as hydroxide ions (OH ), runs
according to the following mechanism in the presence of steam [68]:

H2 O.g/ C VO;.BZY/ C OO;.BZY/ ! 2OHO;.BZY/ (9.4)
A major disadvantage of the PCCEL process is the fact that the BZY electrolyte is
permeable to oxygen ions as well as protons [69], meaning that the Faraday effi-
ciency of the proton conducting cell is greatly reduced. In the case of BZY-based
proton conductors, the proton conduction and the oxide ion conduction improve as
the temperature increases. At temperatures well above 600 °C, oxide ion conduc-
tion is comparable to proton conduction and makes a decisive contribution to ion
transport and electrochemical processes. For this reason, the operating temperature
of these cells is limited to approx. 500 to 600 °C.
A cermet of Ni and BZY is used as the cathode material for cell production
and a perovskite of La0.6 Sr0.4 Co0.8 Fe0.2 O3 or Ba0.5 Sr0.5 Co0.8 Fe0.2 O3 is used for the
anode [70]. As nickel is used as the catalyst on the cathode side, small quantities of
hydrogen must be continuously fed into the catalytic converter during operation,
similar to the SOEL process. Due to the low proton conductivity and the high grain
limit resistance in the BZY electrolytes, the cells are designed with the Ni/BZY
substrate as a support mechanism and a thin BZY layer as an electrolyte, which
corresponds to the cathode-supported cell concept (CSC) of a SOEL cell.
The level of technological maturity in PCCEL is less advanced than solid ox-
ide electrolysis. In Europe, the materials for proton conductors and tubular proton
conducting cells have been developed by SINTEF for decades [71]. However, as
in the case of SOEL, planar cell technology also offers a higher power density. De-
velopments in this area are currently being driven by players from outside Europe.
In the meantime, good power densities in industrial CSC prototype cells based on
BZY electrolytes are being demonstrated in Japan (Panasonic, Nippon Shokubai
[72, 73]) and the USA (Fuel Cell Energy [74]). A planar PCCEL stack is structured
in the same way as an SOEL stack. Activities for stack development based on PCC
cells can be found, for example, in [74]. Nevertheless, due to its short service life
and poor Faraday efficiency, this technology is still a few development steps away
from the demonstration phase.
9.3 Other innovative processes for hydrogen production
9.3.1 Steam reforming with carbon capture and utilization
Hydrogen can be obtained from a wide range of hydrogen-containing energy car-

riers by means of steam reforming. The general formula for hydrogen production
from alcohols and hydrocarbons through steam reforming is:
y
Cx Hy Oz C .x z/H2 O ! xCO C x z C H2 (9.5)
2
The reaction is usually carried out at temperatures above 700 °C using a heteroge-
neous catalyst and on an industrial scale in tubular bundle reactors filled with the
catalyst. By the homogeneous combustion of a part of the hydrogen carrier or fuel,
the endothermic steam reforming process is supplied with energy from an external
source. In this way, on a large scale, the majority of the world’s hydrogen produc-
tion can be carried out at low production costs of approx. C 2/kg using natural
gas (48 percent) without competition, while a further approx. 30 percent of the hy-
drogen is produced from crude oil, whereas only approx. 4 percent is produced by
electrolysis processes and the remaining portion is produced by carbon gasification
(Fig. 9.1). Alongside hydrogen, the primary product of steam reforming is carbon
monoxide. The mixture of hydrogen and carbon monoxide is called synthesis gas
and can be used for many chemical syntheses. If the hydrogen is to be obtained
in pure form, the carbon monoxide, which is already partially converted to car-
bon dioxide in the above-mentioned water-gas-shift reaction during the reforming
process, is converted in a water gas-shift reaction in a separate reactor, which is
operated at lower temperatures. In addition to the gas washing, pressure swing ad-
sorption systems are mainly used for hydrogen purification. These systems enable
the separation of all other gases by adsorption at high operating pressure. The sep-
arated gases are released again at a lower operating pressure (desorption). These
separated gases contain proportional amounts of carbon monoxide and some hy-
drogen and can be used to supply energy for steam reforming.
The chemical equilibrium of the steam methane reforming (Eq. 9.5 where x D 1,
y D 4 and z D 0) is shifted toward the side of the natural gas by the increased
operating pressure of the pressure-swing adsorption systems (see Table 9.2). This
means that unconverted methane is always present in the separated gas mixture.
In fact, the increased operating pressure is a disadvantage, but it is accepted if the
hydrogen is to be purified.
An alternative to pressure swing adsorption is the separation of the hydrogen
by membranes. The most widespread membranes are made of palladium, in which
9.3 Other innovative processes for hydrogen production 237
Table 9.2 Equilibrium conversion of methane reforming at a steam-to-carbon (S/C) ratio

(i.e. the molecular ratio of steam to methane in the feed) of 2.5 at various pressures and
temperatures
p [bar]
T [°C] 1 5 10
600 71.5 41.9 32.3
700 95.4 70.1 55.9
800 99.6 92.6 82.1
the hydrogen dissolves atomically and moves to the other side of the membrane as
a result of the increased pressure, where it recombines to form molecules. However,
these membranes are very expensive and have not become widely established on an
industrial scale. After combustion, the separated gas mixture contains only carbon
dioxide and water. The latter can be easily separated, and the carbon dioxide is
reused as process gas or for chemical synthesis. Furthermore, grouting in caverns
has been discussed and tested, but possible long-term negative effects are not yet
sufficiently understood.
If the carbon dioxide is separated and permanently stored by CCS processes,
the hydrogen produced is called blue hydrogen. Fig. 9.21 shows a block diagram of
a corresponding system that additionally generates an excess of electrical energy. If
carbon dioxide is obtained from the air or from industrial processes such as cement
production, it can be converted to methane by means of electrolytically produced
Fig. 9.21 Block diagram of a system for methane steam reforming with integrated CO2
separation. (Fraunhofer IMM according to [75])
hydrogen by the reverse reaction of steam reforming (methanation) and used as

