![Map showing the 220/380-kV lines of the German high-voltage grid. ª Forum Network Technology/Network Operation in the VDE (FNN) 2016, VDE Association for Electrical, Electronic & Information Technologies [6]. The other power levels are too fragmented to be displayed clearly in this figure. Solid lines: existing power lines; dashed lines: power lines under construction; dotted lines: projected power lines; magenta lines: high voltage direct current power lines; filled dots: power transformation stations; open dots: large cities.](https://www.researchgate.net/publication/318179264/figure/fig21/AS:668322769166338@1536351973240/Map-showing-the-220-380-kV-lines-of-the-German-high-voltage-grid-Forum-Network_Q320.jpg)
![. Network-attached storage; projects under construction/planned in parentheses [10].](https://www.researchgate.net/publication/318179264/figure/tbl1/AS:668322773352452@1536351974591/Network-attached-storage-projects-under-construction-planned-in-parentheses-10_Q320.jpg)
![Storage locations for natural gas, crude oil products, and liquefied gas in Germany; ª State Office for Mining, Energy and Geology [10]. Natural gas is stored in pore storage (circles) which are marked as operational (red) or planed (yellow) with their respective capacity in Mio. m 3 (Vn); natural gas cavern storage is marked by elliptical symbols with the same color coding; cavern storage for mineral oil, oil-based products and liquefied natural gas are marked by green square symbols; the number indicates the number of individual caverns.](https://www.researchgate.net/publication/318179264/figure/fig1/AS:513474339053568@1499433230100/Storage-locations-for-natural-gas-crude-oil-products-and-liquefied-gas-in-Germany_Q320.jpg)
Available via license: CC BY-NC 4.0
Content may be subject to copyright.
Energy Storage as Part of a Secure Energy Supply
Florian Ausfelder
[1,],
*, Christian Beilmann
[2]
, Martin Bertau
[3]
, Sigmar Bra
¨uninger
[4]
, Angelika Heinzel
[5]
,
Renate Hoer
[6]
, Wolfram Koch
[6]
, Falko Mahlendorf
[5]
, Anja Metzelthin
[7]
, Marcell Peuckert
[8]
, Ludolf Plass
[9]
,
Konstantin Ra
¨uchle
[3]
, Martin Reuter
[10]
, Georg Schaub
[11]
, Sebastian Schiebahn
[12]
, Ekkehard Schwab
[4]
,
Ferdi Schu
¨th
[13]
, Detlef Stolten
[14]
, Gisa Teßmer
[15]
, Kurt Wagemann
[1]
, Karl-Friedrich Ziegahn
[7,16]
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 144
Abstract
The current energy system is subject to a fundamental trans-
formation: A system that is oriented towards a constant
energy supply by means of fossil fuels is now expected to
integrate increasing amounts of renewable energy to achieve
overall a more sustainable energy supply. The challenges
arising from this paradigm shift are currently most obvious
in the area of electric power supply. However, it affects all
areas of the energy system, albeit with different results. With-
in the energy system, various independent grids fulfill the
function of transporting and spatially distributing energy or
energy carriers, and the demand-oriented supply ensures
that energy demands are met at all times. However, renew-
able energy sources generally supply their energy indepen-
dently from any specific energy demand. Their contribution
to the overall energy system is expected to increase signifi-
cantly. Energy storage technologies are one option for tem-
poral matching of energy supply and demand. Energy sto-
rage systems have the ability to take up a certain amount of
energy, store it in a storage medium for a suitable period of
time, and release it in a controlled manner after a certain
time delay. Energy storage systems can also be constructed as
process chains by combining unit operations, each of which
cover different aspects of these functions. Large-scale
mechanical storage of electric power is currently almost
exclusively achieved by pumped-storage hydroelectric power
stations. These systems may be supplemented in the future
by compressed-air energy storage and possibly air separation
plants. In the area of electrochemical storage, various tech-
nologies are currently in various stages of research, develop-
ment, and demonstration of their suitability for large-scale
electrical energy storage. Thermal energy storage technolo-
gies are based on the storage of sensible heat, exploitation of
phase transitions, adsorption/desorption processes, and
chemical reactions. The latter offer the possibility of perma-
nent and loss-free storage of heat. The storage of energy in
chemical bonds involves compounds that can act as energy
carriers or as chemical feedstocks. Thus, they are in direct
economic competition with established (fossil fuel) supply
routes. The key technology here – now and for the foresee-
able future – is the electrolysis of water to produce hydrogen
and oxygen. Hydrogen can be transformed by various pro-
cesses into other energy carriers, which can be exploited in
different sectors of the energy system and/or as raw materials
for energy-intensive industrial processes. Some functions of
energy storage systems can be taken over by industrial pro-
cesses. Within the overall energy system, chemical energy
storage technologies open up opportunities to link and inter-
weave the various energy streams and sectors. Chemical
energy storage not only offers means for greater integration
of renewable energy outside the electric power sector, it also
creates new opportunities for increased flexibility, novel
synergies, and additional optimization. Several examples of
specific energy utilization are discussed and evaluated with
respect to energy storage applications.
The article describes various technologies for energy storage
and their potential applications in the context of Germany’s
Energiewende, i.e. the transition towards a more sustainable
energy system. Therefore, the existing legal framework
defines some of the discussions and findings within the
article, specifically the compensation for renewable electri-
city providers defined by the German Renewable Energy
Sources Act, which is under constant reformation. While the
article is written from a German perspective, the authors
hope this article will be of general interest for anyone
working in the areas of energy systems or energy technology.
Keywords: Energy storage technology, Energy supply, Optimization
Received: March 09, 2017; accepted: March 10, 2017
DOI: 10.1002/cben.201700004
{
ª2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA. This is an open access article under the terms of the Creative Commons
Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly
cited and is not used for commercial purposes.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 145
Table of Contents
1 Introduction 2
Demands on the Energy Supply 2
2 Overall Concept of the Energy System 3
2.1 Structure of the German Energy System 3
2.2 Integration of Renewable Energies into the Energy Supply 3
2.3 Future Structure and Challenges of the Energy Supply 3
2.4 European Perspective 4
2.5 Definition of Energy Storage 5
2.6 Function and Contribution of Power Grids to a Secure
Energy Supply 5
3 Evaluation Criteria and Scenarios for the Integrated Use of
Energy Storage Technologies 12
3.1 Tasks and Functions of Energy Storage 12
3.2 Energetic Potential Analysis 14
3.3 Economic Evaluation 16
4 Storage Technologies 17
4.1 Technologies for Storage as Mechanical Energy 17
4.2 Technologies for the Storage of Energy in the Form of
Chemical Energy 22
4.3 Technologies for Storage in the Form of Electrochemical
Energy 36
4.4 Technologies for Energy Storage in the Form of Thermal
Energy 41
4.5 Providing Flexibility through Energy-Intensive Processes as
Storage Systems 46
4.6 Storage as a Grid Service Provider and Grid Transfer
Technologies 48
5 Reference Cases for the Use of Energy Storage 50
5.1 Electricity Supply of a Household through Photovoltaics 50
5.2 Industrial Site 55
5.3 Seasonal Balancing Storage through Power to Gas 57
6 Conclusion and Outlook 62
Acknowledgment 63
References 63
1 Introduction
1.1 Demands on the Energy Supply
A secure energy supply is one of the basic requirements of a
modern economy. The goal is to ensure an affordable and reli-
able energy supply for residential, commercial, and industrial
consumers. A modern energy system should minimize the
emission of greenhouse gases and through its price structure
also be attractive for energy-intensive companies in inter-
national competition.
These requirements result in a series of measures to address
the challenges. For the smoothest possible supply of energy, the
provision and demand-oriented expansion of cost-effective,
high-performance, and efficient energy infrastructure, i.e., pri-
marily the supply grids, are indispensable. The strategic
reserves of conventional fuels are able to cover short- and
medium-term supply shortages and guarantee stable primary
supply. Diversification of energy sources, carriers, and pro-
viders makes the energy supply more resistant to political
changes. The increased use of renewable energy leads to an
even more sustainable energy supply due to the lower con-
sumption of fossil raw materials and the consequent reduction
in emissions of greenhouse gases.
Energy supply and consumption must be balanced spatially
and temporally. The former is achieved through transport,
transmission, and distribution grids, which ensure demand-
oriented, comprehensive, and trouble-free supply of consumers
with the required energy or energy carrier. The grids with their
international connections allow demand and supply to be
balanced by importing missing and exporting excess energy.
—————
[1]
Dr. Florian Ausfelder (corresponding author), Prof. Dr. Kurt Wage-
mann
DECHEMA Gesellschaft fu¨ r Chemische Technik und Biotechno-
logie e.V., Theodor-Heuss-Allee 25, 60486 Frankfurt am Main,
Germany.
