Fraunhofer ISE Study Paths To A Climate Neutral Energy System

FRAUNHOFER INSTITUTE FOR SOLAR ENERGY SYSTEMS ISE
Paths to a Climate-Neutral Energy System

The German Energy Transition
in its Social Context
27%
18%
21%
12% 37%
6%
ANNUAL
GROWTH
23%
H2
GAS
BHKW 12%
PATHS TO A CLIMATE-NEUTRAL
ENERGY SYSTEM
The German Energy Transition
in its Social Context
Philip Sterchele, Julian Brandes, Judith Heilig,

Daniel Wrede, Christoph Kost, Thomas Schlegl,
Andreas Bett, Hans-Martin Henning
Fraunhofer Institute for Solar Energy Systems ISE, Freiburg
Freiburg, February 2020

2
Foreword
The study “Paths to a Climate-Neutral Energy System – The German Energy Transition in
its Social Context” was conducted by the Fraunhofer Institute for Solar Energy Systems
ISE as an in-house research project. Part of the work for this study was performed within
the Kopernikus ENavi project, financed by the German Federal Ministry of Education and
Research. The main motivation behind the study was to investigate how specific social be-
havior and attitudes influence the course of the energy transformation and what effect this
has on the investments and costs required for the transition of the German energy system.
We would like to thank all the employees of Fraunhofer ISE who assisted in this study.
This group of participants is far larger than the authors named on the title page. The vast
expertise on the development of the future costs and performance of significant techno
logies within the institute was extremely important and helpful and provided a solid basis
for the simulations.
With the analysis and results presented here, we hope to make a valuable contribution
to an objective, factual discussion on the feasibility of a climate-neutral energy system,
based on the use of renewable energy and the provision of higher efficiency for energy
conversion and utilization.
Freiburg, Germany
February 13, 2020
Hans-Martin Henning Andreas Bett
Fraunhofer ISE | Paths to a Climate-Neutral Energy System 3

Summary
In this study, potential development paths for the German energy system were examined,
which lead to a reduction of at least 95 % of energy-related CO2 emissions, compared to
1990 levels, by 2050. Two additional scenarios with the aim of achieving 100 % reduction
in energy-related CO2 emissions by 2050 were also considered.
Besides the questions of technical feasibility and costs, societal behavior plays a significant
role in determining whether and in what form the energy transition can be implemented.
To take this aspect into account, four main scenarios, each describing different behavior
and attitudes of society, were analyzed. The societal aspects play a dominant role in the
various scenarios and thus set the framework for the further development of the energy
transition.
The Persistence scenario is characterized, in particular, by strong resistance to the use of

new technologies in the private sector. The Non-Acceptance scenario is defined by a lack
of acceptance among the population for the further expansion of large infrastructures.
By contrast, the Sufficiency scenario describes a development in which behavioral changes
are visible in a large majority of the population, resulting in a noticeable decrease in energy
demand. These three scenarios are compared to the Reference scenario, in which no condi-
tions to either promote or impede achieving the climate targets are specified.
The simulation and optimization of the various scenarios are carried out using the energy
system model REMod (Renewable Energy Model). This model was developed almost ten
years ago by the Fraunhofer Institute of Solar Energy Systems ISE and since then has been
intensively further developed and optimized.
The most important results and findings derived from the simulations are as follows:
1.
From a technical and systemic point of view, the achievement of the climate
protection targets in the German energy supply is feasible on the basis of
renewable energy sources.
Using dynamic simulations carried out in hourly time steps from 2020 up to 2050, the
model showed that a secure energy supply is guaranteed hour-by-hour in all consumption
sectors despite a high share of variable renewable energy sources in the electricity mix.
At the same time, the results from the various scenarios demonstrate that the differences
in expenditure and costs are strongly dependent on the framework conditions.
2.
Electricity from renewable energy sources will dominate the energy supply, with
wind and solar being the largest primary energy sources.
In the scenarios considered, the total installed capacity of wind and photovoltaic power
plants combined lies between 500 GWel and over 750 GWel in 2050, or about five to
seven-fold of the installed capacity today. All of the development scenarios considered
show that solar and wind power plants shall cover between 50 % and 60 % of the primary
energy demand in 2050.
4
3.
Due to lower losses in the conversion chains, mostly as a result of sector coupling,
the primary energy consumption decreases significantly.
In the investigated scenarios, the primary energy consumption for the energy sector in
2050 lies between 1750 TWh (Sufficiency scenario) and 2500 TWh (Persistence scenario),
thus appreciably lower than today’s value (almost 3400 TWh). This is despite the assumed
increase in useful energy consumption over time considered in most of the scenarios. The
reduction is mostly due to a more or less pronounced displacement of combustion-based
technologies (boilers, thermal power plants, internal combustion engines) by electricity-
based technologies and the resulting improvements in conversion efficiency throughout all
sectors. Therefore, sector coupling, i.e. the increased direct – or in the case of synthetic en-
ergy carriers produced by renewable electricity – indirect use of electricity, is a key element
of the energy system transformation throughout all consumption sectors.
Considering the above (points 2 and 3), it can be concluded that the core building blocks
needed for reaching the CO2 emission targets in the most cost-effective manner are: the
successive reduction of fossil fuel use in the heat (building, processes) and transport sectors
in conjunction with a greater use of electricity and an accelerated reduction of specific
emissions from electricity generation.
4. Greater flexibility in the generation and use of electricity becomes a key element
of the system development
The growing share of variable renewable energy sources for power generation necessitates
a paradigm shift in the supply model. The predominantly demand-based power supply
relying on large power plants is increasingly being replaced with a system in which con-
tinuous energy balancing is carried out between energy supply from renewable energy
sources, characterized by limited controllability and predictability, and a consumption that
should be as flexible as possible. This results in a complex interplay between energy supply
and time-adjusted (through load shifting, demand response) energy consumption, including
stronger coupling of the electricity, heat and transport sectors as well as temporary use
of flexible production systems such as electrolyzers for hydrogen production and storage
systems of all types.
Multimodal heating networks fed from different generators in combination with large
heat storage units have also proved to be an effective measure for flexibility and load
management, especially in dense urban areas.
Stationary battery storage facilities are another important element of flexibility. For the
scenarios investigated the installed capacity of battery storage in 2050 lies between
50 GWhel and 400 GWhel.
Last but not least, controllable power generators are also needed in the long term,
especially highly flexible gas turbines with a total installed capacity between 100 GWel
and over 150 GWel.
5.
Electrolysis combined with hydrogen utilization for various applications is a key
building block of the future energy supply.
Electrolyzers can be used to produce electricity-based energy carriers such as hydrogen,
methane or liquid fuels, making them an important option for renewable electricity use.
Although these plants achieve higher full-load hours at foreign locations with a higher
availability of renewable power than Germany, their installation in Germany has proved
advantageous in the context of a cost-optimized energy system transformation. There are
three main reasons for this: Firstly, the secondary energy carriers produced can be used in
applications where a direct use of electricity is difficult to implement. Examples of this are
liquid fuels for maritime transport, air traffic or heavy-duty transport as well as hydrogen
and hydrocarbons for the chemical industry.