a hydrogen carrier in the same way as fossil natural gas. This is possible without
conversion of the natural gas network and in a concept that is fully compatible with
fossil natural gas and allows a gradual conversion to renewable synthetic methane.
9.3.2 Methane pyrolysis
In contrast to the steam reforming described above, methane pyrolysis involves the
conversion of methane (from natural gas or biogas) without adding water. The en-
dothermic reaction produces so-called turquoise hydrogen (Fig. 9.2). The methane
is decomposed into solid carbon and hydrogen, and no carbon dioxide is produced.
At atmospheric pressure, methane is no longer thermodynamically stable above
1000 °C. At increased pressure, this temperature limit increases considerably (to
approx. 2000 °C at 10 bar). The thermal decomposition can be achieved well above
this temperature without a catalyst, for example at temperatures of approx. 2000 °C
in plasma. The production of carbon black by thermal decomposition for the tire in-
dustry, for example, is a process that has been established on an industrial scale for
about 100 years, albeit with a low efficiency. It has already been used for hydrogen
production using natural gas from oil production. As described, the decomposition
reaction only takes place at high to very high temperatures. This creates a techni-
cal problem, namely that the solid carbon has to be transported out of a very hot
reactor in a moving bed, which nevertheless must be gas tight.
If the methane pyrolysis is carried out using a catalyst, it is called thermocat-
alytic decomposition. This reaction already takes place below 1000 °C. The lowest
reaction temperatures can be achieved with nickel-containing catalysts. However,
the catalyst is usually rapidly deactivated by the carbon produced. One solution
to this problem is the use of fluidized bed reactors, in which the catalyst can be
moved relatively easily between different reactors. If the intention is to regener-
ate the catalyst, the carbon has to be oxidized, thereby creating undesired carbon
oxides (Fig. 9.22).
With regard to the technical maturity of the systems, plasma processes are cur-
rently the most advanced. However, they are mainly aimed at the production of
carbon. Hydrogen production is currently not economically viable, mainly due to
the high market price of gray hydrogen. It is only used as an energy source for
the process. If regenerative electricity is used to generate the plasma, the plasma
processes have a high potential to enable the production of turquoise hydrogen in
large quantities in the future. The resulting carbon can then be used in a variety
of technological processes, which could compensate for the higher manufacturing
Fig. 9.22 Block diagram of a system for the thermocatalytic decomposition of natural gas
with integrated catalyst regeneration. (Fraunhofer IMM according to [76])
costs of the hydrogen. If the energy supply of the decomposition process originates
from renewable sources, the ecological footprint is reduced considerably (Fig. 9.2).
The projected production costs of hydrogen through the thermal or thermocat-
alytic conversion of natural gas are currently only slightly higher than the costs of
approx. C 2/kg resulting from steam reforming at approx. C 3/kg. In summary,
the handling of the solid product in the thermal decomposition of methane and the
large amount of carbon formed on the catalyst during thermocatalytic decomposi-
tion remain the main technical challenges to be tackled.
9.3.3 Photocatalytic systems
In another process, sunlight is used to directly produce green hydrogen by allowing

semiconductors to absorb the light and then catalytically split water on their sur-
face. This is referred to as a photoelectrochemical (PEC) or photocatalytic process,
because the charge carriers are generated directly in the semiconductor, which then
provide for the reduction to hydrogen or oxidation to oxygen [77]. There are only
a few materials that are suitable for both coupled processes, which is why two dif-
ferent semiconductors are usually combined for oxygen and hydrogen production.
This means that each can be better adapted to the different requirements and, in
addition, improved with specific co-catalysts for a greater level of activity.
For the future, the direct PEC splitting of water using light promises simple
structures with low system complexity. However, in practice, the technology is at
an early stage: In addition to the search for new materials for the individual re-
actions, there are only isolated studies looking at upscaling the cells, and actual
setups to determine the overall efficiency of the process chain from solar radiation
energy to hydrogen often suffer from low levels of energy efficiency [78, 79]. The
Fraunhofer-Gesellschaft is involved as part of its internal anticipatory research,
aiming to build bridges between scientific theory and application for a possible
technology of the future. Several concepts are being pursued: With the active co-
operation of Fraunhofer ISE, a multi-layered cell structure with highly active III–V
semiconductor materials based on gallium, indium and rhodium, among others, has
been developed. This cell has been able to demonstrate solar-to-hydrogen efficien-
cies of 19 percent—the highest value achieved to date [80]. A current project aims
to scale these cells to surface areas of up to 36 36 cm2 .
In a second approach taken by the Fraunhofer Institutes IKTS, IST and CSP, on
the other hand, robust and low-cost materials with an average efficiency of 10 per-
cent are used. They are constructed economically in integrated tandem systems
by means of coating processes established in the field of photovoltaics. Moreover,
they are built to be inherently scalable. Cells with an active surface area of as big
as one square meter will be created here. In general, in a tandem cell, the sunlight
passes through a semi-transparent anode that absorbs the short-wave light and hits
a cathode that absorbs the long-wave light. If both layers are applied to the oppo-
site sides of a glass substrate and are electrically connected (Fig. 9.23), hydrogen
is released on the cathode side and oxygen is released on the anode side when the
cell is exposed to sunlight. A thin film of water covers the surfaces [81], which may
be circulated if necessary. This structure allows higher efficiencies than separated
half-cells, and automatically results in important physical separation in order to
avoid the formation of explosive oxyhydrogen gas and expensive separation steps.
For such a tandem cell with a high degree of efficiency in the overall system,
the semiconductors used must be precisely matched to each other: The power per
unit of surface area is limited by the less active single electrode, and it is more
favorable if the same electrolytes and a similar pH value are used in the water for
both half-cells. Such adaptations of semiconductor properties are often achieved
by chemical composition, doping, or by using a specific structure and morphology.
The process conditions during sputtering and possible post-treatment steps can be
used to influence specific features of the microstructure of the separated layers.
Fig. 9.23 Structure of a tandem cell, the central element of a PEC unit. (Fraunhofer IKTS)
Fig. 9.24 (a) Targets for the sputtering process of different PEC semiconductors and
(b) a layer deposited with them; (c, d) layers with different structures in the scanning electron
microscope. (Fraunhofer IKTS)
In order to use carefully coordinated materials, semiconductor materials are being

developed and manufactured at Fraunhofer IKTS as targets for the generation of
these layers (Fig. 9.24).
The investigation of large PEC cells poses a problem: How can we get the sun-
light into the laboratory? For a reproducible measurement, it is very important to
accurately replicate natural light conditions. Fraunhofer CSP therefore operates
Fig. 9.25 Test bench for