E-Mail: ausfelder@dechema.de
[2]
Dr. Christian Beilmann
Helmholtz Gemeinschaft, Anna-Louisa-Karsch Strasse 2, 10178
Berlin, Germany.
[3]
Prof. Dr. Martin Bertau, Dr. Konstantin Ra
¨uchle
Technische Universita
¨t Bergakademie Freiberg, Akademiestrasse
6, 09599 Freiberg, Germany.
[4]
Dr. Sigmar Bra
¨uninger, Dr. Ekkehard Schwab
BASF SE, Carl-Bosch-Strasse 38, 67063 Ludwigshafen, Germany.
[5]
Prof. Dr. Angelika Heinzel, Dr. Falko Mahlendorf
Universita
¨t Duisburg-Essen, Forsthausweg 2, 47057 Duisburg,
Germany.
[6]
Dr. Renate Hoer, Prof. Dr. Wolfram Koch
GDCh Gesellschaft Deutscher Chemiker e.V., Varrentrappstrasse
40–42, 60486 Frankfurt am Main, Germany.
[7]
Dr. Anja Metzelthin, Dr.-Ing. Karl-Friedrich Ziegahn
Deutsche Physikalische Gesellschaft e.V., Hauptstrasse 5, 53604
Bad Honnef, Germany.
[8]
Prof. Dr. Marcell Peuckert
Katzenlu¨ ckstrasse 11, 65719 Hofheim am Taunus, Germany.
[9]
Dr. Ludolf Plass
Parkstrasse 11a, 61476 Kronberg im Taunus, Germany.
[10]
Dr. Martin Reuter
VCI Verband der Chemischen Industrie e. V., Mainzer Landstrasse
55, 60329 Frankfurt am Main, Germany.
[11]
Prof. Dr.-Ing. Georg Schaub
Karlsruher Institut fu¨ r Technologie, Engler-Bunte-Institut, Engler-
Bunte-Ring 1, 76131 Karlsruhe, Germany.
[12]
Dr. Sebastian Schiebahn
Lyatenstrasse 3, 52382 Niederzier, Germany.
[13]
Prof. Dr. Ferdi Schu¨th
Max-Planck-Institut fu¨ r Kohlenforschung, Kaiser-Wilhelm-Platz 1,
45470 Mu¨ lheim an der Ruhr, Germany.
[14]
Prof. Dr.-Ing. Detlef Stolten
Forschungszentrum Ju¨ lich GmbH, Wilhelm-Johnen-Strasse,
52428 Ju¨ lich, Germany.
[15]
Dr. Gisa Teßmer
DGMK Deutsche Wissenschaftliche Gesellschaft fu¨ r Erdo
¨l, Erdgas
und Kohle e.V., U
¨berseering 40 (RWE-Haus), 22297 Hamburg,
Germany.
[16]
Dr.-Ing. Karl-Friedrich Ziegahn
Karlsruher Institut fu¨ r Technologie, Campus Nord, Hermann-von-
Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany.
{
Updated and complemented English version of DOI: 10.1002/
cite.201400183
While grids ensure the local distribution of energy or fuels,
energy storage systems contribute to making energy available
on demand. They can temporarily absorb excess energy or
energy sources and make them available again later. A tempo-
rary excess of one form of energy can be transformed into
another, which can then be used or stored. However, conver-
sion losses must be taken into account.
The existing energy storage systems are sufficient for the cur-
rent structure of the energy system. Strategic stockpiling of oil
and gas can compensate temporary bottlenecks in transport,
power plants, and heat supply. Pumped-storage hydroelectric
power stations in combination with flexible operation of power
plants can adequately respond to fluctuations in electricity con-
sumption. However, the increasing expansion of renewable
energy with feed-in priority, the dependency relationships in
the supply of energy sources, and new concepts of mobility
suggest an important role for energy storage systems in the me-
dium term. Energy storage can therefore be characterized as an
important, but not urgent problem. Nevertheless, research and
development are already required now to be prepared for the
time when energy storage is needed on a larger scale.
The fact that we are currently able to cope with the existing
energy storage systems cannot be taken as granted for the
future. Instead, it provides the opportunity and an obligation
to future generations to initiate the necessary developments
with the aim of moving to a sustainable energy system.
Energy storage systems are a possible, and sometimes crucial
option for dealing with a fluctuating energy supply. However,
they are not the only way to respond to fluctuations in supply.
The forecasts for the earnings from renewable energy sources
are becoming more accurate and allow for improved coordina-
tion and more efficient use. Additionally, the demand side can
be addressed to provide a contribution to system stability. The
availability of spare generation capacity is another way to meet
demand even in times of low supply of renewable energy.
Finally, a higher-performance grid can compensate for local
temporal variations by transport to and from other regions.
In the spirit of a forward-looking energy policy that develops
options and is open to flexible implementation, the scientific
and technological development of energy storage systems
should currently be pursued in order to have this option avail-
able when it is needed.
2 Overall Concept of the Energy System
2.1 Structure of the German Energy System
The use of primary energy in Germany covers three major
areas of need. With around 50 % of the final energy consump-
tion, the heating market represents the largest share [1], where-
by mainly natural gas and fuel oil are used as energy sources.
In second place, the use of petroleum derivatives such as gaso-
line, diesel fuel, and kerosene in the transport sector follows. In
third place is the use of electricity, generated by coal, gas,
nuclear and renewable sources. All of these fuels and their
usage forms are made available to the consumer through con-
version and supply systems. Thereby the challenge is to provide
the energy needed in the right form at any time and any place.
Currently, the energy system in Germany is still dominated by
fossil fuels, but the contribution of renewable energy is rising
steadily, especially in the electricity sector.
2.2 Integration of Renewable Energies
into the Energy Supply
Germany places strong emphasis on the development of
renewable energy. The central objective is to build a sustainable
energy system by reducing the emission of greenhouse gases
(especially CO
2
). Furthermore, the dependence on unreliable
suppliers should be reduced and the expected price increases of
fossil fuels should partially be offset. The assumption of in-
creased fossil fuel prices has been the basis of many scenarios
for implementation of renewable energies. However, within re-
cent past, this trend could not be observed and the challenge of
a transformation towards a more renewable energy supply has
likely to be met with low prices of fossil fuels in the future.
In particular in the area of power supply, the increased feed-
in of renewable energies that are not suitable for meeting base-
load requirements leads to increased exchange and a greater
coordination of activities with neighboring countries. Also, the
reduction of emission of greenhouse gases is currently not
achievable because old coal-fired power plants with a relatively
high output of CO
2
can be operated more economically than
modern gas-fired power plants. The reason is the very low
price of CO
2
emissions certificates, currently around
5–7 € t
–1
CO
2
in the European Emissions Trading Scheme
(ETS), and the relatively high price of natural gas. The CO
2
balance of power and fuels generated from renewable resources
is controversial. The relevant renewable energies are listed in
Tab. 1. All renewable energies are subject to geographic restric-
tions, arising either from the energy conversion technology
itself or from its political and social acceptance. Renewable
energy can be used centrally or locally.
2.3 Future Structure and Challenges
of the Energy Supply
The further expansion of renewable energies is the declared
objective of the energy transition. Its realization takes place,
besides fixed quotas for biofuels, mainly in the electricity sector
through guaranteed purchases at fixed prices. On the basis of
the regulations of the German Renewable Energy Sources Act
(Erneuerbare-Energien-Gesetz: EEG), the related costs are, with
some exceptions, transferred to the consumer (EEG surcharge).
The increasing expansion of renewable energies is forcing a
paradigm shift in the traditional power distribution structures.
The existing electricity system is demand-driven and is sup-
ported by fossil and nuclear energy sources in large-scale plants
with their own utilities and consumer structure. These changes
are most clearly evident in the electricity grid. On the genera-
tion side, the classic division into base-load, intermediate, and
peak-load supply is becoming increasingly blurred [2].
At the same time, renewable energies cannot provide the
necessary guaranteed performance (i.e., electricity generation
capacity) that must be available at all times. Owing to the irreg-
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 146
ular contributions of renewable energy from centralized and
decentralized power producers, a greater gap is created between
the temporally and locally available generation capacity and
the consumer demand. This discrepancy must be eliminated in
the most efficient and cost-effective manner.
The heat and mobility sectors (with the exception of electric
mobility) as well as energy-intensive industrial processes are
currently not in the focus of energy policy measures. Never-
theless, it is recognized that only through the inclusion of these
areas can the energy transition be successful, especially with
regard to the reduction of CO
2
emissions.