Secondly, those electrolyzers that enable rapid start-up and shut-down can be used as
flexible loads, thus furthering the potential for integrating renewable electricity into the
energy system.
Thirdly, these processes offer the possibility of producing electricity-based energy carriers
from renewables during times of high electricity feed-in. These secondary energy carriers
can then be stored nearly loss-free over a period of several days or months. Subsequently
they can be used, for example, in dispatchable power plants when grid feed-in from re-
newable electricity is low. Thus, a reliable power production is guaranteed without the
use of fossil fuels. For the scenarios investigated, the installed capacity of electrolyzers
ranges between around 50 GWel and 120 GWel in 2050.
6.
The efficient use of energy and a reduction in consumption resulting from
implemented technical measures contributes significantly to reaching the
climate targets, especially in the area of heat provision.
In all of the investigated scenarios, increasing the rate of energy retrofits in buildings was
shown to be an important factor in decreasing the space heating requirement. This goes
hand in hand with the conversion of many heating systems to a lower temperature level,
which accommodates the use of heat pumps and solar thermal systems. There are also
numerous possibilities in the industry for more efficient energy utilization, which in part
can be realized by the direct use of electricity.
7.
Behavioral changes in a majority of society that result in greater energy
conservation can have a significant effect on the restructuration of the
energy system and substantially reduce costs.
In the Sufficiency scenario, a development was investigated in which behavioral changes in
a large part of society resulted in considerable reductions in energy consumption. A shift
in values, driven by a growing awareness of the dangers posed by climate change, could
bring about such behavioral changes. If this were to occur, a significantly smaller number
of technical systems for the conversion, storage, distribution and consumption of renew
able energy would be required. As a result, the necessary investments and costs would be
lower. With an average value of about 50 € per tonne CO2 over the next thirty years, the
carbon avoidance costs in this scenario are far lower than for all other models calculated in
this study. At the same time, it can be assumed that smaller expansion of renewable energy
plants and other technical facilities in the energy system would also lead to greater accept-
ance of the changes associated with the energy system transformation.
8.
The import of energy in the form of electricity and synthetic chemical energy
carriers produced abroad with renewable electricity is an important part of
reaching the German climate protection targets.
In the scenarios investigated, the amount of imported synthetic chemical energy carriers
produced with renewable electricity abroad and consumed in 2050 varies between ca.
75 TWh (Sufficiency scenario) and 500 TWh (Persistence scenario). Naturally, these values
are strongly dependent on the assumed price development for such energy carriers.
Although the absolute quantities are subject to a great deal of uncertainty, the large
difference resulting from different social behavior and attitudes is clear. In particular,
an adherence to the use of familiar combustion-based technologies for heat supply and
transport (Persistence scenario) results in large import quantities of such energy carriers.
6
9.
From a system perspective, it would be possible to increase the use of photovoltaics
if wind capacity is not expanded to its optimal level. However, this would lead to a
greater need for energy storage and also to higher CO2 avoidance costs than would
be the case for a cost-optimal development.
In the Reference scenario, which assumes cost optimization without external intervention,
total onshore and offshore wind capacity is around 40 % (263 GWel) and photovoltaic
capacity is around 60 % (414 GWel) of the total installed capacity of plants generating
electricity from variable renewable energy sources. This means that about two thirds of
the total electricity generated by these plants is from wind turbines and around one third
from photovoltaics. On the other hand, in the Non-Acceptance scenario, which is char-
acterized by a strong resistance to further expansion of large infrastructures, the total
installed wind capacity (on and offshore) is 115 GWel while the photovoltaic capacity
increases to 645 GWel. As a result, a number of different measures are required to integrate
the increased solar power into the energy system in a useful way also for periods of high
solar radiation in Germany. For example, the orientation of photovoltaic installations shall
be distributed from east to west in order to widen the feed-in profile over the course of
the day and avoid an extreme midday peak. At the same time, with 400 GWhel of installed
capacity considerably more stationary battery storage units will be installed by 2050 than
in the Reference scenario (150 GWhel). Furthermore, there will be 300 TWh of imported
chemical energy sources, i.e. around twice as much, and the average CO2 avoidance costs
will increase from around € 150 per tonne (Reference scenario) to € 162 per tonne over the
entire period under consideration.
10.
The transformation of the energy system involves additional costs compared to
a development that is not oriented towards compliance with climate protection
targets. A large part of these additional costs are for investments that are necessary
for the development and reconstruction of the energy system..
Compared to a Business-as-Usual (BAU) scenario, the net additional expenditures range
between € 440 billion for the Sufficiency scenario and € 2330 billion for the Persistence
scenario. These results underline that societal behavior and attitudes have a significant
influence on the costs associated with the transformation of the energy system towards a
climate-neutral energy supply. In the scenarios, the annual additional expenditures range
from 0.4 % (Sufficiency scenario) to around 1.5 % (Reference and Non-Acceptance scenar-
ios) to around 2 % (Persistence scenario) of Germany’s 2019 gross domestic product (GDP).
Another comparative measure for the annual additional costs is the turnover attributed to
the Christmas business, which in Germany for 2019 was just under € 102 billion, i.e. about
twice as high as the average annual expenditure for the energy system transformation in
the Reference and Non-Acceptance scenarios.
The results show that in almost all scenarios far more than half of the additional expend-
iture is incurred for investments necessary for restructuring the energy system. Once this
system transformation is (for the most part) completed in 2050, investments will decrease
significantly, since from then on only replacement investments will be necessary.
When considering costs, it should also be noted that the analysis carried out here did
not consider external costs for the different scenarios, nor did it consider the total macro
economic impact, which include added value and employment effects. In particular, it does
not take into account the costs and effects associated, for example, with a sharp rise in
atmospheric temperature that would result if climate protection targets are not met.

8
Table of Contents
Foreword 3
Summary 4
1 Introduction 10
2 Methodology 12
2.1 The Renewable Energy Model 12
2.2 Setting the Climate Policy Targets 17
2.3 Construction of Consistent Scenarios 18
2.3.1 Reference Scenario 19
2.3.2 Persistence Scenario 20
2.3.3 Non-Acceptance Scenario 20
2.3.4 Sufficiency Scenario 20
2.3.5 Reference100 Scenario 21
2.3.6 Sufficiency2035 Scenario 21
3 Results 22
3.1 Transformation of the Energy Supply 22
3.1.1 Primary Energy and Final Energy 22
3.1.2 Development of Electricity Demand 23
3.1.3 Electricity Generation 25
3.1.4 Balancing Variable Electricity Generation 27
3.1.5 Fossil and Synthetic Energy Carriers 32
3.2 Sector-specific Analysis 35
3.2.1 Industrial Process Heat 35
3.2.2 Transport 38
3.2.3 Building Heat 41
3.2.4 CO2 Reduction by Sector 46
3.3 Complete Reduction of Energy-related CO2 Emissions 48
3.3.1 Complete Reduction of Energy-related CO2 Emissions by 2050 48
3.3.2 Complete Reduction of Energy-related CO2 Emissions by 2035 50
3.4 Cost Analysis 52
3.4.1 Total Cumulative Expenditure for the Energy Transformation 52
3.4.2 CO2 Avoidance Costs 56
4 Conclusion 58
Bibliography 60

1
Introduction
Globally occurring phenomena such as forest fires on an unprecedented scale, storms,
melting glaciers and the dramatic decline in glaciation in the northern polar region provide
strong indications of changes in the global climate that are very likely attributable to climate
change. At the same time, especially teenagers and young adults in many countries and
regions of the world are raising their voices with Fridays for Future and calling for a massive
policy changes and effective measures to significantly reduce the climate-relevant trace
gas emissions worldwide. It is becoming clear that an 80 % reduction in greenhouse gas
emissions in Germany compared with 1990 levels by the middle of the century will not be
sufficient. More stringent reductions are necessary from Germany in order to achieve the
climate goals agreed upon in December 2015 Paris accord and thus limit the global temper-
ature increase to an average of 2 °C, or better 1.5 °C, compared with pre-industrial levels.
The European Commission, newly formed in 2019, has also increased the European Union’s
climate protection targets, and the so-called “European Green Deal” aims to achieve a
climate-neutral Europe by 2050 [1].
The energy transition is a highly complex and ambitious large-scale project that mandates
a substantial restructuring of the present energy supply. In this study, we examine possible
development pathways for the German energy system starting from today up to the middle
of the century that comply with the ambitious goals mentioned above. To this end, this study
examines scenarios that lead to a reduction of energy-related CO2 emissions by 95 % and
100 % by 2050 (and in one case also by 2035) compared to the 1990 level. Today energy-
related CO2 emissions account for around 86 % of total German greenhouse gas emissions.
Therefore, restructuring the energy system to achieve a climate-neutral energy supply, the
so-called “Energiewende” is the most important measure towards reaching comprehensive
climate neutrality. The results of this study relate only to energy-related CO2 emissions,
since the simulation and optimization model REMod used to calculate possible transforma-
tion paths maps only the energy sector fully.
It is becoming increasingly obvious that, in addition to questions of technical feasibility