PEC cells. (Fraunhofer
CSP)
a test stand for large PEC cells with LED solar simulators, which produces a ho-
mogeneous light field of 20 cm 20 cm with high spectral quality and is used for
the examination of the electrode materials and cells. A 2 m2 solar simulator is to
be used in a future expansion stage.
The PEC technology for the production of green hydrogen is at a very early
stage of its development. Key questions about efficiency, scalability and longevity
still need to be addressed. The Fraunhofer-Gesellschaft is conducting research on
this in various projects and also contributes its expertise in the development of
modular systems in order to help a promising technology go from laboratory scale
to a possible application.
9.3.4 Biological procedures
In addition to the above-mentioned processes for hydrogen production, hydrogen

can be produced from biologically available materials or directly through biolog-
ical processes. Biomass gasification is already a very advanced process used on
an industrial scale. In such a thermochemical conversion (partial combustion) of
biological material at temperatures of up to 900 °C, an oxidizer—such as steam,
air and/or oxygen—produces hydrogen, carbon monoxide, but also CO2 , methane
and steam. However, with prices of approx. C 7/kg of hydrogen, the costs of the
process are significantly higher than the costs of producing gray hydrogen [82].
In contrast to biomass gasification, the use of photosynthetic or (photo)fermen-
tative microorganisms offers a purely biological method for hydrogen production.
Some prokaryotic agents of these microorganisms (e.g., bacteria) and also a few
eukaryotic single cells (e.g. green algae) form hydrogen as a metabolic product.
Table 9.3 Hydrogen production pathways by microorganism [84]

Energy source Electron source Known organisms
Dark fermentation Organic molecules Clostridia,
(e.g. acids and sugars) enterobacteriaceae
Anoxygenic Light Acids, H2 S Purple bacteria
photosynthesis
Oxygenic Light Water Cyanobacteria,
photosynthesis single-cell green algae
By analogy with electrolysis and fuel cells, hydrogen serves as a natural storage
mechanism and source of electrons in the organism [83].
Basic metabolic pathways for biological hydrogen production are summarized
in Table 9.3. The enzyme hydrogenase plays a key role in the biological produc-
tion of hydrogen. Depending on the metal content, a distinction is made between
[Fe], [FeNi] and [FeFe] hydrogenases (Fig. 9.26; [85]). However, [FeFe] hydro-
genases are by far the most efficient systems for hydrogen production [86]. There
are numerous metabolic processes in which hydrogen is formed under anoxic con-
ditions, e.g., by the intestinal bacterium Escherichia Coli [87] or clostridia [88].
Hydrogenases in prokaryotic microorganisms in particular may become more im-
portant for the biological production of hydrogen in the future. These are able to
use wastewater and nutrients as an energy source for hydrogen formation. In the
future, purple bacteria may also be used for photofermentative hydrogen formation
[89] and cyanobacteria may be used for photobiological hydrogen formation [90].
In this context, it is remarkable that the hydrogenases of algae, such as the green
algae Chlamydomonas reinhardtii, enable photosynthetic hydrogen formation.
Fig. 9.26 Active centers of [FeFe] hydrogenases (left) and [FeNi] hydrogenases (right).
(Color code: gray: carbon, yellow: sulfur, red: oxygen, blue: nitrogen, brown: iron, turquoise:
nickel) [91]
Although theoretically, the energy conversion efficiency of these hydrogen pro-