2.4 European Perspective
Energy supply and energy policy that operate only from a
national perspective are in danger of ignoring the opportunities
that arise through cooperation among national energy systems to
build a greater European energy network. Already today, there
are far-reaching networks of international energy supply struc-
tures. This will further increase in the future, whereby also
national strategic interests will significantly influence the energy
policy in the medium term. These include, e.g., the preference for
domestic energy sources, maintaining a nuclear component,
agreements for the supply of natural gas, and regarding the Ger-
man energy transformation as an exclusively national project.
Indeed, much of the energy infrastructure is already linked
throughout Europe, and isolated decisions and actions have an
impact on the European partner countries and their energy sys-
tems. At the same time, opportunities are created, especially in
the use of renewable energy sources, whereby countries can
share their geographical advantages and thus can contribute to
the security of supply, cost effectiveness, and reducing green-
house gas emissions from the energy sector. This requires
expansion and maintenance of the European energy infrastruc-
ture, rules and procedures that enable fair access and secure
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 147
Table 1. Overview of the capabilities and limitations of renewable energies.
Parameter /
Technology
Base load Predictability Geographical
limitations
Energy grid
connection
Area of Application Notes
Run-of-river
power plants
Yes Yes River course Power grid
(transmission
and distribution
network)
Regional/national Established
technology,
repowering
Hydroelectric
pumped storage
No Yes Altitude difference Power grid
(transmission
network)
Regional/national Established
technology,
repowering
Onshore
wind power
No Conditionally Wind yield/
landscape and
nature conservation
Power grid
(transmission
and distribution
network)
Regional Established
technology,
repowering
Offshore
wind power
No Conditionally Wind yield/out of
sight of the coast
Power grid
(transmission
network)
National Ensure grid
connection
Photovoltaics No Conditionally Sunlight Power grid
(distribution
network)
Regionaocal Fragmented
generation
structure
Solar thermal No Conditionally Direct sunlight No (district
heating networks
possible)
Local Fragmented
generation
structure
Central solar
thermal (CSP)
with/without steam
power plants
Yes (with
thermal storage)/
no (without thermal
storage)
Conditionally Direct sunlight District heating
networks,
power grid
Regionaocal Depending
on location
Geothermal Yes Yes Geological structure No (district heating
networks possible)
Local Problematic social
acceptance
Biogas with
cogeneration plant
Yes Yes Cultivated area
biomass
Gas network after
separation of CO
2
Local Generation of heat
and electricity
Biomass Yes Yes Cultivated area
biomass
Power grid, district
heating network
Local Generation of heat
and electricity
Biofuels Yes Yes Cultivated area
biomass
No, transportation
via rail tank cars
National Vehicles
exchange between stakeholders, and coherence of national and
European rules. The European Union addresses this challenge
via its Energy Union initiative.
2.5 Definition of Energy Storage
The term ‘‘energy storage’’ has a number of different meanings.
It describes both the storage container (e.g., a gas cavern) as
well as the storage medium (e.g., hydrogen as energy storage
material), in some cases even both (e.g., batteries). Some stor-
age media (especially those that store energy in the form of
chemical bonds, e.g., biomass and liquid synthetic fuels) are
produced during the storage process and consumed during dis-
charge, while other storage media are preserved (e.g., in batter-
ies). Often, the entire periphery of an energy storage system,
which defines the possible applications of the storage system, is
also summarized under this term.
In the context of this article, which mainly deals with the
function of energy storage systems and their application, the
following definition of energy storage is used: An energy stor-
age system is one that can absorb an amount of energy in a
controlled manner (loading), store it for a suitable time period
in a storage medium (storage), and deliver it in a controlled
manner (discharge) at a desired time (cf. the definition pro-
posed by BDEW [3]).
From the above definition, the following consequences,
among others, follow:
– The energy forms for charging, storage, and discharging may
be different.
– These functions (charging, storage, discharging) may be inte-
grated into one system or be implemented in a process chain
of different, spatially separated components.
– If the storage medium is transportable, the discharging can
also be performed at a different location and/or being mobile.
– Chemical energy storage is its own storage medium, whereas
other forms of energy require an additional storage medium.
– Rechargeability with respect to the storage system, but not nec-
essarily with respect to the storage medium, should be given.
– This definition also includes fossil fuels and biomass.
2.6 Function and Contribution of Power Grids
to a Secure Energy Supply
The ability of grids to ensure the transportation and local
distribution of energy or energy carriers is central to a bal-
ance between energy supply and consumption, which is one
of the basic requirements of efficient energy use. Each of the
possible energy supply grids has its own infrastructure that is
tailored to the grid-specific functions. To date, the require-
ments of the grids were often considered separately. In this
context, a report by the Bundesnetzagentur (Federal Network
Agency) in May 2012 [4] pointed out the critical link
between electricity and gas supply. In a recent study by the
Fraunhofer ISE [5], the possibility of an energy supply based
nearly entirely on renewable energy by 2050 is outlined,
which depends largely on the combination of heat and elec-
tricity supply.
At present, it is still difficult to sufficiently estimate the syn-
ergies that could result from better interlinking between differ-
ent grids. For the purposes of supply security and flexibility,
these intergrid links should be regarded as an opportunity,
because they open up the possibility of combining the individu-
al advantages of the respective grids and forms of energy. Smart
combination and a systematic approach could result in a more
cost-effective potential for energy efficiency than isolated con-
sideration of each energy form. In the following, the existing
energy grids are briefly summarized.
2.6.1 Electricity Grid
Grid structure: EU, national (high-voltage grid, 220–380 kV),
regional (medium-voltage grid), local (low-voltage grid).
Grid function: Secure supply of electricity to private, commer-
cial, industrial, and institutional consumers.
Storage technologies: Hydroelectric pumped storage power
plants (commercialized), diabatic (one plant) and adiabatic
compressed-air energy storage (dormant demonstration pro-
ject), batteries (an energy reservoir based on Li-ion batteries
has been connected to the grid in Schwerin).
Function of storage: Improved load management and grid sta-
bilization, peak shaving, load leveling by providing positive and
negative regulation energy for balancing production and con-
sumption, especially with respect to the favored supply of
renewable energies unsuitable for base-load supply.
Notes: The grid fulfills primarily a transport function. The
power grid itself has no storage capacity. The rotating masses
of conventional power plants can be understood as flywheel
accumulators, albeit with the same restrictions of low energy
storage density.
Improved load management reduces peak prices and fluctua-
tions, which worsens the economic conditions for the imple-
mentation of new storage systems with increasing storage
capacity. Other measures for grid stabilization include con-
sumption control (consumer) as well as import and export of
electricity, EU-wide grid development (especially nodes), and
enhanced mutual coordination. Economically, it is probably
wise to combine different grid stabilization measures in a struc-
tured fashion. This raises questions regarding the connections
and regulation of the market. An expansion of network struc-
tures at national and international levels reduces the economic
efficiency and the need for storage systems. Fig. 1 illustrates the
high-voltage grid in Germany, while Fig. 2 shows schematically
the relationship between the different voltage levels.
2.6.2 Gas Grid
Grid structure: EU, national, regional, local.
Grid function: Secure supply of natural gas to consumers.
Storage technologies: Cavern and pore storage, grid capacity,
supply of purified biogas, possible feeding of H
2
(research,
development, and demonstration). Capacity (working gas
volume) of pore storage: 9.8 billion m
3
, cavern storage:
14.3 billion m
3
. The storage capacity covers approximately
20 % of the annual consumption in Germany, which is highly
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 148
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 149
Figure 1. Map showing the 220/380-kV lines of the German high-voltage grid. ªForum Network Technology/Network Op-
eration in the VDE (FNN) 2016, VDE Association for Electrical, Electronic & Information Technologies [6]. The other power lev-
els are too fragmented to be displayed clearly in this figure. Solid lines: existing power lines; dashed lines: power lines under
construction; dotted lines: projected power lines; magenta lines: high voltage direct current power lines; filled dots: power
transformation stations; open dots: large cities.
seasonal due to consumption by the heating market. The stor-
age locations are shown in Fig. 3.
Function of storage: Worldwide, Germany has the fourth largest
natural gas storage facilities, which are also expandable. Gas
storage systems compensate the seasonal (pore storage) and
daily fluctuations (cavern storage). Supply optimization by tak-
ing advantage of fluctuating gas prices also plays an increasing
role. Strategic storage is not currently perceived as a political
necessity, since the natural gas supply is considered to be secure
[8]. Storage systems are therefore operated from a purely com-
mercial point of view. However, at present, strong expansion
of storage is taking place with the prospect of developing
Germany into a hub for the European gas market.
Notes: The gas grid itself, as a transport network of a chemical
energy carrier, can provide both transport and storage func-
tions. The storage capacity of the pipelines is considerable and
can compensate for short-term fluctuations without having to
exploit actual storage (Fig. 4).