and costs, societal behavior plays a decisive role in whether and in what form the energy
transition can be implemented. Public attitudes vary over a broad spectrum: On the one
hand, more and more people are taking on behavior and lifestyles that follow the most
economical use of energy and the lowest possible greenhouse gas emissions. On the other
hand, in many places resistance is emerging to the further expansion of wind power or
electricity grids, i.e. infrastructures that are imperative for the successful transformation of
the energy system. Different trends can be observed even in private investment decisions,
which have significant consequences for energy consumption and CO2 emissions and are
often based on reasons beyond purely economic considerations. These range from a rather
persistent behavior that resists the use of new technologies (e.g. for the supply of heat to
buildings and in motorized private transport) up to initiatives, on the other hand, calling for
energy efficiency measures or the installation of renewable energy systems on the private,
community and civic levels.
In order to take this aspect into account, we have based the four main scenarios in this
study on different narratives in which certain behaviors and attitudes play a dominant
role. These provide then the framework for the development of the energy transition. The
Persistence scenario is characterized, in particular, by a strong resistance to the use of new
technologies in the private sector, e.g. by an insistence to combustion technologies both
for the heat supply of buildings and for motorized private transport. The Non-Acceptance
scenario is characterized, above all, by strong resistance to the further expansion of large
infrastructures, especially wind turbines and electricity grids, but also, the installation of
overhead contact lines above the autobahn, for example. The Sufficiency scenario describes
a development in which behavioral changes take effect in large parts of society, resulting
in a noticeable reduction in energy consumption. A widespread change in values, driven by
10
a growing awareness of the dangers posed by climate change, could, for example, bring
about such behavioral changes. These three scenarios are contrasted with a scenario in
which no specific conditions are defined that would promote or impede achievement of the
climate targets (Reference scenario).
To simulate and optimize the scenarios described above, REMod (Renewable Energy Model)
is used for the calculations. This model was created almost 10 years ago at the Fraunhofer
Institute for Solar Energy Systems ISE and has been intensively developed since then. RE-
Mod has been used to carry out numerous studies and analyses for various clients, includ-
ing model calculations for sector coupling within the framework of the BMBF-funded
academy project ESYS (Energy Systems of the Future).1 Since the publication of the study
“What Will the Energy Transformation Cost?” in 2015, significant extensions have been
added to the model, e.g. in the area of industrial processes. Since more and more new
technologies are reaching market maturity or are expected to do so in the near future,
REMod’s technology portfolio has been expanded and refined in step with these devel-
opments. For example, the cost and efficiency curves of all technologies were adapted to
current market analyses and the state-of-the-art research. For power plants and several
other system components, an optimized start-up behavior was mapped in the model. In
addition to the detailed mapping of demand sectors, mapping the imports of CO2-neutral
energy sources from abroad was improved. The available weather data sets were increased
to five, making the results more robust with regard to different weather conditions with
extreme events.
Following this introduction, the study is divided into two main chapters. The methodo-
logical approach is described in Chapter 2. This includes a short presentation of REMod,
followed by the prescribed paths for the reduction of energy-related CO2 emissions up to
2050. Then the scenarios mentioned above and the assumptions and boundary conditions
on which they are based are described. Chapter 3 provides a detailed presentation of the
results. In Chapter 3.1, the transformation paths in the various scenarios are compared,
followed by an analysis of the results for the different consumption sectors (industry,
transport, building heating) in Chapter 3.2. The results from these scenarios are based on
a 95 % reduction in energy-related CO2 emissions by 2050, compared to 1990 levels. For
two additional scenarios, calculations were carried out that model a complete, or 100 %,
reduction of energy-related CO2 emissions by 2050 and 2035 respectively (Chapter 3.3).
Finally, a cost analysis was performed for all six scenarios, ending with a calculation of the
CO2 avoidance costs (Chapter 3.4).
A detailed appendix with data on the performance and cost developments of all the tech-
nologies used in the model calculations as well as other data included in the calculations
is available on the Internet.2
1 The position paper “Sector Coupling - Options for the next phase of energy system transformation” and the
analysis “Sector Coupling - Studies and reflections on the development of an integrated energy system”,
published in 2017. Download at https://energiesysteme-zukunft.de/themen/sektorkopplung
2 For more information, see: https://www.ise.fraunhofer.de/klimaneutrales-energiesystem and https://www.
energy-charts.de

2
Methodology
The objective of the scenarios is to describe possible, consistent future developments
based on currently available knowledge, but not to predict future developments. The
transformation paths shown as the result of model calculations are based to the best of
our knowledge on the technologies available today and their cost and performance pro
jections. The main aim of modeling is to analyze, from a systems engineering perspective,
the development of overall systems that are constrained by a fixed upper limit on emissions
and to determine their associated costs. The optimization algorithm calculates the develop-
ment paths that lead to the lowest possible total costs. The configuration of the energy
system that is calculated can vary to a greater or lesser extent depending on the set bound-
ary conditions. To take the associated uncertainties into account, several consistent scenar-
ios are defined and analyzed. In this way, the effects of the specified boundary conditions
on the energy system can be shown, robust technological developments identified, and
recommendations for action derived. In contrast, the model does not allow for any state-
ments to be made about the appearance of future business models for market participants,
how prices will be set or how costs will be distributed among different social groups.
The Renewable Energy Model (REMod) developed by the Fraunhofer Institute for Solar
Energy System ISE was used to calculate the scenarios in this study. The basic idea of the
model is to identify technically and economically feasible transformation paths for the
German energy system. At the same time these paths shall comply with the climate policy
goals as defined for reducing energy-related CO2 emissions. By simultaneously optimizing
all sectors of the energy system (electricity, building heating, industrial process heat and
transport), the mutual influence of these sectors is taken into account. All relevant energy
sources, converters, storage facilities and all consumption sectors are mapped in the model.
The aim of the calculations is to describe for each scenario a cost-optimized path for the
transformation of the current energy system until 2050, as defined by given boundary
conditions and parameters. The final CO2 reduction target (e.g. 95 % reduction compared
to 1990) is defined by a CO2 path given for each year. This CO2 path represents the upper
limit of permitted CO2 emissions and is a fundamental boundary condition for optimization.
This chapter describes the structure and functionality of the model as well as the input
data and the specific assumptions and boundary conditions relevant for constructing the
scenarios.
2.1 The Renewable Energy Model

A central feature of REMod is the simultaneous optimization of all consumption sectors of
the energy system with a high temporal resolution [2, 3]. The simulation considers all hours,
from today up to 2050. The calculated scenarios represent the technically possible, cost-
optimized transformation paths of the German energy system that comply with a given
CO2 reduction path up to 2050. In addition to assumptions on the development of the
various sectors, plausible development forecasts on the costs and efficiencies of the con-
sidered technologies are stored. To create the scenarios, the model optimizes the energy
system development based on the available technologies so that all given boundary condi-
tions are met while simultaneously optimizing the costs to the greatest extent possible. The
hourly calculation ensures that all converters, storage devices and consumers are dimen-
sioned in such a way that each consumption sector is supplied every hour with sufficient
energy. For the optimization, technical components are accordingly removed or replaced by
alternative components (e.g. oil-fired boilers by heat pumps), taking their expected service
life into account. The consideration of different historical weather data (five years with
hourly values of all the relevant variables) ensures a statistical range that includes weather
extremes, e.g. years with longer phases of low renewable energy availability.
12
Figure 1 shows a schematic overview of the model, which illustrates the most important
energy conversion technologies and consumption sectors. They are broken down into the
main forms of (traditional) electricity applications (e.g. lighting, information and commu-
nication technology, refrigeration, mechanical energy), building heating (space heating,
domestic hot water), transport and industrial process heat.
Energy sources Energy conversion/storage Consumption