duction systems is around 10 percent, the biological hydrogen production is usually
linked with other metabolic pathways [92]. This reduces the conversion efficiency,
meaning that hydrogen is usually not freely accessible in large quantities. The bi-
ological systems are also often sensitive to oxygen, which usually leads to the
inhibition of the enzymatic system, and thus to the cessation of hydrogen forma-
tion. However, the yield of free hydrogen can be increased and the direct biological
utilization inhibited through targeted genetic manipulation of the organisms [93].
There are also approaches to reduce the oxygen sensitivity of these systems [94].
In addition to hydrogen formation through hydrogenases, hydrogen is primar-
ily produced as a by-product during the natural nitrogen fixation (conversion of
atmospheric nitrogen to ammonia) by the enzyme nitrogenase [95].
These biological processes are still at a relatively low technology readiness level
(TRL), making them the subject of research in more theory-based projects. It will
take a great deal of time before they are ready for use on an industrial level.
9.4 Summary and outlook
CO2 -free water electrolysis, based primarily on renewable energies, will become
the key technology for sector coupling and the third phase of the energy transition
in Germany. From a cost perspective, green hydrogen cannot yet compete with
conventional gray hydrogen. However, the reasons for this are not so much related
to the technology involved, but are mainly due to the unsuitable way that the energy
market is designed.
With rising CO2 prices, cost reductions in electrolysis production and a growing
supply of cost-effective electricity from wind and solar energy worldwide, green
hydrogen is becoming more and more competitive. Blue or turquoise hydrogen, i.e.
non-renewable, but CO2 -free or low-CO2 hydrogen, is currently being discussed as
an interim option. This is because the necessary electrolysis capacities can neither
be built up at the appropriate speed, nor will the expansion of renewable energies
be sufficient in the next few years. How long that interim period will last remains
the subject of debate. It must also be stressed that the process routes of blue and
turquoise hydrogen production have also not yet been widely tested and are not
yet available, and that the long-term consequences of CCS technology and the
associated CO2 storage remain difficult to estimate.
It is not yet clear to what extent the low-temperature technologies (alkaline elec-
trolysis or PEM electrolysis) will establish themselves on a large scale. Whereas
9.4 Summary and outlook 245
alkaline electrolysis, the longest-used and most advanced technology, is the es-
tablished standard today, advances are being made in the (even more expensive)
field of PEM electrolysis, which promises improved flexibility, more compactness
and greater efficiency at the same power density, with the goal of reducing costs
through mass production. In the short term, alkaline electrolysis will continue to
account for the largest share of electrolysis capacities, but PEM electrolysis is
expected to become increasingly significant, especially in the case of small and
medium power outputs.
In the medium to long term, high-temperature electrolysis will also become es-
tablished on the market, provided that production costs can be drastically reduced,
and the service life continues to increase. The pure material costs are significantly
lower than PEM electrolysis. Especially in applications where existing waste heat
can be used, the advantages of this technology in terms of efficiency will come
to the fore. Alkaline membrane electrolysis remains another pertinent option. This
combines the advantages of classical alkaline electrolysis with the compactness
and improved efficiency of PEM electrolysis. However, some developments and
cost reductions are still needed in this regard to make the technology competitive.
Blue and turquoise hydrogen act as bridging technologies and are expected to
play a greater role abroad (at coal and natural gas sites as well as CO2 or carbon
sinks). They will fulfill their role as hydrogen imports in Germany on a temporary
basis, but the operation of such plants in Germany seems unlikely. Photocatalytic
and biological processes are still at an early stage of development and cannot yet
be fully evaluated in terms of their potential marketability.
In the near future, hydrogen will play an increasing role in the energy system in
Germany, especially in sector coupling. On the consumer side, mobility (air, ship,
heavy goods traffic) and sustainable process routes will be the biggest drivers in
industry. In industry, a reduction in CO2 emissions can be achieved in the short
term by replacing gray hydrogen with green hydrogen in refineries and ammonia
production. The challenges and opportunities of the industrial use of hydrogen are
described in more detail in Chap. 5.
However, the increasing demand for hydrogen will still largely have to be met
through gray hydrogen, at least in the early stages. In the medium term, it will
be covered by domestic green hydrogen and hydrogen imports (blue, green and
possibly turquoise). In the long term, the goal will remain to use only CO2 -free,
primarily green hydrogen.
References
1. Herz G., Rix C., Jacobasch E. et al. (2021): Economic assessment of Power-to-Liquid
processes—Influence of electrolysis technology and operating conditions. Applied En-
ergy 292: 116655. https://doi.org/10.1016/j.apenergy.2021.116655
2. van Troostwijk P., A, Deiman J. R. (1790): Beschrijving van een electriseermachine en
van proefnemingen met dezelve in het werk gesteld. Journal de physique 1790, 130–158
3. Nicholson W. (1800): Beschreibung des neuen electrischen oder galvanischen Apparats
Alexander Volta’s, und einiger wichtigen damit angestellten Versuche (Description of
Alexander Volta’s new electronic or galvanic apparatus, and some important experiments
with it). Annalen der Physik (Annals of physics) 6 (3): 340–359. https://doi.org/10.1002/
andp.18000060306
4. Schmidt O. (1900): Über elektrische Wasserzersetzung im Grossen (Electrical decom-
position of water on a large scale). Zeitschrift für Elektrochemie 7 (20): 295–300
5. Kreuter W., Hofmann H. (1998): Electrolysis: The important energy transformer in
a world of sustainable energy. International Journal of Hydrogen Energy 23 (8):
661–666. https://doi.org/10.1016/S0360-3199(97)00109-2
6. Smolinka T., Garche J. (2021): Electrochemical power sources, fundamentals, systems,
and applications. Hydrogen Production by Water Electrolysis. Elsevier, Amsterdam
7. Russell J. H., Nuttall L. J., Fickett A. P. (1973): Hydrogen production by solid polymer
electrolyte water electrolysis. American Chemical Society Division of Fuel Chemistry.
Preprints 18: 24–40
8. Grot W. (1972): Perfluorierte Ionenaustauscher-Membrane von hoher chemischer und
thermischer Stabilität (Perfluorinated ion exchange membrane with high chemical and
thermal stability). Chemie Ingenieur Technik (Chemical engineer technology) 44 (4):
167–169. https://doi.org/10.1002/cite.330440408
9. Nutall L. (1977): Conceptual design of large scale water electrolysis plant using
solid polymer electrolyte technology. International Journal of Hydrogen Energy 2 (4):
395–403. https://doi.org/10.1016/0360-3199(77)90046-5
10. Stucki S., Scherer G.G., Schlagowski S., Fischer E. (1998): PEM water electrolysers:
evidence for membrane failure in 100 kW demonstration plants. Journal of Applied Elec-
trochemistry 28 (10): 1041–1049. https://doi.org/10.1023/A:1003477305336
11. Möbius H.-H. (1997): On the history of solid electrolyte fuel cells. Journal of Solid State
Electrochemistry 1 (1): 2–16. https://doi.org/10.1007/s100080050018
12. Dönitz, W. (1985): High-temperature electrolysis of water vapor—status of develop-
ment and perspectives for application. International Journal of Hydrogen Energy 10 (5):
291–295. https://doi.org/10.1016/0360-3199(85)90181-8
13. Winter C., Fuchs M. (1991): HYSOLAR and Solar-Wasserstoff-Bayern. Interna-
tional Journal of Hydrogen Energy 16 (11): 723–734. https://doi.org/10.1016/0360-
3199(91)90069-U
14. Brookes H.C., Carruthers C.M., Doyle T.B. (2005): The electrochemical and electro-
catalytic behaviour of glassy metals. Journal of Applied Electrochemistry 35 (9):
903–913. https://doi.org/10.1007/s10800-005-4726-5
15. Crnkovic F.C., Machado S.A.S., Avaca L.A. (2004): Electrochemical and morphological
studies of electrodeposited Ni–Fe–Mo–Zn alloys tailored for water electrolysis. Inter-
References 247
national Journal of Hydrogen Energy 29 (3): 249–254. https://doi.org/10.1016/S0360-