Favored supply of critical power plants for the reliable supply
of electricity is not ensured. The natural gas storage facilities are
currently operated commercially in compliance with legal regu-
lations. Political problems leading to disruptions of supply may
require exports to neighboring European countries.
2.6.3 District and Local Heating, Steam Grids
Grid structure: Local.
Grid function: Secure supply of process heat or space heating to
consumers (Fig. 5).
Storage technologies: Grid capacity, thermal storage.
Function of storage: Stability of the heat supply in municipal or
industrial sites, decoupling of the operating modes of combined
heat and power (CHP) plants, balance of seasonal fluctuations.
Notes: Heating grids can provide both transport and storage
functions. However, the storage capacity of the network itself is
dependent on the losses that occur, e.g., by incomplete insula-
tion. It can therefore only be used for short-term storage. The
amount of heat supplied depends on the consumption. It is
possible to take up energy from the power grid via resistance
heating. Strong cross-linking of electricity and heat grids exists,
e.g., in Denmark. The combination of industrial sites and local
district heating networks is often already a reality. The grid
load is subject to seasonal fluctuations.
2.6.4 Fuel Grids
Grid structure: EU, national (CEPS and NEPS, operator: FBG)
Grid function: Ensuring strategic supply of major civil and
military consumers (airports, ports, refineries) with fuels.
Storage technologies: Fuel storage, grid capacity. Petroleum
storage facilities in Germany have a capacity of approximately
27.6 million m
3
, which guarantees a supply for around two
months. The stockpiling of petroleum products is regulated by
law. The law calls for stocking of the respective petroleum
products based on the average net imports [9] for 90 days by
the Erdo¨lbevorratungsverband (Petroleum Stockpiling Agency),
which can be done either in Germany or in other countries of
the EU. At present, this requirement is met with a reserve of
23.4 million t of crude oil and petroleum products [10]. A stra-
tegic Federal reserve for crude oil no longer exists. By decision
of the Federal Government, it was dissolved from 1997 to 2001.
The storage facilities in Germany are shown in Fig. 4. An exam-
ple of the military infrastructure is the CEPS (Central Europe-
an Pipeline System) with more than 5500 km of pipelines,
which connects 30 NATO depots and six civilian depots, as well
as military and civilian bulk consumers. The system has a total
storage capacity of 1.22 million m
3
and is shown in Fig. 6 [12].
Function of storage: Strategic security, civil/military transport
infrastructure.
Notes: The fuel grid itself, as a transport network of material
energy carriers, can provide both transport and storage func-
tions. The storage capacity of the pipelines and transport con-
tainers (tank trucks, tank wagons, filling stations, etc.) is con-
siderable and can compensate for short-term fluctuations,
without having to exploit the actual storage.
The fuel supply of other consumers is ensured via a logistical
infrastructure distributing the fuel, starting from the major
ports with large storage facilities (e.g., Hamburg, Rotterdam),
by barge, rail transport, and trucks. At the nodes and distribu-
tion centers, fuels are stored above ground in large tanks.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 150
Figure 2. Functions and structures of the various power levels in the power network [7]. ªF.A.Z.-Grafik/Karl-Heinz Do
¨ring. Wired land-
scape: Large-scale power plants feed the maximum-voltage transmission network (red), medium-sized power generators feed the high-
voltage grid (orange), and smaller municipal power plants feed the medium-voltage grid (green). From there the electricity flows into
the low-voltage lines (blue), which end in our home outlets. Through the substations between networks, the energy can principally flow
in either direction (double arrows). The railway operates a separate high-voltage grid. It is powered by its own power plants or at least its
own generators in other power plants.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 151
Figure 3. Storage locations for natural gas, crude oil products, and liquefied gas in Germany; ªState Office
for Mining, Energy and Geology [10]. Natural gas is stored in pore storage (circles) which are marked as op-
erational (red) or planed (yellow) with their respective capacity in Mio. m
3
(Vn); natural gas cavern storage is
marked by elliptical symbols with the same color coding; cavern storage for mineral oil, oil-based products
and liquefied natural gas are marked by green square symbols; the number indicates the number of individ-
ual caverns.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 152
Figure 4. Overview of the German gas pipeline network at a glance; ªTSOs [11]. Solid blue lines: existing transmission pipelines; dashed
orange lines: existing pipelines; orange symbols: gas compression stations for which operation has not started yet; dashed blue lines:
connections gas storage in neighboring countries; solid blue squares: domestic gas storage facilities.
2.6.5 Summary of the Current Status of Storage and Grids
Currently, the different energy supply and distribution grids
are operated independently. Stronger coupling of the energy
networks raises the question of a common communication and
regulation structure for the necessary mutual coordination.
An overview of the grids and storage facilities is given in
Tab.2. It shows that the gas and fuel grids have large storage
facilities, whereas capacities for storing electricity and heat are
insufficient. The technical difficulty is the low storage density
of electricity- and heat-storage systems, each of which requires
a storage medium. In contrast, as chemical storage systems, the
energy resources fuel and natural gas have high energy den-
sities.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 153
Figure 5. Structure of the combined local and district heating networks of Neuburg on the Donau to integrate industrial waste heat in muni-
cipal district heating; ªeta energy consulting GbR [13]. Utilization of industrial waste heatand renewable heat to supply public,private and
industrial customers (total amount of required heat: 270 GWh
th
a
–1
); letters and numbers: supply regions; red symbols: industrial waste heat
sources; green symbols: renewable heat sources; magenta symbols: customers requiring steam; blue symbols: district heating customers.
Table 2. Network-attached storage; projects under construction/planned in parentheses [10].
Grid Storage type Amount Capacity Power Discharge time Notes
Electricity Pumped-storage hydro-
electric power plants
35 (+14) 40 (+>13) GWh 7 (+6) GW 6 h Annual energy demand ca.
600 TWh; pumped storage
covers ca. 35 min. of this
requirement.
Diabatic compressed air
energy storage
1 600MWh 321 MW 2 h
Natural gas Pore storage 20 locations 9784 million m
3
(working gas volume)
7 085 000 m
3
h
–1
58 d
Cavern storage 30 locations,
260 storage
sites
14 315 million m
3
(working gas volume)
20 167 000 m
3
h
–1
30 d
Heat No major storage
Oil Cavern storage for crude
oil, petroleum products,
and liquefied petroleum
gas
103
(12 locations)
27 601 (+3800)
(million m
3
)
Covers total demand of oil
for about 59 d.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 154
Figure 6. Map of the NATO pipeline network CEPS; ªNATO [12].
3 Evaluation Criteria and Scenarios for
the Integrated Use of Energy Storage
Technologies
Energy storage can be based on different technologies and serve
the most diverse requirements. Some energy storage technolo-
gies are summarized in Tab. 3, and their diversity already
clearly indicates the complexity of the subject ‘‘energy storage’’.
3.1 Tasks and Functions of Energy Storage
For a secure and efficient energy supply, supply and consump-
tion must be in temporal and spatial balance. This task is per-
formed mainly by the grids; if necessary, the energy supply is
adjusted. The situation is critical above all for the electric
power grid, since it has no inherent capacity to compensate for
imbalances that affect the grid frequency. The example of the
electricity supply can be taken to illustrate the functions,
advantages, and limitations of energy storage systems.
3.1.1 Definition: Residual Load
In this article, the term ‘‘residual load’’ refers to the difference
between the load (demand) and production (supply) and has
the units of power (e.g., kW). A positive residual load means
that demand exceeds supply (additional power plants must be
used), while a negative residual load indicates excess produc-
tion (power stations must be switched off).
3.1.2 Definition: Load-Duration Curve
The load-duration curve describes the residual load over a
selected period of time (e.g., a year). The values of the residual
load are not presented chronologically, but sorted in descend-
ing order (see the figures in the next section). The conclusion
would then be: e.g., in the year under consideration (8760 h),
there were 2000 h with a positive residual load, that is, more
demand than supply, 5000 h with a balanced load, and 1760 h
with an excess supply of energy.
3.1.3 Smoothing of Random Fluctuations
The random behavior of consumers cannot be fully predicted.
Therefore, this automatically leads to fluctuations, the so-called
load forecast error, of the readily predictable mean value of the
residual load, which can be described as a normal distribution
and must be balanced [14]. In the case of normally distributed
fluctuations, a significant attenuation of the peaks can already
be achieved with a relatively small rechargeable storage unit. A
prerequisite is that charging and discharging of the storage are
fast with respect to the fluctuations. This is illustrated in Fig. 7.