Regenerative Energy conversion Energy Consumption
Storage sectors
Wind Wind turbine
Sun Run-of-River PP Vehicle battery
Transport
Hydropower PV plants
Environmental Geothermics Stationary

heat battery
Solar thermal
energy Electricity
Biomass
Electrolysis Pumped
storage PP
Imports
Biofuel GAS
Electricity Process heat

Gas
Power-to-liquid
Synthetic
fuels
Conventional PP
H2
Fossil & nuclear Hydrogen
Building heat
CHP
Coal and oil Gas

H2 O2 Fuel cell
Heat
Biomass
Biomass
Natural gas Figure 1:
Nuclear PP Liquid Fuels
Schematic representation of
Hydrogen
the REMod model (PP: power
Nuclear Methanization Electricity plant, CHP: combined heat
Heat Coal and power generation) [3].
The input data used for the calculations are divided into two groups of different time reso-
lutions. Five data sets based on real weather data for the years 2011 to 2015 are stored as
hourly profiles. These describe, among other things, the temporal course of outdoor tem-
perature, solar irradiation, feed-in profiles of weather-dependent renewable energy sources,
process heat demand or driving profiles in the transport sector. The second database con-
tains the technical and economic parameters of technologies and energy carriers as well as
boundary conditions for each year from 1990 to 2050. The inventory of the German energy
system is given from 1990 to 2019 and includes, e.g. age, efficiency and installed capacity
or number of power plants and other energy converters, storage facilities, heating tech-
nologies in buildings as well as the number of buildings and vehicles. For the optimization
process carried out in the model between 2020 and 2050, assumptions about the devel-
opment of the costs and efficiency of all available technologies as well as their technical
life time are used as input. Additionally, the cost development of fossil fuels and of energy
carriers produced abroad using renewable energy sources are assumed on a yearly basis.

Based on this database, the future development of the plant fleet across all sectors will
be optimized. This optimization includes conventional power plants, variable renewable
electricity generators, electrical storage facilities, thermal and chemical energy carriers, the
vehicle fleet divided into cars and trucks with different powertrain technologies as well as
different technologies for converting electricity into hydrogen (H2), methane (CH4), liquid
fuels or heat. The optimization is based on the assumption that all technical equipment will
be replaced at the end of its specified life time. At this point, the optimization algorithm
decides whether, for example, in the space heating sector, a gas boiler shall be replaced by
another gas boiler or, rather, e.g., by a heat pump. The power plant fleet is simulated using
an iterative optimization process constrained by boundary conditions based on the security
of supply (the energy balance must be met every hour), climate protection (the CO2 path
must comply with the specified target values) and economic efficiency (lowest possible
costs for the entire optimization period from 2020 to 2050).
The final energy demand is divided accordingly into the four sectors: building heating
(space heating, hot water), traditional electricity applications, industrial process heat and
transport.
In REMod, the hourly demand for space heating and hot water in the building sector
is calculated according to DIN EN 13790 [4] using the so-called “Simply-Hourly-Method”,
or SHM. Within this sub-model a distinction is made between residential and non-residen-
tial buildings that are subdivided into a total of 19 characteristic building types. In total
there are nine classifications for residential buildings (three age classes multiplied by three
building types) and ten classifications for non-residential buildings. The input data used
are the weather data from 2011 to 2015, building cubature, average living space or useful
area per building type, number of buildings and age-dependent heat transfer coefficients.
The values from the German industrial standard DIN EN 13790 are used for the energy
efficiency and the calculation of the building energy demand.
For space heating supply, a distinction is made between two classes of heating systems
with their corresponding temperature levels (standard heating system with radiators, low
temperature heating e.g. radiative heating). When combining these with the eleven tech-
nologies for supplying heat, there are twenty different heat supply options for the build-
ing sector. The available heat supply technologies include conventional boilers based on
methane, biomass or oil, engine-driven combined heat and power (CHP) units, electrical,
fuel-based and hybrid heat pumps (with air or ground as heat source) and fuel cell systems
based on methane or hydrogen. All technologies can be optionally supplemented with hot
water storage tanks and solar thermal collectors.
In addition to these systems, the model also includes the option of supplying heat via heat
grids. These can supply heat from geothermal or CHP plants, large electric heat pumps,
(peak load) gas boilers or also solar thermal collectors. In addition, large thermal energy
storage systems can be used in heat grids. Many of these technologies have the ability to
react flexibly to the requirements of the overall system. For example, if power availability is
low at a certain time, the thermal energy storage can be discharged instead of operating
large electric heat pumps. Conversely, large electric heat pumps may be put into operation
if there is a particularly high feed-in from photovoltaic or wind energy systems during the
hour under consideration. In this case, electric heat pumps or simple heating rods can also
be used to raise the temperature in thermal energy storage systems.
Traditional electricity applications describe the load occurring in the network today 3
excluding the electricity demand for heat and road transport. The load profiles in the
model are based on data from the European Network of Transmission System Operators
for Electricity (ENTSO-E), which also includes the electricity demand of rail transport and
3 Reference is hereby made to the year 2016, as the relevant statistical values are available for that year.
14
the mechanical energy industry. In addition to the base electricity load, the model also
takes into consideration additional electricity consumption in the course of future system
development, such as battery electric vehicles or electric heat pumps.
The electricity generators implemented are conventional power plants using lignite and
hard coal as fuel, nuclear power plants (until their phase out in 2022), oil-fired power
plants, gas turbines (with CH4 or H2 fuel), CHP plants, combined cycle gas turbine plants
(CCGT) and fuel cell systems based on hydrogen or methane. Renewable electricity can
be generated from onshore and offshore wind, photovoltaic and run-of-river power
plants. Photovoltaic plants are divided into three different types: ground-mounted systems
facing south and roof-mounted systems facing south, east or west. The differing costs
(ground-mounted are cheaper than roof-mounted systems) as well as various feed-in pro-
files based on orientation are considered. In addition, depending on the parameterization,
different interconnections are available for the import and export of electricity.
The energy demand for transport is first converted into traction energy (kinetic energy
necessary for the movement of vehicles) on the basis of BMWi data and then distributed to
each hour of the year using hourly resolved travel or demand profiles. The simulation model
then determines the final energy demand for the respective fuels or electricity using the
efficiency of the respective powertrain technologies. The energy demand for aviation and
shipping 4 as well as fuel-based railway traffic are also taken into account in the balance.
The transport sector is represented in detail by passenger cars and trucks (only one class
size each), which are each described by seven drivetrain concepts and the corresponding
energy demand. In addition to the currently established combustion engines using gasoline,
diesel or methane as fuel, battery and hydrogen-electric (fuel cell) powertrain concepts are
implemented. All concepts can also be part of the solution as plug-in hybrid variants. In
addition to the powertrain concepts mentioned above, overhead-line trucks can be used
as an option for freight transport, depending on the scenario framework. Battery-powered
powertrain concepts can be used to increase flexibility in the system, i.e. as electricity stor-
age that can be charged or discharged to provide grid support. The degree of usability of
this flexibility option is determined by a predefined proportion of users, who allow flexible
control of the battery up to a specified maximum depth of discharge.
Energy used to provide process heat in industry is divided into different temperature
levels. For the temperature range below 500 °C an hourly demand profile close to the base
load is assumed. For applications with a temperature higher than 500 °C, a constant ener-
gy demand is assumed. The total energy demand over one year is based on data from the
BMWi [5] and assumptions specific to each scenario.
The energy system can use electrical energy storage devices in the form of stationary
and mobile batteries in vehicles or pumped storage power plants. In addition, hydrogen
storage and thermal hot water storage of various sizes are taken into account. For methane
storage, the simplified assumption is made that the storage capacities already existing
today (including the grid, approx. 210 TWh [6]) will continue to be available to the system
in the future. Their future size is therefore not a result of the optimization.
Hydrogen can be used in transport, industry (materially or energetically to generate process

heat) or in buildings (by means of H2 fuel cells). In addition to feeding it into the natural gas
grid, the use of hydrogen is an option for re-generating electricity in gas turbines. Further-
more, hydrogen can be converted in methanation or Power-to-Liquid plants into synthetic
natural gas or liquid fuels respectively for use in power plants, heating technologies or for
transport. Hydrogen can either be imported or produced domestically from biomass, by
steam reforming of methane or by electrolysis from renewable energy. The capacities of the
respective technologies are determined in the model within the scope of the optimization.
4 The energy demand of aviation covers the aviation fueled in Germany. The energy demand of shipping only
includes domestic German transport.