3199(03)00212-X
16. Huot J.-Y., Trudeau M., Brossard L., Schulz, R. (1989): Electrochemical and Electro-
catalytic Behavior of an Iron-Base Amorphous Alloy in Alkaline Solutions at 70 °C.
Journal of the Electrochemical Society 136 (8): 2224–2230. https://doi.org/10.1149/1.
2097269
17. Jakšic J., Vojnovic M., Krstajic N. (2000): Kinetic analysis of hydrogen evolution at
Ni–Mo alloy electrodes. Electrochimica Acta 45 (25–26): 4151–4158. https://doi.org/
10.1016/S0013-4686(00)00549-1
18. Jothi V.R., Karuppasamy K., Maiyalagan T. et al. (2020): Water Splitting: Corrosion and
Alloy Engineering in Rational Design of High Current Density Electrodes for Efficient
Water Splitting. Advanced Energy Materials. 10 (24): 2070107. https://doi.org/10.1002/
aenm.202070107.
19. Krstajic N., Jovic V., Gajickrstajic L. et al. (2008): Electrodeposition of Ni–Mo alloy
coatings and their characterization as cathodes for hydrogen evolution in sodium hy-
droxide solution. International Journal of Hydrogen Energy 33 (14): 3676–3687. https://
doi.org/10.1016/j.ijhydene.2008.04.039
20. Schalenbach M., Speck F.D., Ledendecker M. et al. (2018): Nickel-molybdenum alloy
catalysts for the hydrogen evolution reaction: Activity and stability revised. Electrochim-
ica Acta 259: 1154–1161. https://doi.org/10.1016/j.electacta.2017.11.069
21. Schmidt A., Schubert T., Röntzsch L. et al. (2012): Rapidly solidified Fe-base alloys as
electrode materials for water electrolysis. International Journal of Materials Research
103 (9): 1155–1158. https://doi.org/10.3139/146.110804
22. Vračar L., Conway B.E. (1990): Electroactive intermediate states in anodic O2 evolution
at NiMo-containing glassy metals. Electrochimica Acta 35 (11–12): 1919–1931. https://
doi.org/10.1016/0013-4686(90)87101-7
23. Wang S.L., Kim D.-H., Yi S. (2011): Electrocatalytic properties of Fe-based bulk metal-
lic glasses for hydrogen evolution reaction. Korean Journal of Chemical Engineering. 28
(8): 1672–1676. https://doi.org/10.1007/s11814-011-0034-4
24. Pletcher D., Li X. (2011): Prospects for alkaline zero gap water electrolysers for hy-
drogen production. International Journal of Hydrogen Energy 36 (23): 15089–15104.
https://doi.org/10.1016/j.ijhydene.2011.08.080
25. Zeng K., Zhang D. (2010): Recent progress in alkaline water electrolysis for hydro-
gen production and applications. Progress in Energy and Combustion Science 36 (3):
307–326. https://doi.org/10.1016/j.pecs.2009.11.002
26. Castro, E. (2004): Electrochemical characterization of porous nickel-cobalt oxide elec-
trodes. International Journal of Hydrogen Energy 29 (3): 255–261. https://doi.org/10.
1016/S0360-3199(03)00133-2
27. Rosalbino F., Delsante S., Borzone G., Scavino G. (2013): Electrocatalytic activity of
crystalline Ni–Co–M (M = Cr, Mn, Cu) alloys on the oxygen evolution reaction in an
alkaline environment. International Journal of Hydrogen Energy 38 (25): 10170–10177.
28. Xiao C., Li Y., Lu X., Zhao C. (2016): Bifunctional Porous NiFe/NiCo2O4/Ni Foam
Electrodes with Triple Hierarchy and Double Synergies for Efficient Whole Cell Water
Splitting. Advanced Functional Materials. 26 (20): 3515–3523. https://doi.org/10.1002/
adfm.201505302.
29. Bernäcker C.I., Rauscher T., Büttner T. et al. (2019): A Powder Metallurgy Route to
Produce Raney-Nickel Electrodes for Alkaline Water Electrolysis. Journal of the Elec-
trochemical Society 166 (6): F357–F363. https://doi.org/10.1149/2.0851904jes
30. Rauscher T., Bernäcker C.I., Mühle U. et al. (2019): The effect of Fe as constituent
in Ni-base alloys on the oxygen evolution reaction in alkaline solutions at high current
densities. International Journal of Hydrogen Energy 44 (13): 6392–6402. https://doi.org/
10.1016/j.ijhydene.2019.01.182
31. Kiemel S., Smolinka T., Lehner F. et al. (2021): Critical materials for water electrolysers
at the example of the energy transition in Germany. International Journal of Energy
Research. https://doi.org/10.1002/er.6487
32. Merle G., Wessling M., Nijmeijer K. (2011): Anion exchange membranes for alkaline
fuel cells: A review. Journal of Membrane Science 377 (1–2): 1–35. https://doi.org/10.
1016/j.memsci.2011.04.043
33. Varcoe J.R., Atanassov P., Dekel D.R. et al. (2014): Anion-exchange membranes in
electrochemical energy systems. Energy & Environmental Science 7 (10): 3135–3191.
https://doi.org/10.1039/C4EE01303D
34. Maurya S., Shin S.-H., Kim Y., Moon S.-H. (2015): A review on recent developments
of anion exchange membranes for fuel cells and redox flow batteries. RSC Advances. 5
(47): 37206–37230. https://doi.org/10.1039/C5RA04741B
35. Pan J., Chen C., Li Y. et al. (2014): Constructing ionic highway in alkaline polymer
electrolytes. Energy & Environmental Science 7 (1): 354–360. https://doi.org/10.1039/
C3EE43275K
36. Hnát J., Paidar M., Schauer J. et al. (2011): Polymer anion selective membranes for
electrolytic splitting of water. Part I: Stability of ion-exchange groups and impact of the
polymer binder. Journal of Applied Electrochemistry 41 (9): 1043–1052. https://doi.org/
10.1007/s10800-011-0309-9
37. Bauer B., Strathmann H., Effenberger F. (1990): Anion-exchange membranes with im-
proved alkaline stability. Desalination 79 (2–3): 125–144. https://doi.org/10.1016/0011-
9164(90)85002-R
38. Pavel C.C., Cecconi F., Emiliani C. et al. (2014): Highly efficient platinum group metal
free based membrane-electrode assembly for anion exchange membrane water electrol-
ysis. Angewandte Chemie (Applied chemistry) (International ed. in English) 53 (5):
1378–1381. https://doi.org/10.1002/anie.201308099
39. Faraj M., Boccia M., Miller H. et al. (2012): New LDPE based anion-exchange mem-
branes for alkaline solid polymeric electrolyte water electrolysis. International Journal
of Hydrogen Energy 37 (20): 14992–15002. https://doi.org/10.1016/j.ijhydene.2012.08.
012
40. Xing B., Savadogo O. (2000): Hydrogen/oxygen polymer electrolyte membrane fuel
cells (PEMFCs) based on alkaline-doped polybenzimidazole (PBI). Electrochemistry
Communications 2 (10): 697–702. https://doi.org/10.1016/S1388-2481(00)00107-7
41. Zarrin H., Jiang G., Lam G.Y.-Y. et al. (2014): High performance porous polybenzim-
idazole membrane for alkaline fuel cells. International Journal of Hydrogen Energy 39
(32): 18405–18415. https://doi.org/10.1016/j.ijhydene.2014.08.134
42. Kraglund M.R., Carmo M., Schiller G., Ansar S.A., Aili D., Christensen E., Jensen J.O.
(2019): Ion-solvating membranes as a new approach towards high rate alkaline elec-
References 249
trolyzers. Energy & Environmental Science 12 (11): 3313–3318. https://doi.org/10.1039/