The storage unit can both emit energy (dark rectangle) and
thus reduce a positive residual load (load surplus) as well as
absorb energy (light rectangle) and thus reduce a negative
residual load (production surplus). The height of the rectangles
represents the performance of the storage unit, and the area is
equal to the capacity. The load-duration curves sort the load
curves according to their absolute values. In the case of a nor-
mal distribution around a mean value, the variation curves are
symmetrical. The effect of energy storage, which compensates
80 % of the variation, is reflected mainly in the central regions
of the load-duration curve, where deviations are compensated.
Storage cannot counterbalance the fluctuations if they exceed
its power consumption and dissipation. If the capacity of the
energy storage system allows no further energy uptake or
release, then positive or negative fluctuations cannot be sub-
dued until the storage has had the opportunity to be either
discharged or charged.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 155
Table 3. Examples of energy storage.
Storage technology Form of energy loading Form of energy discharge Storage of energy as
Hydroelectric pumped storage Electricity Electricity Potential energy (height difference)
Battery Electricity Electricity Electrochemical potential (charge separation)
Flywheel storage Electricity Electricity Kinetic energy (velocity difference)
Warm-water storage Heat Heat Sensible heat (temperature difference)
Compressed-air reservoir Electricity Electricity Inner energy (pressure difference)
Phase-change materials Heat Heat Latent heat (phase transition enthalpy)
Thermochemical storage Heat Heat Chemical bond (enthalpy)
Power-to-gas/liquid/X (hydrogen,
methane, methanol, etc.)
Electricity Heat and possibly electricity
or mechanical energy
Chemical bond (enthalpy)
Fossil fuels: coal, oil, natural gas Solar energy Heat and possibly electricity
and mechanical energy
Chemical bond (enthalpy)
Biomass Solar energy Heat and possibly electricity Chemical bond (enthalpy)
Biodiesel Solar energy Heat and mechanical energy Chemical bond (enthalpy)
The design of the energy storage system depends on the
desired smoothing of the curve and must be economically opti-
mized, especially with regard to smart combination with other
means to smooth the residual load (e.g., import/export of
energy).
3.1.4 Compensating a Systematically Higher Supply
If the supply of a grid is continuously higher than the demand,
compensation can no longer be carried out exclusively in the
framework of the considered grid. The available excess energy
supply must then either be reduced, e.g., by shutting down
power plants, or the excess energy is converted to another form
of energy that is either consumed or stored.
A storage technology can only meet this requirement if it
removes the excess energy from the original grid permanently
by transforming it into another form of energy, which may also
be more readily stored. This is illustrated in Fig. 8. The effect
on the residual load is made clear by the smoothing in the form
of the gray line in relation to the original residual load. The
area of the rectangle represents the total energy that is received
by the storage system. The height of the rectangle corresponds
to the maximum power consumption of the storage system. In
this example, the average excess of the residual load is identical
to the maximum power consumption of the storage system.
The average power consumption is equivalent to 75 % of the
maximum. The fluctuations of power and load are normally
distributed around the average values with a standard deviation
equal to half of the average excess. The load-duration curves
clearly show the effect of the additional decrease in perform-
ance. The reduction is carried out exclusively asymmetrically in
the central region of the positive residual load.
For the subsequent use of this stored excess energy, the
options of the new energy forms are available, such as coupling
to another energy grid and material use if a storage material
can be used as a raw material.
Storage technologies for this function will in general be a
combination of individual solutions. For example, power-to-
gas concepts would convert the excess electricity to hydrogen
by electrolysis of water. This hydrogen could power a gas tur-
bine to generate electricity, albeit with low efficiency, be used
for mobility in fuel-cell cars, or, after conversion with carbon
dioxide to methane, be fed into the natural gas grid, or be
stored in the form of liquid fuels. Alternatively, the hydrogen
could also be used in metallurgical, chemical, or petrochemical
processes.
This example illustrates how different technology chains can
fulfill the function of energy storage and how coupling with
other forms of energy opens up new degrees of freedom in the
use of energy that were not accessible to the original form of
energy.
3.1.5 Compensating a Systematically Higher Demand
Similar to the above case, it is also impossible to achieve a bal-
ance between supply and demand solely in the energy system
under consideration if demand is consistently higher than the
corresponding supply. In this case, the needed additional ener-
gy must be supplied from outside the system. This can be done,
e.g., through the supply of additional power plant capacity for
the electrical grid.
Another contribution may be provided by storage technol-
ogy that feeds the energy of an independent system with higher
supply than demand, optionally after a transformation process,
into the system under consideration. This is analogous to the
case considered above and illustrated in Fig. 9. The effect is
illustrated by the smoothing in the form of the gray line in rela-
tion to the original residual load. The area of the rectangle rep-
resents the total energy that is output from the storage system,
and the height of the rectangle corresponds to its maximum
power output. In this example, the mean deficit of the residual
load is identical to the maximum power output of the storage
unit. The average power output is equivalent to 75 % of the
maximum. The fluctuations of power and load are normally
distributed with a standard deviation equal to half the average
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 156
Figure 7. Impact of storage on the smoothing of the residual
load. The residual load fluctuates statistically with a normal dis-
tribution (standard deviation: 10 %) around the mean.
Figure 8. Effect of an additional load decrease on systematic
oversupply.
excess about the mean values. A combination of the two energy
systems into a whole system may, under certain circumstances,
be an attractive alternative in terms of efficient use of energy.
Whether this option is also cost-effective, must be checked on
a case-by-case basis.
3.1.6 Structured Fluctuations
Structured fluctuations can be regarded as combinations of dif-
ferent supply/demand situations that can be compensated within
an observed time interval. As shown in Fig. 10, it is not primarily
about smoothing of the profile, but about the postponement of
the resulting surplus energy for supply in times of inadequate
load coverage. The total storage capacity is given by the larger of
the two areas: either the sum of the light gray rectangles (power
output) or the dark gray rectangle (power) defines the required
capacity. The requirements for power consumption and dissipa-
tion are different. On the load-duration curve it is clearly evident
that a surplus occurs only during daylight hours.
Therefore, the storage capacity must be high enough that
load coverage is secured during the entire period. It is also
noteworthy that the requirements for power uptake and release
are usually not identical. An example is a hot-water tank for
radiators, which is charged quickly, whereas the heat is emitted
over a longer period with lower power.
The power uptake is defined by the difference between the
maximum power and the load, while the power output in this
example is given by the daily course of the electricity demand.
In comparison to the smoothing of random fluctuations, the
storage capacity is much greater due to the time-structured en-
ergy production, but the number of cycles is lower.
Thus, storage systems whose investment costs are signifi-
cantly defined by their capacity (e.g., batteries) in this case are
generally not economically favored. The different requirements
for power uptake, power output, and capacity suggest that this
storage function should not be met by a single storage unit, but
by an integrated system of individual elements that can cover
the various requirements separately through optimal technical
solutions.
3.1.7 Energy Storage Systems: Requirements
and Functions
The demands on energy storage systems vary greatly depend-
ing on the purpose for which they are used. In contrast to the
above idealized cases, the balance between supply and demand
in a real energy system can be regarded as a combination of
these ideal cases, each of which makes different contributions.
In Tab. 4, the individual cases and their respective requirements
are summarized.
3.2 Energetic Potential Analysis
3.2.1 Fundamental Considerations for Energy
Conversion
The storage of energy in the same form is often neither possible
nor advantageous. However, transformation to another form of
energy is always associated with losses. As part of the process
chain starting from the use of the primary energy carrier,
through storage, to the use of the desired energy form, each
step can be assigned an efficiency value. In real systems this
efficiency is always < 1. The overall efficiency of the entire
sequential process chain corresponds to the product of the effi-
ciencies of the individual process steps. This results in the fol-
lowing three prerequisites for energetically efficient use of the
process chain:
– Minimizing the number of process steps from the primary
energy carrier to the consumer.
– Carrying out the individual process steps in the process
chain with the highest possible efficiency.
– Minimizing the number of conversions between various
forms of energy and energy carriers, since they are relatively
inefficient.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 157
Figure 9. Effect of an additional load supply on systematic un-
dersupply.
Figure 10. Schematic representation of a storage system to
compensate for structured variation; example of a daily load
curve of a photovoltaic system and the electricity demand.
3.2.2 Properties of Storage Technologies
Capacity
The capacity of a storage unit (expressed in units of energy,
e.g., Wh, J, TCE, TOE, etc.) indicates how much energy the
storage medium can take up. It is defined by the energy density
of the storage medium and its energy form. The mass-normal-
ized storage density of heat and electricity storage is generally
rather low, whereas that of chemical storage is relatively high.
Thus, electrical energy is stored in, e.g., batteries by electro-
chemical charge separation, the extent of which depends on the
particular combination of materials and their quantities.