Biomass can be used either directly or after conversion into another energy source. For
example, wood can be used in boilers to provide process heat for industry or to generate
low-temperature heat in the building sector. In addition, biogas plants, gasification plants
with subsequent synthesis into hydrogen, methane or liquid fuels as well as biodiesel plants
are implemented for the conversion of biomass.
In addition to domestic production, synthetic energy carriers produced from renewable

energy sources can be imported from abroad. Among these energy carriers are hydrogen,
methane and liquid fuels. A maximum available quantity with a specific price is specified
as an annual assumption. The specific price is based on the production costs and the costs
necessary for transporting the energy carrier. The amount of the available quantity that is
actually used each year is determined by the optimization in the model.
The mathematical optimization identifies cost-optimized transformation paths that com-

ply with the given boundary conditions such as the upper limit of permissible CO2 emis-
sions and other scenario-dependent conditions. The algorithm used for this purpose is the
so-called Covariance Matrix Adaptation Evolution Strategy procedure [7]. The term “cost
optimization” requires an explanation here: The calculations take all investments for new
plants and for the replacement of old plants (replacement investments) into account as well
as the corresponding capital costs (assumed interest rate of 7 %). The operating and main-
tenance costs for all components are also considered. Finally, costs for cultivated biomass
and imported energy carriers (electricity, fossil fuels and synthetic fuels produced abroad
from renewable energy sources) are included. The cost calculation is based on an annuity
method, whereby all payments are discounted in the future (assumed discount rate of 2 %).
For each of the scenarios investigated, the cumulative expenditure (investments and costs)
for the reconstruction, expansion and operation of the total energy system for the period
from 2020 to 2050 is determined and then minimized in the optimization process.
16
2.2
Setting the Climate Policy Targets
The model calculations in this study are based on the CO2 reduction targets of the German
government. Since the modeling only addresses the energy sector, only energy-related CO2
emissions are included in the balance. A reduction of energy-related CO2 emissions by 95 %
compared to 1990 levels was set as a minimum target, since this reduction is most compat-
ible with the European Commission’s Green Deal goal of a climate-neutral Europe by 2050.
Additionally two scenarios were calculated in this study that consider zero energy-related
CO2 emissions (100 % mitigation) by 2050.
In 2019, CO2 emissions in Germany were 35 % lower than in 1990.5 It is therefore unlikely
that in 2020 the target of a 40 % reduction in CO2 emissions compared to 1990 levels will
be be met [8]. To account for the resulting additional emissions, we used a combination of
a CO2 reduction path and a CO2 budget in this study. The reduction path is set so that CO2
emissions are reduced linearly from 2020 to 2030 in order to meet the reduction target of
Figure 2:
55 % by 2030. This means, however, that more emissions are released in the period from
Historical development of
2020 to 2030 than would have been the case if a reduction to 40 % had already been
energy-related CO2 emissi-
achieved in 2020 (see Figure 2, area between the blue and green lines between 2020 and
ons [5]; The German federal
2030). To compensate for these additional emissions, we adjusted the CO2 reduction target
government’s CO2 reduction
for 2040 and the CO2 reduction path from 2030 to 2050 so that the path is steeper than in
targets of 95 % compared to
the original plans of the German government. As a result, the sum of emissions from 2020
1990 levels [9] and the targets
to 2050 is equal to the sum in the federal government’s Energy Concept 2010 for a target
considered in the study up to
of 95 %.
2050.
Mio. t (Energy-related CO2 emissions)

1000
900
800
700
600
500
400
300
200
100
0
1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Energy-related CO2 emissions

100% target in 2035 100% target in 2050 95% target in 2050 Target of German Fed. Gov.
5 Achieved reduction of emissions in 2019 is based on [5]. Final values determined by the Federal Environment
Agency are not yet available for this year.

2.3
Construction of Consistent Scenarios
The current climate policy debate in Germany shows that the energy transformation is not
only a question of technical and economic feasibility, but also of social and political will. In
order to include effects of influencing variables such as consumer behavior, insistence on
using today’s technologies or willingness to change, acceptance of infrastructural changes
as well as aspects of sufficiency and reduction of consumption, four scenarios with a CO2
reduction target of 95 % are defined. In addition, two further scenarios are presented with
a CO2 reduction target of 100 %, which are derived from the main scenarios. The central
assumptions for these scenarios and their corresponding names, which are used hereafter,
are shown in Table 1.
Table 1: Description of key boundary conditions in the scenarios examined6 7
Sufficiency2035
Reference100
Acceptance
Persistence
Sufficiency
Reference
Non-
CO2 Reduction Targets 95 % 95 % 95 % 95 % 100 % 100 %
Upper Limit for In-
stalled Capacity in GWel
Photovoltaics 530 6 530 800 7 530 530 530
Wind Onshore 230 230 80 230 230 230
Wind Offshore 80 80 40 80 80 80
Electricity imports 40 40 20 40 40 40
Development in
Consumption
Traditional electricity constant constant constant decreasing constant decreasing
applications
Transport volume / increasing increasing increasing decreasing increasing decreasing
Guidelines
some without
internal overline
combusion trucks
Heated Building Spaces / increasing increasing increasing decreasing increasing decreasing
Guidelines
Some gas
boilers
Process Heat slightly slightly slightly decreasing slightly slightly
declining declining declining declining declining
The scenarios listed in Table 1 are described in more detail below. For all of the scenarios
(except the Reference scenario), only the assumptions that differ from the Reference scenar-
io are explained.
6 The 530 GW limit is made up of 143 GW for ground-mounted systems and 387 GW for roof-mounted sys-
tems. These values are based on the studies [10, 11]. However, there are different analyses of land potential,
from which other potential limits can also be derived..
7 According to a bottom-up analysis of the German building stock (residential, commercial) carried out at KIT,
roof areas in particular have a significantly higher potential than that indicated in almost all other sources
[12]. For the Non-acceptance scenario, the available potential was therefore increased to a total of 800 GW,
of which 190 GW are for ground-mounted systems and 610 GW for roof-mounted systems..
18
2.3.1 Reference Scenario
In the Reference scenario, the cost-optimal transformation of the energy system is exam-
ined without assuming pronounced social behavior with regard to acceptance of change or
altered consumption. The main assumptions that characterize this scenario are listed below
by sector:
Energy Supply: The technical installation potential of photovoltaic systems totals 530
GWel. This consists of 143 GWel ground-mounted systems and 387 GWel roof-mounted
systems, which can be oriented to the south or east and west. The installation potential for
onshore wind turbines was set at 230 GWel [13]. A high potential of 80 GWel was assumed
for the offshore wind energy installations. Coal-fired power generation will be phased
out by 2035 8 and nuclear power by 2022 [14]. The interconnector capacity for electricity
imports will increase from just under 17 GWel today to 40 GWel in 2050 [15]. Synthetic fuels
(hydrogen, methane and liquid energy carriers) produced abroad from renewable electricity
can be imported as of 2030.
Electricity Sector: The electricity demand for lighting, cooling, information and communi-
cations technology (ICT) and mechanical energy is constant up to 2050.
Transport: Moderate development. By 2050, the total mileage of motorized private transport
will increase by 3.5 % and that of road freight transport by 27 % [16]. An overhead-line
infrastructure is being built for electrified freight transport. Battery electric vehicles cannot
exceed 80 % of the annual new registrations.9 10 % of all users agree with the flexible
control of their car battery (grid-to-vehicle G2V, vehicle-to-grid V2G) to balance the electric-
ity load [17]. Air traffic, which is recorded in the balance sheet, will remain constant
until 2050.
Building Heat: Moderate development. Heated living area in Germany will increase by a
total of 8 % between today and 2050 [18]. Heat pumps cannot exceed 85 % of the annual
new installations.10 The maximum rate of refurbishment increases from 1 % to 3 % during
the period under consideration.
Industrial Process Heat: Moderate development. Due to technological innovations and