C9EE00832B
43. Miller H.A., Bouzek K., Hnat J. et al. (2020): Green hydrogen from anion exchange
membrane water electrolysis: a review of recent developments in critical materials and
operating conditions. Sustainable Energy Fuels 4 (5): 2114–2133. https://doi.org/10.
1039/C9SE01240K
44. Chen D., Chen C., Baiyee Z. M. et al. (2015): Nonstoichiometric Oxides as Low-Cost
and Highly-Efficient Oxygen Reduction/Evolution Catalysts for Low-Temperature Elec-
trochemical Devices. Chemical reviews 115 (18): 9869–9921. https://doi.org/10.1021/
acs.chemrev.5b00073
45. Li X., Walsh F.C., Pletcher D. (2011): Nickel based electrocatalysts for oxygen evolu-
tion in high current density, alkaline water electrolysers. Physical chemistry chemical
physics: PCCP 13 (3): 1162–1167. https://doi.org/10.1039/C0CP00993H
46. Lim C.S., Chua C.K., Sofer, Z. et al. (2015): Layered transition metal oxyhydroxides as
tri-functional electrocatalysts. Journal of Materials Chemistry. A 3 (22): 11920–11929.
https://doi.org/10.1039/C5TA02063H
47. Fawcett W.R. (2020): Interfacial Electrochemistry. In: Wieckowski A. (Ed.): The-
ory, Experiment, and Application. Marcel Dekker: New York. https://doi.org/10.1021/
ja004720h
48. Safizadeh F., Ghali E., Houlachi G. (2015): Electrocatalysis developments for hydrogen
evolution reaction in alkaline solutions—A Review. International Journal of Hydrogen
Energy 40 (1): 256–274. https://doi.org/10.1016/j.ijhydene.2014.10.109
49. Liu Z., Sajjad S.D., Gao Y. (2017): The effect of membrane on an alkaline water elec-
trolyzer. International Journal of Hydrogen Energy 42 (50): 29661–29665. https://doi.
org/10.1016/j.ijhydene.2017.10.050
50. Carmo M., Fritz D.L., Mergel J., Stolten D. (2013): A comprehensive review on
PEM water electrolysis. International Journal of Hydrogen Energy 38 (12): 4901–4934.
51. Su H., Bladergroen B.J., Pasupathi S. et al. (2012): Performance Investigation of Mem-
brane Electrode Assemblies for Hydrogen Production by Solid Polymer Electrolyte
Water Electrolysis. International Journal of Electrochemical Science 7 (5): 4223–4234
52. Siegmund D., Metz S., Peinecke V. et al. (2021): Crossing the Valley of Death: From
Fundamental to Applied Research in Electrolysis. JACS Au 1 (5): 527–545. https://doi.
org/10.1021/jacsau.1c00092
53. Lickert T., Kiermaier M.L., Bromberger K. et al. (2020): On the influence of the anodic
porous transport layer on PEM electrolysis performance at high current densities. In-
ternational Journal of Hydrogen Energy 45 (11): 6047–6058. https://doi.org/10.1016/j.
ijhydene.2019.12.204
54. Smolinka T. (2021): Grundlagen der Wasserelektrolyse zur Wasserstofferzeugung (Ba-
sics of water electrolysis for hydrogen production). In: Graf F., Schoof R., Zdrallek M.
(Ed.): Power-to-Gas: Grundlagen—Konzepte—Lösungen (Power-to-gas: fundamentals,
concepts, solutions). Vulkan Verlag, Essen
55. Kusnezoff M., Beckert W., Trofimenko N. et al. (2015): Electrochemical MEA Charac-
terization: Area Specific Resistance Corrected to Fuel Utilization as Universal Charac-
teristic for Cell Performance. ECS Transactions. 68 (1): 2555–2563. https://doi.org/10.
1149/06801.2555ecst
56. Ge B., Ma J. T., Ai D., Deng C., Lin X., Xu J. (2015): Sr2FeNbO6 Applied in Solid Ox-
ide Electrolysis Cell as the Hydrogen Electrode: Kinetic Studies by Comparison with
Ni-YSZ. Electrochimica Acta 151, 437–446. https://doi.org/10.1016/j.electacta.2014.
11.078
57. Borglum B., Tang E., Pastula M. (2011): Development of Solid Oxide Fuel Cells at Versa
Power Systems. ECS Transactions. 35 (1): 63–69. https://doi.org/10.1149/1.3569979
58. Ghezel-Ayagh H., Borglum B.P. (2017): Review of Progress in Solid Oxide Fuel Cells
at FuelCell Energy. ECS Transactions. 78 (1): 77–86. https://doi.org/10.1149/07801.
0077ecst
59. Geipel C., Hauptmeier K., Herbrig K. et al. (2019): Stack Development and Industrial
Scale-Up. ECS Transactions. 91 (1): 123–132. https://doi.org/10.1149/09101.0123ecst
60. Santhanam S., Ullmer D., Wuillemin Z. (2019): Experimental Analysis of a 25 kWe
Solid Oxide Fuel Cell Module for Co-Generation of Hydrogen and Power. ECS Trans-
actions. 91 (1): 159–166. https://doi.org/10.1149/09101.0159ecst
61. Hartvigsen J., Elangovan S., Frost, L. (2019): OxEon Energy Demonstration of Manned-
Mission Scale ISRU Process Systems. ICES-2019-257, 49th International Conference
on Environmental Systems: Boston, Massachusetts 2019
62. Noponen M., Torri P., Göös J. et al. (2020): Progress of SOC development at Elcogen.
A0304. In: Proceedings of 14th European SOFC & SOE Forum 2019. European Fuel
Cell Forum AG
63. Posdziech O., Schwarze K., Brabandt J. (2019): Efficient hydrogen production for in-
dustry and electricity storage via high-temperature electrolysis. International Journal
of Hydrogen Energy 44 (35): 19089–19101. https://doi.org/10.1016/j.ijhydene.2018.05.
169
64. Schwarze K., Posdziech O., Mermelstein J., Kroop S. (2018): Operational Results of an
150/30 kW RSOC System in an Industrial Environment. A0604. In: Proceedings 13th
European SOFC & SOE Forum 2018. European Fuel Cell Forum AG
65. Peters R., Tiedemann W., Hoven I. et al. (2020): Operating Experience with a 5 kW-class
Reversible Solid Oxide Cell System. A0803. In: Proceedings of 14th European SOFC &
SOE Forum 2019. European Fuel Cell Forum AG
66. Küngas R., Blennow P., Heiredal-Clausen et al. (2020): Power-to-X activities at Haldor
Topsoe: our approach for electrification of the chemicals industry. A0305. In: Proceed-
ings of 14th European SOFC & SOE Forum 2019. European Fuel Cell Forum AG
67. Herz G., Reichelt E., Jahn M. (2018): Techno-economic analysis of a co-electrolysis-
based synthesis process for the production of hydrocarbons. Applied Energy 215,
309–320. https://doi.org/10.1016/j.apenergy.2018.02.007
68. Meng Y., Gao J., Zhao Z. et al. (2019): Review: recent progress in low-temperature
proton-conducting ceramics. Journal of Materials Science 54 (13): 9291–9312. https://
doi.org/10.1007/s10853-019-03559-9
69. Leonard K., Okuyama Y., Takamura Y. et al. (2018): Efficient intermediate-temperature
steam electrolysis with Y:SrZrO3 –SrCeO3 and Y:BaZrO3 –BaCeO3 proton conducting
perovskites. Journal of Materials Chemistry A 6 (39): 19113–19124. https://doi.org/10.