In addition to the mass-normalized storage density, depend-
ing on the application, the volume-normalized storage density
may also play a central role. Solids and liquids (e.g., gasoline
and coal) have a higher storage density than gases (e.g., hydro-
gen and methane), which must be compressed to obtain
acceptable volumetric energy densities.
In general, the application profile of a storage technology
determines the demands on the capacity. Storage for mobile
applications should be small, light, and compact, with a clearly
defined, limited capacity (e.g., a fuel tank). In contrast, a sta-
tionary power-to-gas plant for the production of hydrogen
would have a virtually freely scalable capacity that would be
limited only by the size of the connected storage cavern or the
amount of gas that can be fed into the gas grid.
Power Input/Output
An important performance indicator of a storage system is the
amount of energy absorbed or released by the system per unit
time (expressed in terms of energy per unit time, e.g., W).
Depending on the application, the requirements for power
uptake and discharge may be different. Pumped-storage hydro-
electric plants, which inter alia are intended to offset the ran-
dom fluctuations in the power grid, are designed such that
power uptake and release are of similar magnitude. In contrast,
the power consumption of seasonal heat storage devices is rela-
tively high compared to their power output. The storage unit is
loaded relatively quickly, whereas discharge is extended over a
longer period of time.
Moreover, power absorption and release may involve differ-
ent forms of energy. In the above example of a heat accumula-
tor, charging can be done electrically, while discharge occurs as
thermal energy by means of a heat carrier (e.g., hot water or
steam). It is therefore essential to clearly define the perform-
ance requirements of the storage system related to the purposes
of the application in order to select the appropriate system.
Full-Load Cycle Number
The full-load cycle number (expressed as frequency in units of
reciprocal time, e.g., h
–1
) is a calculated parameter (similar to
the full-load hours of a power plant) which describes how often
a storage system has been fully charged and discharged per unit
time. Partial charging and discharging are added up. Conclu-
sion about the time-resolved storage demand cannot be made
on the basis of the full-load cycle number. It is limited by the
materials used and defined by the desired application. In prin-
ciple, the highest possible full-load cycle number is desirable
for economic operation of energy storage systems.
Some applications, such as the compensation of random
fluctuations, lead to frequent stress on the storage system, and
thus to a high full-load cycle number at a low storage capacity,
while a seasonal heat storage is charged and discharged only
once a year. The full-load cycle number is a property of the
overall system. Thus, in a power-to-gas system, a cycle is only
completed when the produced gas is used.
Especially electrochemical storage technologies are charac-
terized not only by the calendar life, but also by the cycle life.
Charging and discharging cycles lead to irreversible changes in
the material structure that, after a certain number of cycles, will
make it unusable for further use as a storage material. Unlike
electrochemical systems, the cycle life of mechanical storage is
practically unlimited, provided the appropriate maintenance
and the replacement of defective parts is performed.
Efficiency
The efficiency of a process can be expressed as a percentage or as
a fraction. It is the ratio of the energy output to the input. The
efficiency itself is generally based on a particular product or a
particular form of energy. It defines, e.g., how much of the energy
used is incorporated in the product or how much electrical ener-
gy can be generated in a power plant by a combustion process in
comparison to the chemical energy of the fuel itself. The efficien-
cy of a storage system indicates the amount of energy that is
available for discharge relative to the amount stored.
The theoretical efficiency, e.g., of a chemical reaction, is
always greater than that achieved in its technical implemen-
tation, since additional losses occur. For technical processes the
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 158
Table 4. Overview of the different storage functions and their requirements.
Storage task Storage function Requirements power Requirements capacity Technology examples
Random fluctuations in
the residual load profile
around the mean
Smoothing of the residual
load profile
Relatively (equal) high
power and release
Low capacity Pumped-storage power
plants, batteries, flywheel,
sensible thermal storage
Systematic surplus Providing additional load Adequate power input Mostly unlimited capacity Electrolysis
Systematic deficit Providing additional power Adequate power output Mostly unlimited capacity Power plants
Structured fluctuations Postponement of the power
provided (load levelling)
Differently dimensioned
power consumption and
release
High capacity Redox flow batteries,
phase-change materials
validity of a specified efficiency is severely limited unless other
operating parameters are clearly specified. Frequently, it is im-
plicitly assumed that the process is operated at its optimum
operating point. The implementation of a process often in-
cludes peripheral processes, e.g. gas purification, which also
increase the cost, and thus reduce the overall system efficiency.
If it is not stated whether the energy requirements of the
peripheral processes have been included, specification of the
efficiency is often not very meaningful.
Assessment of different efficiencies and their comparison
with one other is therefore only possible if the exact conditions
of the efficiencies are known. Since this is not guaranteed for
all sources in this document, the efficiencies given here should
rather be considered in terms of orders of magnitude and not
as exact values.
The overall efficiency of a sequential process chain is given by
the product of the efficiencies of the individual process steps,
whereby the efficiency definitions must be compatible with each
other. Since there are always energy losses in power conversion,
the efficiency of a process chain is always less than 100 %.
For a more detailed discussion, the reader is referred to the
example of the efficiency definition for an electrolyzer used for
the electrolysis of water [15], in which its meaningfulness and
its limits are discussed.
In the following sections, the following efficiency definitions
are used:
– The electrical efficiency refers to the amount of electricity
supplied compared to the input amount (e.g., battery stor-
age).
– Efficiency (LHV, lower heating value) refers to the efficiency
of a combustion or conversion process based on the net
calorific value, whereby the heat of condensation of the
water vapor cannot be used (e.g., fuels).
– Efficiency (HHV, higher heating value) refers to the efficien-
cy of a combustion or conversion process on the basis of the
gross calorific value, whereby the heat of condensation of
water vapor can be used (e.g., chemical processes).
Response Time
The power uptake and release of the selected storage technolo-
gy must be sufficiently fast to respond to the variability of the
application under consideration. Fluctuations in the power
grid, e.g., caused by fluctuating feed of wind energy, can be
compensated quickly enough through the use of batteries,
whereas sensible-heat storage would react too slowly to gener-
ate steam with a connected turbine. If the storage function is
performed by process chains, usually the slowest step deter-
mines the response time (expressed in units of time, e.g., s) of
the entire chain.
3.3 Economic Evaluation
3.3.1 Fundamentals of the Cost-Effectiveness
of Storage Technologies
The main focus of this position paper lies on the scientific and
technological feasibility rather than an economic analysis. The
scientific and technological feasibility is a necessary but not
sufficient prerequisite for the use of storage technology. The
decision to implement a particular storage technology should
be made only after a thorough economic and business evalua-
tion. This will always be done on a case-by-case basis. However,
there are some aspects of a more general nature.
3.3.2 Grid Services
Storage technologies can contribute to the stability of the sup-
ply grid through an improved balance between supply and
demand and facilitate more flexibility. By providing additional
balancing power, they allow a decrease in reserve capacity,
reduce the required grid expansion, and help to eliminate bot-
tlenecks. They can increase the service quality of the entire grid
and thus provide an universal grid service, regardless of the
specific degree of utilization of individual storage units, from
which all system users benefit.
The power grid is a special case, since a technical possibility
for long-term storage of electricity once it is generated does not
exist at present. Short-term storage is possible, but is not spe-
cifically remunerated as a grid service.
The traditional business model of pumped-storage hydro-
electric power stations is based on the utilization of the price
difference between periods of high and low consumption (the
so-called spread). Currently, this model offers only limited eco-
nomic viability, because the original high-price periods (after-
noons) are now covered by a relatively high contribution of
photovoltaics, which have feed-in priority independent of the
electricity price.
For the supply of electrical energy, the operators of pumped-
storage hydroelectric power stations, but also industrial compa-
nies with flexible processes, usually negotiate contracts with
grid operators. A list of companies that offer reserve-energy
capacity is available on the internet [16]. It is apparent that the
main suppliers of reserve-energy capacity themselves belong to
the energy sector, supplemented by public utilities and site
operators. Individual industrial providers are mainly active in
the metals and chemicals sector. Geographically, not all suppli-
ers are located in Germany, but also in neighboring Alpine
countries, which clearly highlights the European dimension of
power grids.
Conversely, improved grid stability tends to lead to a lower
utilization of storage units, which affects their profitability. It is
therefore to be expected that a too massive expansion of stor-
age technology is not necessarily the most economic method
for grid stabilization and decreases the economic conditions of
the individual storage units.
3.3.3 Storage Usage
In the best-case scenario, a storage unit is charged and dis-
charged as completely as possible with high frequency. In
general, the more frequently a storage unit is used, the more
likely it can be operated economically. This, however, strongly
depends on the particular application. The faster a storage
technology is capable of reacting to the grid, the more likely it
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 159
can exploit short-term price fluctuations and the greater its
potential contribution to grid stability.