the exploitation of efficiency potential (best available technology), industrial process heat
demand is declining continuously by 0.5 % per year [19].
Based on these assumptions, three further scenarios are defined in which the influencing
factors are each specifically changed.
8 By 2038 at the latest, no coal-fired power plants are to be on the grid in Germany. In 2032, it is to be
reviewed whether the phase-out date can be brought forward to 2035 at the earliest in agreement with the
operators. At the time of preparing the calculations for this study, no final decision had yet been made.
9 This assumption is a settlement. It is assumed that purely battery-electric drives will not cover the entire car
population, as there will always be users who for various reasons decide to use other forms of drive (e.g.
range, duration of charging, others).
10 This assumption is also a settlement. It is assumed that electric heat pumps will not cover the entire stock of
heating systems, as there will always be users who for various reasons decide to use other heating technolo-
gies (e.g. temperature requirements in historical buildings, lack of availability of suitable heat sources, others).

2.3.2
Persistence Scenario
In this scenario it is assumed that the willingness of the population to switch to new
technologies is very limited. This means a significant delay in the deployment of new
technologies and an insistence on continuing to use today’s conventional technologies,
i.e. especially heating systems and vehicles with internal combustion engines. The main
changes compared to the Reference scenario are the following:
Transport: Conservative development. Conventional internal combustion engines based

on liquid fuels account for at least half of the annual new registrations in motorized private
transport. There is no willingness in the population to flexibly control the batteries of elec-
tric cars in order to balance the electricity load.
Building Heat: Conservative development. Gas boilers account for at least half of new
installations each year, while electric heat pumps are limited to 20 %. The rate of energy-
related building refurbishments is constant at 1 %.
2.3.3
Non-Acceptance Scenario
Due to complex approval procedures and local resistance to the expansion of large infra-
structures (i.e. transmission grids or wind turbines), their expansion may be significantly
smaller than that corresponding to a cost-optimal development. In this scenario, a devel-
opment is assumed in which larger infrastructure projects cannot contribute to the trans-
formation of the energy system to the extent previously assumed. The most important
changes compared to the Reference scenario are the following:
Energy Provision: The technical installation potential of onshore wind turbines is reduced to
80 GWel [13]. A potential of 40 GWel is assumed for offshore wind energy. To ensure that
renewable energy sources continue to make a significant contribution to the electricity sup-
ply, the technical installation potential of photovoltaics is increased to 800 GWel [12] (See
comments about Table 1). The interconnection capacity for electricity imports will rise to
only 20 GWel by 2050, since here too it was assumed that the corresponding grid expan-
sion would not be accepted.
Transport: For freight transport, there is no option to extend overhead lines on motorways.
Only a small part of the total mileage in freight transport can be covered by battery electric
vehicles.
2.3.4
Sufficiency Scenario
In this scenario, the influence of behavioral changes on a broad scale is examined. These
changes are large enough to lead to a significant reduction in energy consumption. This
beneficial effect could result, for example, from a strong change in values due to the
increasing number of extreme events caused by climate change. The main changes in
the Sufficiency scenario compared to the Reference scenario are based on the study com-
missioned by the Federal Environment Agency “Improving Climate Protection Modelling
with Sufficiency” [20] and [21] and are summarized below:
Electricity Demand: It is assumed that the electricity demand for lighting, cooling, ICT
and mechanical energy will decrease by 45 % compared to today’s level.
Transport: Efficient development. Based on today’s traffic volume, motorized private trans-
port and air traffic are expected to decline by 30 % and 55 % respectively by 2050. Total
mileage in freight transport will remain at today’s level. In addition, end users are expected
to be more willing to flexibly control their car batteries for load balancing.
20
Building Heat: Efficient development. The minimum rate of building energy retrofits
increases from 1 % in 2020 to 2 % in 2050, while the maximum possible rate of energy
refurbishment rises from 1 % to 3 % over the period under consideration.
Industrial process heat: Efficient development. The demand for industrial process heat
will decline continuously by 0.75 % per year until 2050.
2.3.5
Reference100 Scenario
The Reference100 scenario is based on the Reference scenario. However, a complete
reduction of energy-related CO2 emissions by 2050 is specified here as the boundary
condition (i.e. minus 100 % relative to the 1990 value).
2.3.6
Sufficiency2035 Scenario
The Sufficiency2035 scenario is based on the Sufficiency scenario. However, in this
scenario it is assumed that a complete reduction in energy-related CO2 emissions must
be achieved as early as 2035.

3
Results
This chapter presents the results of the model calculations for the scenarios described
above with a target of 95 % reduction in energy-related CO2 emissions by 2050. First the
energy provision is analyzed (Section 3.1) and then the individual consumption sectors of
process heat, transport and building heating are discussed (Section 3.2). This is followed
by a presentation of the results of the model calculations with a complete reduction of
energy-related CO2 emissions (Section 3.3) and finally a section dealing with cost aspects
(Section 3.4).
3.1
Transformation of the Energy Supply
The aim of this chapter is to investigate the energy supply in the changing energy system.
A presentation of the composition of primary energy and final energy in comparison to
today’s situation is followed by a more detailed analysis of the electricity generation, as
electricity from renewable energy sources is becoming the most important primary energy.
3.1.1 Primary Energy and Final Energy

Germany’s primary energy consumption in 2018 was 3641 TWh [5]. Excluding the non-
energetic energy consumption of 247 TWh, 3394 TWh remain that are attributed to final
energy applications and losses in the energy sector. The breakdown between the main
energy sources (fossil, nuclear, biomass, electricity from variable renewable energy (VRE)
and other renewable energy carriers) is shown in Figure 3 above. The final energy consump-
tion was 2494 TWh and is also shown in the figure, divided into the main applications of
low temperature heat (LT low temperature heat; space heating and hot water), heat for
industrial processes (industrial PH), transport and the traditional applications of electricity
(traditional electricity). The conversion efficiency is thus 73 %; 27 % are the losses in the
energy sector. The largest share can be attributed to the conversion of fuels into electricity
in conventional power plants.
In all of the four scenarios considered, primary energy consumption in 2050 is significantly
lower than today, ranging from 52 % (Sufficiency) to 73 % (Persistence) of the current level.
The reasons for the lower values are, on the one hand, the lower final energy de-mand
and, on the other hand, the significantly reduced losses in the energy sector. The reduction
in losses is due to the fact that only a very small proportion of electricity is generated in
thermal power plants with corresponding losses. Depending on the scenario, the conver-
sion efficiency is between 84 % and 92 %. Currently, the total renewable energy share in
primary energy is around 15 %. In the four scenarios considered the share is consistently
around 90 %, with variable renewable electricity, i.e. photovoltaics and wind energy,
accounting for 50 % to 60 %.
The final energy of the four scenarios examined is between 60 % (Sufficiency) and 90 %
(Persistence) of today’s value. This reduction is due in particular to high efficiency on the
use side of heat pumps and of electric drivetrains in road transport, and to lower space
heating requirements due to energy-related refurbishment. Since all these measures have
a much lower impact in the Persistence scenario, the reduction is noticeably smaller. In
the Sufficiency scenario, the reduction in consumption due to behavioral changes has an
additional mitigating effect, which explains the lowest value for final energy in 2050 in a
comparison of the four scenarios.
22
Primary energy
2018
Final energy
73% Figure 3:
Primary energy mix subdivid
Primary energy ed into the primary energy
Reference
Final energy sources, and the final energy
86% subdivided into the main areas
of application today for the
Primary energy
Persistence four scenarios which lead to
Final energy a reduction of 95 % energy-
89% related CO2 emissions by 2050
(VRE electricity: electricity from
Primary energy
Non-Acceptance variable renewable energy
Final energy sources (wind, sun); other RE:
92%
other renewables including
geothermal, solar thermal and
Primary energy
Sufficiency environmental heat; LT heat:
Final energy low-temperature heat, i.e.
84%
space heating and hot water in
0 500 1000 1500 2000 2500 3000 3500 4000
the building sector; industrial
Energy in TWh PH: process heat in industry;
trad. Electricity: traditional
VRE-Electricity Biomass Other RE Fossil Nuclear electricity applications such
LT-Heat Industry-PH Transport Traditional Electricity as lighting, ICT, refrigeration,
stationary motors)
3.1.2 Development of Electricity Demand