1039/c8ta04019b
70. Bi L., Fabbri E., Sun Z., Traversa E. (2011): Sinteractive anodic powders improve
densification and electrochemical properties of BaZr0.8 Y0.2 O3 -• electrolyte films for
References 251
anode-supported solid oxide fuel cells. Energy & Environmental Science 4 (4): 1352.
https://doi.org/10.1039/c0ee00387e
71. Vøllestad E., Strandbakke R., Tarach M. et al. (2019): Mixed proton and electron
conducting double perovskite anodes for stable and efficient tubular proton ceramic elec-
trolysers. Nature materials 18 (7): 752–759. https://doi.org/10.1038/s41563-019-0388-
2
72. Nakamura T., Mizunuma S., Kimura Y. et al. (2018): Energy efficiency of ionic transport
through proton conducting ceramic electrolytes for energy conversion applications. Jour-
nal of Materials Chemistry A 6 (32): 15771–15780. https://doi.org/10.1039/c8ta05373a
73. Leonard K., Lee Y.-S., Okuyama Y., Miyazaki K., Matsumoto H. (2016): Proton Con-
duction Properties of Bzcy and Szcy Perovskites Type Oxides. Addressing the Effect of
Dopant Levels and Proton Content. Meeting Abstracts [Online]
74. Braun R. J., Dubois A., Ferguson K. et al. (2019): Development of kW-Scale Protonic
Ceramic Fuel Cells and Systems. ECS Transactions 91 (1): 997–1008. https://doi.org/
10.1149/09101.0997ecst
75. Antonini C., Treyer K., Streb A., van der Spek M., Bauer C., Mazzotti M. (2020): Hy-
drogen production from natural gas and biomethane with carbon capture and storage—A
techno-environmental analysis. Sustainable Energy Fuels 4 (6): 2967–2986. https://doi.
org/10.1039/D0SE00222D
76. Nikolaidis P., Poullikkas A. (2017): A comparative overview of hydrogen production
processes. Renewable and Sustainable Energy Reviews 67: 597–611. https://doi.org/10.
1016/j.rser.2016.09.044
77. van de Krol R., Grätzel M. (2012): Photoelectrochemical Hydrogen Production 102.
Springer US: Boston, MA
78. Ahmet I.Y., Ma Y., Jang J.-W. et al. (2019): Demonstration of a 50 cm2 BiVO4 tandem
photoelectrochemical-photovoltaic water splitting device. Sustainable Energy Fuels 3
(9), 2366–2379. https://doi.org/10.1039/C9SE00246D
79. Moss B., Babacan O., Kafizas A., Hankin A. (2021): A Review of Inorganic Photo-
electrode Developments and Reactor Scale-Up Challenges for Solar Hydrogen Pro-
duction. Advanced Energy Materials. 11 (13), 2003286. https://doi.org/10.1002/aenm.
202003286
80. Cheng W.-H., Richter M.H., May M.M. et al. (2018): Monolithic Photoelectrochem-
ical Device for Direct Water Splitting with 19% Efficiency. ACS Energy Lett. 3 (8),
1795–1800. https://doi.org/10.1021/acsenergylett.8b00920
81. Döscher H., Geisz J.F., Deutsch T.G., Turner J.A. (2014): Sunlight absorption in wa-
ter—efficiency and design implications for photoelectrochemical devices. Energy &
Environmental Science 7 (9), 2951–2956. https://doi.org/10.1039/C4EE01753F
82. Kircher M., Schwarz T. (Ed.) (2020): CO2 und CO—Nachhaltige Kohlenstoffquellen für
die Kreislaufwirtschaft (CO2 and CO—Sustainable sources of carbon for the circular
economy). Springer, Berlin, Heidelberg
83. Rey F.E., Heiniger E.K., Harwood C.S. (2007): Redirection of metabolism for biologi-
cal hydrogen production. Applied and environmental microbiology 73 (5): 1665–1671.
https://doi.org/10.1128/AEM.02565-06
84. Marx C., Happe T. (2020): Alternative biologische und biotechnologische Verfahren zur
Wasserstoffherstellung (Alternative biological and biotechnological processes for hydro-
gen production). In: Kircher M., Schwarz T. (Ed.): CO2 und CO—Kohlenstoffquellen
für die Kreislaufwirtschaft (CO2 and CO—Sustainable sources of carbon for the cir-
cular economy). Springer, Berlin Heidelberg, p. 39–58. https://doi.org/10.1007/978-3-
662-60649-0_4
85. Lubitz W., Ogata H., Rüdiger O., Reijerse E. (2014): Hydrogenases. Chemical reviews
114 (8): 4081–4148. https://doi.org/10.1021/cr4005814
86. Kleinhaus J.T., Wittkamp F., Yadav S., Siegmund D., Apfel U.-P. (2021): FeFe-
Hydrogenases: maturation and reactivity of enzymatic systems and overview of
biomimetic models. Chemical Society reviews 50 (3): 1668–1784. https://doi.org/10.
1039/d0cs01089h
87. Trchounian K., Poladyan A., Vassilian A., Trchounian A. (2012): Multiple and reversible
hydrogenases for hydrogen production by Escherichia coli: dependence on fermentation
substrate, pH and the F(0)F(1)-ATPase. Critical reviews in biochemistry and molecular
biology 47 (3): 236–249. https://doi.org/10.3109/10409238.2012.655375
88. Li C., Fang H.H.P. (2007): Fermentative Hydrogen Production From Wastewater and
Solid Wastes by Mixed Cultures. Critical Reviews in Environmental Science and Tech-
nology 37 (1): 1–39. https://doi.org/10.1080/10643380600729071
89. Kim D.-H., Lee J.-H., Kang S. et al. (2014): Enhanced photo-fermentative H2 production
using Rhodobacter sphaeroides by ethanol addition and analysis of soluble microbial
products. Biotechnology for biofuels 7: 79. https://doi.org/10.1186/1754-6834-7-79
90. Lopespinto, F. (2002): A brief look at three decades of research on cyanobacterial hy-
drogen evolution. International Journal of Hydrogen Energy 27 (11–12): 1209–1215.
https://doi.org/10.1016/S0360-3199(02)00089-7
91. Weigand W., Apfel U.-P. (Ed.) (2018): Bioorganometallic Chemistry. De Gruyter, Berlin
92. Ghirardi M.L., Togasaki R.K., Seibert M. (1997): Oxygen sensitivity of algal H2-
production. Applied biochemistry and biotechnology 63–65: 141–151. https://doi.org/
10.1007/BF02920420
93. Ozturk Y., Yucel M., Daldal F. et al. (2006): Hydrogen production by using Rhodobacter
capsulatus mutants with genetically modified electron transfer chains. International Jour-
nal of Hydrogen Energy 31 (11): 1545–1552. https://doi.org/10.1016/j.ijhydene.2006.
06.042
94. Plumeré N., Rüdiger O., Oughli A.A. (2014): A redox hydrogel protects hydrogenase
from high-potential deactivation and oxygen damage. Nature chemistry 6 (9): 822–827.
https://doi.org/10.1038/nchem.2022
95. Happe T., Schütz K., Böhme H. (2000): Transcriptional and mutational analysis of
the uptake hydrogenase of the filamentous cyanobacterium Anabaena variabilis ATCC
29413. Journal of bacteriology 182 (6): 1624–1631. https://doi.org/10.1128/jb.182.6.
1624-1631.2000