3.3.4 Investment Costs
The investment costs are written off over the amortization peri-
od, or the total number of cycles and the utilization of the stor-
age. Depending on storage technology, they vary greatly.
Investments with long amortization periods presuppose a con-
fidence in the appropriate political and economic framework.
The technical lifetime is usually longer and ranges from a few
years (batteries) up to 100 years (pumped storage).
3.3.5 Operating Costs
In general, the energy that is fed into a storage unit is not free.
The universal business model for storage technologies utilizes
the difference in the market price of the energy output and
input.
In exceptional cases, electrical energy is traded at negative
prices at the electricity exchange. This excess electricity is cur-
rently still very manageable. It is generally expected that the
amount of excess electricity will increase with the expansion of
volatile renewable energies, that is, wind and photovoltaics
[17]. However, negative electricity prices cannot increase with-
out limit since at a certain price, it will be more favorable for
the producer to switch off the plant than pay for supplying the
surplus electricity. Acceptance fees are currently guaranteed
under the EEG remuneration for renewable energy, and there-
fore the maximum value of the negative electricity price is set
before a down-regulation of the corresponding assets becomes
more favorable. The current version of the EEG (Sect. 51),
restrics the remuneration in times of negative electricity prices
[18].
As a measure of the excess amount of electricity, the amount
of dumped energy from EEG electricity can be used. In 2015,
4.7 TWh was dumped from renewable sources [19]. For 2024, an
amount of dumped energy on the order of 0 – 8.8 TWh (0 – 2.5 %
of the predicted renewable electricity generation) is expected
[17]. However, this amount of dumped energy is manageable.
A dena study estimated, depending on the scenario, from 2.5 to
12 TWh of "non-integrable energy" by 2030 [20]. In compari-
son, the current trade balance, i.e., the net amount of current
exports by Germany to neighboring countries, amounted to
51.8 TWh in 2015 [21].
In this case as well, a strong expansion of storage technolo-
gies would in principle stabilize prices by increased demand
and reduce the surplus quantities of electricity. An uncon-
trolled expansion of storage technologies would lead its own
business model ad absurdum.
This business model is obviously not suitable for long-term
storage, and consequently, stockpiling of strategic reserves of
natural gas and oil is regulated by law.
This reveals a fundamental dilemma. From an economic per-
spective, storage technologies serve to smooth the difference
between the load and generation and, therefore, to smooth out
the price differences. From a business perspective, precisely
these differences between charging and discharging and the
associated price differences constitute the basis for economic
operation of the storage system.
3.3.6 Process Chains
The function of energy storage does not need to be met by a
single storage technology and can be provided as a process
chain of various technologies. A process chain can be signifi-
cantly better adapted to specific storage requirements and espe-
cially to their changes. It also allows cross-links to other areas
of the energy system to be established and thus increases the
flexibility of the overall system. However, the downstream pro-
cess steps must also be considered for a full economic analysis.
3.3.7 Alternative Uses of Chemical Storage
In virtually all cases, there are alternative uses of substances
produced for energy storage. Markets exist in which pricing of
these substances is performed. For example, methanol is a basic
chemical product that is traded and has a global market price.
Since, in principle, the same potential applications exist for the
storage substance as for the original substance produced for
other purposes, they are in direct economic competition with
each other. Its use as a storage substance is in competition with
conventional industrial value chains. However, in addition, the
relative dimensions for use in energy applications and the alter-
native uses must be taken into consideration.
A final evaluation of the efficiency of a storage system is
always dependent on the selected reference case and the associ-
ated conditions. For example, the VDE study ‘‘Energy Storage
in Power Supply Systems with a High Proportion of Renewable
Energy Sources’’ [22] compiles the full costs of the storage sys-
tems under certain operating scenarios. In Sect. 5, simplified
cost assessments are carried out for different reference cases.
4 Storage Technologies
4.1 Technologies for Storage as Mechanical
Energy
Mechanical storage systems form the basis of the storage
reserve in the power supply. The most mature technology is
used in pumped-storage hydroelectric power stations, which
can provide both positive and negative balancing power. Cur-
rently, pumped-storage hydroelectric power stations are the
only available large-scale storage technology for electricity.
They need certain geographical conditions and their relatively
low energy density leads to large storage volumes. The develop-
ment potential is relatively limited due to geographical require-
ments and, above all, the lack of acceptance within the popula-
tion in Germany.
For the use of compressed-air energy storage a wealth of
experience is already available. By combination with thermal
storage, it can be operated adiabatically, which is reflected in a
significantly higher degree of efficiency, but also in higher
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 160
investment costs. For a further expansion of compressed-air
energy storage technology, possible competition with the use of
caverns for the storage of natural gas or hydrogen must be tak-
en into consideration. However, this competition is limited to
caverns of lower depths. Hydrogen and natural gas are usually
stored at high pressures, which require depths of about 1000 m.
This is much deeper than the depth required for compressed
air. Flywheel storage has so far only been used in industrial
applications. Flywheels provide high power output but have
only a small capacity. The different technologies are compared
in Tab. 5.
4.1.1 Pumped-Storage Hydroelectric Power Stations
Pumped-storage hydroelectric power stations have been used
on an industrial scale for more than 80 years. They are linked
to specific geographical conditions. Energy is stored in the form
of the potential energy due to the difference in height of the
upper and lower reservoirs and the mass of the storage
medium water. If the grid has more power than can be used,
water is pumped from the lower reservoir to the upper reser-
voir. During times of increased demand for electricity, water
flows from the upper reservoir through a turbine into the lower
reservoir, generating electricity which is fed into the grid
(Fig. 11).
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 161
Table 5. Storage in the form of mechanical energy [2, 22–24]. The figures serve as benchmarks.
Technology Pumped-storage
hydroelectric
power plant
Pumped-storage
hydroelectric
power plant,
mine
Compressed-air
storage, diabatic
Compressed-air
storage, adiabatic
Flywheel Air liquefaction
Performance range MW–GW MW–GW MW–GW MW–GW 10 kW–20 MW 10–250 MW
Capacity MWh–GWh MWh–GWh MWh–GWh MWh–GWh kWh MWh–GWh
Reaction time min min 15 min 15 min s 20 min
Maximum number
of cycles
20 000 30 000 30 000 1 000 000
Plant life [a]
a)
50–100 < 50 < 50 20 > 30
Cycle efficiency
b)
(electrical) [%]
£80 42 (Huntdorf, D);
54 (McIntosh,
USA)
70 85–90 50–60
Specific energy
density [Wh kg
–1
]
Depending on
height difference,
average ca. 0.7
Depending on
height difference,
average ca. 0.7
< 5 439
Volumetric energy
density [Wh L
–1
]
Depending on
height difference,
average ca. 0.7
Depending on
height difference,
average ca. 0.7
2–5 2.9 10
Specific power
density [W kg
–1
]
275
Investment costs,
power (system)
[€ kW
–1
]
500–1000 1800 1000 1000 100–360 625–1400
Investment costs,
capacity (system)
[€ kWh
–1
]
5–20 25 ‡40 »80 1000
Typical Application Load leveling,
balancing energy,
arbitrage
Load leveling,
balancing energy,
arbitrage
Load leveling,
balancing energy,
arbitrage
Load leveling,
balancing energy,
arbitrage
Power quality,
control energy
Load leveling,
balancing energy,
arbitrage
State of technology Technical
implementation
Research Demonstration Development Technical
implementation
Development
Notes Ca. 40 plants (D);
power: 7 GW;
capacity: 40 GWh
Worldwide: 2
plants; power (D):
321 MW;
capacity (D):
642 MWh
Originally planned:
1 plant; power:
90 MW; capacity:
360 MWh
Self-discharge
3–20 % h
–1
Air liquefaction as
technology is state
of the art, use as a
storage is in
development
a)
Includes sharing of key power plant components;
b)
see discussion in Sect. 3.2.
The technology is very reliable and mature. The electricity-
to-power efficiency of a modern pumped-storage hydroelectric
power station is approximately 80 % [25]. The disadvantages of
pumped-storage power stations lie in the high investment
costs, the geographical conditions, and the associated low
degree of societal acceptance. Currently, Germany has an out-
put of around 7 GW with a storage capacity of 40 GWh.
Geographically, the hydroelectric pumped-storage power plants
are mainly located in the southern and central parts of
Germany (Fig. 12). A new concept integrates water storage
tanks directly into the foundation of a wind mill, in effect a
small hydroelectric power station. A pilot unit with an energy
storage capacity of 70 MWh is being built in Gaildorf (southern
Germany) by Naturspeicher GmbH [26]. Current considera-
tions to take advantage of former mines as underground
pumped-storage power stations are still in the research stage
[27]. The challenges lie in the development of a cost-effective
system concept, whereby in particular the reservoir must be
newly created.