The electricity sector currently accounts for around 40 % of energy-related CO2 emissions
in Germany [22]. In order to reduce the approximately 60 % of emissions from the areas of
process heat, transport and building heat, a fundamental transformation away from fossil
fuel use to renewables is necessary. As shown in the results in Figure 3, the greatest poten-
tial for renewable energy sources lies in wind energy and solar energy. Therefore a greater
use of electricity for heating and transport is a sensible and necessary measure.
The decline of conventional fossil-fuel powered technologies in favor of electricity-based

alternatives can be illustrated by the degree of electrification. This describes the proportion
of energy demand in the individual sectors (traction energy in the transport sector or space
heating in the building sector) that is covered by electricity-based technologies. This includes
both the share of final energy used directly as electricity in the respective sectors as well as
the share used indirectly via electricity-based fuels. A vehicle with an internal combustion
engine thus increases the degree of electrification once it uses an electricity-based fuel.
Compared to process heat, the degree of electrification in 2019 in road transport (less than
1 %) and in the supply of heat for buildings (4 %) is still very low. The reason for this is the
high proportion of liquid burners and gas and oil boilers. Figure 4 shows the degree of
electrification of the process heat, transport and space heating sectors for the years 2030
and 2050 in the scenarios examined.

% (Degree of electrification)
100
80
60
Figure 4:
Development of the degree of 40
electrification for the supply
20
of process heat, space heating
(including domestic hot water) 0 2030 2050 2030 2050 2030 2050 2030 2050
and road transport. In addition
Reference Persistence Non-Acceptance Sufficiency
to direct electricity use, the
degree of electrification also
takes into account the use of Heat, Industry Heat, Buildings Road Transport
electricity-based fuels.
The results show that electrification of the sectors in the Reference scenario is implemented
first for the supply of building heat (space heating and domestic hot water), followed by road
transport and process heat. The high degree of electrification, including road transport and
process heat provision, shows that the use of electricity will become increasingly important in
these sectors by 2050. This change leads to a corresponding increase in electricity demand,
which is summarized for the four scenarios in years 2030 and 2050 in Table 2.
Table 2: Annual electricity demand of the four scenarios analyzed in the years 2030 and
2050, including the electricity demand of Power-to-X technologies
Electricity Demand in TWhel Reference Persistence Non-Acceptance Sufficiency

2030 683 718 656 546
2050 1447 1464 1282 1068
In 2030 the Sufficiency scenario shows the lowest electricity demand with 483 TWhel,
which is about 120 TWhel less than demand in 2018 (597 TWhel) [23]. This is due to two
opposing effects: On the one hand, the increase in power-based technologies leads to
an increase in electricity demand. On the other hand, total energy demand is significantly
reduced by behavioral changes, including a reduction in distances travelled in road trans-
port, in heated living spaces and in the consumption of industrial process heat. In all other
scenarios, however, the demand for electricity increases continuously. In the Sufficiency
scenario the electricity demand in 2050 is 1.7 times today’s demand and in the Persistence
scenario just under 2.5 times (See Table 2). This electricity demand includes both direct
electricity and the indirect electricity needed for domestic hydrogen conversion and other
synthetic energy carriers (Power-to-Gas, Power-to-Liquid). The indirect electricity use will
account for about a quarter of the stated electricity demand in 2050.
24
3.1.3 Electricity Generation
An increase in electricity in the energy system only makes sense if electricity demand is
met as emission-free as possible. An overview of the technologies in the electricity supply
(in TWhel per year) is shown in Figure 5 for the Reference scenario.
TWhel (Electricity supply)
1500
1250 Photovoltaics Gas turbine PP
Onshore wind CHP (Heat grid)

1000
Offshore wind CHP (decentral)
750 Run-of-river PP Oil PP
500 Electricity imports Nuclear PP

Figure 5:
Combined cycle Lignite PP Development of the electricity
250
gas turbine (CCGT)
Hard coal PP supply by technology in the
PP
0 Reference scenario
2020 2030 2040 2050
(PP = power plant)
Today, a large part of the total electricity demand, which is currently about 597 TWhel,
is met by nuclear power plants as well as lignite and hard coal-fired power plants. When
the calculations were performed for this study, the decision has been made to phase-out
nuclear power by 2022; however, the law to phase-out coal-fired power generation by
2038 at the latest had not yet been adopted [24]. The calculations assumed a more am-
bitious coal phase-out by 2035. There will be a slight increase in electricity imports from
around 2030 onwards, in addition to the strong expansion of wind and solar power plants
to replace the lost electricity. An additional compensatory effect has the electricity genera-
tion by decentralized CHP and fuel cells as well as large combined cycle gas plants (CCGT),
which operate in combined heat and power generation and will contribute to the heat
supply in heat grids from around 2030 onwards.
An overview of the share of renewable energy in electricity generation and the correspond-
ing CO2 factors for all four scenarios is shown in Table 3. The share of variable renewable
energy sources in the electricity supply increases continuously over the period under con-
sideration. In the Reference scenario, the renewable share rises from almost 40 % [25] in
2019, to 71 % in 2030 and to 93 % in 2050. One of the consequences of this transforma-
tion is that the CO2 factor of electricity, which starts at 462 gCO2 /kWhel in 2018 [26], drops
to less than half at 147 gCO2 /kWhel in 2030. The electricity generation is nearly without
CO2 emissions in 2050, with the CO2 factor of electricity decreasing to 3 gCO2 /kWhel.
Table 3: Composition of the electricity mix in 2030 and 2050 for the four scenarios ana-
lyzed (RES: Renewable Energy Systems).