Producing Hydrogen Through Electrolysis and Other Processes

Uploaded by

Copyright:

Available Formats

Producing Hydrogen Through Electrolysis and Other Processes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Producing Hydrogen Through Electrolysis and Other Processes

Uploaded by

Copyright:

Available Formats

Producing hydrogen through electrolysis

and other processes 9

In addition to expanding renewable energy sources and implementing electrical

© Springer Nature Switzerland AG 2022 203

as well as in the heating sector to a certain degree. It will only be possible to

9.1 Hydrogen production processes

Reforming natural gas with steam

Reducing greenhouse gas

Thermal breakdown ofmethane (pyrolysis) with

Water electrolysis with electricity from

found in biotechnological processes and strategies for obtaining solar hydrogen

9.2 Hydrogen production by electrolysis

9.2.1 Fundamentals of water electrolysis

AEMEL 50–70 °C 2H2 O C 2e ! H2 C 2OH OH 2OH ! 12 O2 C H2 O C 2e

PEMEL 50–80 °C 2H+ C 2e ! H2 H+ H2 O ! 12 O2 C 2H+ C 2e

PCCEL 450–600 °C 2H+ C 2e ! H2 H+ H2 O ! 12 O2 C 2H+ C 2e

SOEL 650–850 °C H2 O C 2e ! H2 C O2 O2 O2 ! 12 O2 C 2e

CoSOEL 700–900 °C H2 O C 2e ! H2 C O2 O2 O2 ! 12 O2 C 2e

the heating value of hydrogen. According to the Gibbs-Helmholtz equation, the

HR D GR C TSR (9.2)

25 percent, with the theoretical decomposition voltage V rev dropping to values of

GR .p; T; H2 Ox/ RT xH2 O

In addition, additional loss mechanisms occur when a current flow is applied,

9.2.2 Alkaline water electrolysis

In addition to the peripheral systems (gas drying, compressors, pumps, rectifiers,

Fig. 9.7 Basic structure of an alkaline electrolysis cell. (Fraunhofer ISE)

an aqueous solution, usually of 30% KOH, is used as the electrolyte at typical

The coating of 3D substrate structures with electrocatalytically active layers

a After heat treatment b After leaching

9.2.3 PEM electrolysis

dred micrometers to approx. 1 mm) through milling, punching, deep drawing or

9.2.4 AEM electrolysis

Two technologies are currently used in the classical production of hydrogen

9.2.5 High-temperature steam electrolysis

High-temperature electrolysis (HTEL) is generally understood to mean electroly-

Fig. 9.16 Schematic repre-

Fig. 9.17 Components and structure of an SOC stack. (Fraunhofer IKTS)

Electrolyte supported cells (ESCs) based on a thin electrolyte substrate (60 to

Fig. 9.18 Production steps for an electrolyte-supported cell. (Fraunhofer IKTS)

Fig. 9.19 Production steps for an SOEL stack. (Fraunhofer IKTS)

The production of an HTEL stack is shown schematically in Fig. 9.19. The

9.2.6 Co-electrolysis of water and carbon dioxide

9.2.7 High-temperature ceramic-based proton-conducting

9.3 Other innovative processes for hydrogen production

9.3.1 Steam reforming with carbon capture and utilization

Hydrogen can be obtained from a wide range of hydrogen-containing energy car-

Table 9.2 Equilibrium conversion of methane reforming at a steam-to-carbon (S/C) ratio

hydrogen by the reverse reaction of steam reforming (methanation) and used as

9.3.2 Methane pyrolysis

9.3.3 Photocatalytic systems

In another process, sunlight is used to directly produce green hydrogen by allowing

In order to use carefully coordinated materials, semiconductor materials are being

Fig. 9.25 Test bench for

9.3.4 Biological procedures

In addition to the above-mentioned processes for hydrogen production, hydrogen

Table 9.3 Hydrogen production pathways by microorganism [84]

Although theoretically, the energy conversion efficiency of these hydrogen pro-

9.4 Summary and outlook

national Journal of Hydrogen Energy 29 (3): 249–254. https://doi.org/10.1016/S0360-