4.1.2 Compressed-Air Reservoirs (Diabatic, Adiabatic)
Compressed-air reservoirs use electricity for the compression
of air in storage caverns (Fig. 13 a). When electricity is
required, the compressed air is decompressed via a gas turbine
and thus generates electricity, which is then fed back into the
grid. Storage caverns can be, among others, salt caverns, espe-
cially in northern Germany. During compression of the air, a
significant amount of heat is generated, which is emitted as
waste heat into the environment in the case of a diabatically
operated reservoir. Conversely, during the expansion of the
working gas through a gas turbine additional heat is required
since the gas is strongly cooled. In a diabatic compressed-air
reservoir, this heat is supplied by means of gas burners. Ulti-
mately, the heat losses and the necessary heat input limit the
(electricity-to-electricity) efficiency of diabatically operated
compressed-air storage to about 50 %.
Worldwide, only two diabatic compressed-air energy reser-
voirs are in use. The compressed-air reservoir in Huntorf
(Fig. 13 b) has been in operation since 1978, with an output of
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 162
a) b)
c)
Figure 11. (a, b) Operating principle of a pumped-storage power plant and (c) Raccoon Mountain Pumped Storage Plant at Tennessee
River, USA; ªVoith GmbH [31].
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 163
a)
b)
Figure 12. (a) Pumped-storage power plants in Germany and neighboring states as well as (b) planned projects [32]; ªBDEW; Stauin-
halt: Water capacity; Inbetriebnahme: Beginning of operation.
321 MW [28] and a storage capacity of 2 h of full load at an
electrical efficiency of 42 %. The McIntosh power plant in Ala-
bama, USA, has been in operation since 1991 with an output of
110 MW and a storage capacity of 2860 MWh [29], equaling
approximately 26 h of full load. The McIntosh power plant has
a recuperator, which recovers the heat of the combustion off-
gas for preheating, thereby increasing the electrical efficiency of
the plant to around 54 %. In contrast, adiabatic compressed-air
energy-storage plants use a combination of a compressed-air
reservoir and thermal storage (see Sect. 4.4) to store the heat of
compression and make it available for decompression. Thereby,
no external heat supply is needed. Adiabatic compressed-air
energy storage plants are currently not applied on an industrial
scale. However, they are being extensively studied in the
ADELE and ADELA-ING projects (Fig. 14). The basis of these
projects is a plant with an output of 90 MW and a capacity of
360 MWh aiming at an electrical efficiency of around 70 %
[30].
The projects were able to demonstrate scientific and techno-
logical feasibility. Due to the current conditions of the electric-
ity market, especially low spreads and low electricity prices,
however, the cost effectiveness could not be demonstrated [33].
Large-scale implementation has been suspended until further
notice.
4.1.3 Flywheels
In the case of flywheel storage, electrical energy is used to accel-
erate a flywheel. Conversely, it is discharged by deceleration of
the flywheel. Flywheel energy storage is used locally and is
characterized by high full-load cycle numbers and high effi-
ciency. Flywheels exhibit high output power but only low
capacity. The gyrating masses of conventional power plants can
act as flywheels and thus represent a (small) independent stor-
age capacity in the power grid (Fig. 15).
4.1.4 Air Liquefaction as Mechanical Storage
While in the case of a compressed-air reservoir, cooling of the
air during expansion in the turbine is a problem, this effect is
exploited in the case of air liquefaction. The storage principle is
based on the difference in volume between liquid and gaseous
air. The production of 1t of liquid air requires 439 kWh of
energy, which can be reduced to 210–230kWh by recycling
cold released at the power recovery phase during the evapora-
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 164
a) b)
Figure 13. (a) Operating principle of a diabatic compressed-air reservoir and (b) view of the storage in
Huntorf [34]; ªUniper SE.
Figure 14. Operating principle of the adiabatic compressed-air
accumulator Adele; ªDLR [35].
tion of liquid air. If oxygen and nitrogen are separated,
549 kWh are required to produce of 1 t of liquid nitrogen [36],
which again can be reduced by recycling cold released during
discharge.
By utilizing low-temperature heat and a turbine, the expan-
sion of liquid air can achieve a heat-to-electricity efficiency of
50–60 %. For grid-scale systems a heat to power efficiency of
around 75 % is expected. The high efficiency is made possible
by the low temperature of liquid air. To achieve a high power
output, either an external source of heat is required or heat
produced at the compression stages of the air liquefaction
process can be harvested and stored in thermal storage tanks
for later use, enhancing the power output (Fig. 16). A pilot
plant with an output of 350 kW has been successfully tested in
Slough, UK [36], but has now relocated to the University of
Birmingham. Highview and project part-
ners, Viridor, were awarded funding from
the UK government to build a 5 MW LAES
technology demonstration plant. The plant
is currently undergoing commissioning.
The pilot plant has successfully proven
the concept of LAES technology and has
already provided some services to the grid.
The new 5 MW demonstration plant is
expected to prove the technology at grid-
scale. In principle, both nitrogen and air
are feasible energy vectors. Since nitrogen
storage would require an additional air sep-
aration unit, storage of liquid air has a rela-
tive economic advantage.
Due to the high energy demand, air-sep-
aration units are also suitable for storing
energy in chemical processes (see Sect. 4.5).
4.1.5 Advantages of Mechanical
Storage Systems
Mechanical storage systems have a number
of advantages. They can be designed to
cover a wide range of power input/output
and storage capacity. The technologies are
very robust and highly stable. Especially for
the power grid they provide relatively inex-
pensive positive and negative balancing
power and allow a restart of the collapsed
power grid (black-start capability).
4.1.6 Disadvantages of Mechanical
Storage Systems
Mechanical storage systems have a relative-
ly low energy density. Therefore, they are
almost exclusively designed and operated
as large stationary installations (with the
exception of flywheels).
4.2 Technologies for the Storage of Energy
in the Form of Chemical Energy
For storing large quantities of power over extended periods
of weeks to months, conversion to chemical energy storage
is appropriate, mainly for reasons of convenience. The
exploitation of the energy of chemical bonds enables high
energy densities to be maintained, together with the possibili-
ty for storage in geological formations, which has advantages
over other storage technologies in terms of space and cost.
While various conversion chains are conceivable, the current
debate focuses on hydrogen and its resulting conversion
chains. The central underlying technology is water electroly-
sis, which employs electricity to split water into hydrogen
and oxygen.
www.ChemBioEngRev.de ª2017 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA ChemBioEng Rev 2017,4, No. 3, 144–210 165
High performance flywheel T4
Steel sha with bearings
Burst
protecon ring
made of CFRP
Stator winding for
power up to 200
kW water-cooled
Rotor with
NbFeB-magnets
bandaged with
CFRP
Flywheel made of
carbon-fiber reinforced
composite, rotates up to
50.000 revoluons per
minute and stores 2 kWh
Vacuum casing for
pressures down to
0.001 mbar
Water cooling
Flywheel
Power
electronics
Incoming
supply secon
Power supply
400 V, 220 V, 24 V
(masked)
Brake resistor
a)
b)
Figure 15. (a) Cross-section through a flywheel accumulator and (b) example of its tech-
nical implementation; ªrosetta GmbH [37].
4.2.1 Production of Hydrogen from Renewable Energy
in Electrolytic Processes
For the production of hydrogen from water and electricity var-
ious electrochemical processes are possible. All are based on
the splitting of water into its hydrogen and oxygen constituents
by applying an electrical voltage. Thus, this is technically com-
parable to a fuel-cell system operating in reverse direction to
the current flow. The electrolytic processes considered here are
alkaline, proton-exchange membrane (PEM), and high-tem-
perature electrolysis. The costs of hydrogen produced in this
manner are dependent on the investment costs of the electro-
lyzer, the scenario under consideration, the resulting full-load
hours per year, and the operating costs, especially the costs of
procuring the required electricity.
Alkaline Electrolysis
Alkaline electrolysis is already used for the pro-
duction of hydrogen on an industrial scale.
Production rates can reach up to 740 m
3
h
–1
per
stack and up to 30 000 m
3
h
–1
for total plants
[15]. In industrial settings, 20–40 % KOH solu-
tion is used at an operating temperature of
70–90 C to reduce the overpotential [40]. The
anode and cathode areas are separated by a dia-
phragm, allowing the KOH solution to trans-
port the OH
–
ions. At the cathode, water is
cleaved into H
2
and OH
–
, and the latter is con-
verted to O
2
and H
2
O at the anode. Catalysts
are usually composed of nickel, cobalt or iron.
The current densities present under operating
conditions are generally between 0.2 and
0.4 A cm
–2
[40].
Anode: 2OH
fi0:5O
2þH2Oþ2e
Cathode: 2H
2Oþ2e