Share RES CO2 Factor Share RES CO2 Factor Share RES CO2 Factor Share RES CO2 Factor
% gCO2 /kWhel % gCO2 /kWhel % gCO2 /kWhel % gCO2 /kWhel
2030 71 147 69 149 70 154 65 200
2050 93 3 93 5 92 7 92 9

In all of the scenarios, the CO2 factor of electricity has similar values in the years 2030 with
about 150 gCO2 /kWhel and in 2050 with about 5 gCO2 /kWhel. To meet the CO2 reduction
target, a renewable energy share of more than 70 % in the electricity mix is calculated for
all scenarios in 2030 without assuming a decrease in consumption (Sufficiency). The share
of renewable energy set by the German government is 65 % in 2030 [27]. The fact that
this value is exceeded in all scenarios suggests that it will not be sufficient to achieve the
targeted 55 % reduction in CO2 emissions compared to 1990. Even in 2050, the target of
an 80 % renewable energy share [27] in the electricity supply is clearly exceeded in all four
scenarios; the average share is calculated to be 93 %. Electricity from wind and solar energy
play a central role here.
Figure 6 shows the cumulative installed capacity for onshore and offshore wind turbines
and photovoltaic systems, which are classified into east/west facing and south facing
roof-mounted systems and south facing ground-mounted systems.
GWel (Installed capacity)

800
700
600
500 PV (2019)
PV ground, south
400
PV roof, east-west
300
PV roof, south
200
Onshore wind
100
Figure 6: Offshore wind
Cumulative installed capacity 0
e
ce
ce
ce
y
y
nc
nc
nc
nc
nc
of photovoltaic systems and

n
an
ta
cie
re
te
re
te
cie
pt
ep
fe
fe
sis
sis
ffi
ffi
ce
Re
Re
wind turbines in 2030 and

c
r
Su
Su
Pe
Pe
Ac
Ac
n-
n-
2050 for the four scenarios

No
No
examined 2019 2030 2050
Around 54 GWel onshore and 8 GWel offshore wind and 49 GWel photovoltaic plants are
installed in Germany in 2019 [25]. This amounts to a total capacity of 111 GWel of variable
renewable energy sources for electricity generation. The results of the scenarios show a
doubling in the installed capacity by 2030 – with the exception of the Sufficiency scenario.
In the Reference scenario, this means an average net increase in photovoltaic capacity of
around 9 GWel per year, as well as an annual increase of 3 GWel for onshore wind and
1.5 GWel for offshore wind, based on current values.
In the Reference and Persistence scenarios, the installed capacity of wind energy (onshore
and offshore) increases to around 260 GWel and that of photovoltaics to around 415 GWel
in 2050. This amounts to a total electricity generation of around 1300 TWhel from photo-
voltaics and wind energy systems. In the Sufficiency scenario, the installed capacity is about
480 GWel, or about 200 GWel lower. This means that almost 30 % less electricity is provided
by RES. Due to the lack of acceptance for changes in large infrastructures assumed in the
Non-Acceptance scenario, about 140 GWel less wind capacity is installed here, but there-
fore, an additional 230 GWel photovoltaic capacity compared to the Reference scenario. As
a result, the total installed capacity of RES is around 90 GWel higher than in the Reference
scenario. Since wind turbines have higher full-load hours than photovoltaic systems, around
230 TWhel less electricity is provided from RES in the Non-Acceptance scenario than in
the Reference scenario, despite an overall higher installed capacity. In addition, electricity
generation from photovoltaic plants is characterized by a pronounced daily and seasonal
26
pattern. This poses additional challenges to the Non-Acceptance scenario in order to guar-
antee a secure supply every hour of the year. Accordingly, systemic adjustments must be
made to compensate for a limited expansion of wind energy. What effect this has on the
system development with respect to storage and synthetically produced energy carriers will
be described in detail in the following chapters
3.1.4 Balancing Variable Electricity Generation

Variable renewable energy sources (VRE) play a key role in the success of the energy tran-
sition and will become the mainstay of the energy supply in 2050. However, their further
expansion requires that the variable electricity feed-in can be used sensibly in the energy
system whenever possible. To achieve this, it is necessary to reconcile electricity demand
and generation at all times. In order to meet the electricity demand when the feed-in from
variable renewable energy sources or the discharge of short-term storage facilities such as
batteries or pumped storage power plants is not sufficient, the operation of controllable
power plants is also necessary in the long term. Their installed capacity is shown in Figure 7
for the years 2030 and 2050. To compare, the cumulative installed capacity of conventional
power plants in Germany is currently around 85 GWel.

Gas turbine (CH4)
180
Gas turbine (H2)
160
Combined cycle
140 gas turbine (CCGT) PP
120 Oil PP
100 Nuclear PP
80 Gas PP
60 Combined cycle
gas turbine (CCGT) PP
40 (Heat grid)
Lignite PP
20 Figure 7:
Hard coal PP
Cumulative installed
0
capacity of conventional
ce
ce
ce
ce
y
nc
nc
nc
nc
n
an
an
cie
cie
re
te
re
te
pt
pt
or controllable power plants

fe
fe
sis
sis
ffi
ffi
ce
ce
Re
Re
r
r
Su
Su
Pe
Pe
Ac
Ac
in 2030 and 2050 in the

n-
n-
No
No
four scenarios examined

2019 2030 2050
(PP: power plant)
In the Reference and Persistence scenarios, the installed capacity increases to 95 GWel
and 112 GWel by 2030, respectively. On the other hand, the installed capacity in the Non-
Acceptance scenario remains roughly at today’s level. In this scenario, both the expansion
of wind turbines and the electricity trade with neighboring countries are restricted. This
combined with the earlier and greater expansion of both photovoltaic systems and elec-
tricity storage (see Figure 6 and Figure 8) leads to a reduction in the required capacity of
controllable power plants. In the Sufficiency scenario, the assumed decline in electricity
consumption throughout all sectors causes a slight decrease in the corresponding power
plant capacity.
In addition to a decline in coal-fired power plants from 45 GWel in 2018 to 25 GWel in 2030
[25], a significant increase in highly flexible gas turbines is seen within all scenarios con-
sidered. Due to their comparatively short start-up times, these technologies are becoming
increasingly relevant in a system that is largely characterized by variable renewable energy.
This becomes more evident in 2050. Since coal-fired power plants no longer play a role at
this time, the majority of controllable power plant output will be provided by combined

cycle gas turbine CCGT power plants (with or without heat recovery) and gas turbine power
plants. Overall, the cumulative generation capacity of controllable power plants in 2050
varies between 100 GWel and 160 GWel, depending on the scenario. A reserve capacity of
20 % was always used as a basis for dimensioning. The increasing share of variable renewa-
ble energy in the electricity supply and their prioritized feed-in to the grid have a significant
influence on the operation of the controllable power plant park. Thus, their task is to increas-
ingly cover peak loads instead of base loads, i.e. to provide power instead of energy. This
means that the annual full-load hours of thermal power plants are continuously decreasing,
a trend that can already be deduced from the development over the last few years [5].
Aside from thermal power plants to cover peak loads, electrical storage systems can be
used to balance load and generation. In addition to the pumped storage power plants
already in use today, stationary battery storage systems represent an important option.
The development of the installed capacity of stationary battery storage is shown from
2020 to 2050 for the four scenarios in Figure 8.
GWhel (Battery capacity)

400
350
300
250
200
150
100
50
Figure 8:
0
Development of the installed
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
capacity of stationary battery
storage in the four scenarios
examined
Short-term electricity storage plays an important role in all scenarios. While storage capaci-
ty reaches a value of around 150 GWhel in the Reference scenario and around 50 GWhel in
the Sufficiency scenario by 2050, the installed capacity is significantly higher at 300 GWhel
for the Persistence scenario and 400 GWhel for the Non-Acceptance scenario. This high
value in the Non-Acceptance scenario is due to the fact that the ratio of installed wind
power and solar power is not optimal and thus results in an imbalanced energy supply.
The daytime peak of electricity generation by photovoltaics can be partially compensated
by battery storage, achieving a shift to evening and night hours. In this scenario, a limited
expansion of electricity transmission capacity to neighboring countries was also assumed.
As a result, this reduces the possibility of importing electricity for load balancing. Here too,
more storage capacity can improve the situation.
The comparatively high expansion of stationary battery storage in the Persistence scenar-
io is also due to limited options for load balancing. In this scenario, for example, it was
assumed that the use of vehicle storage units as flexible loads and generators, thus com-
pensating for stationary battery storage, would not be accepted. In addition, the flexible
operation of electricity-based heat generators in residential buildings, such as heat pumps
in combination with thermal energy storage which enable excess electricity to be stored in
the form of thermal energy, is not possible. In the Sufficiency scenario, on the other hand,
battery storage facilities first become necessary towards the end of the considered period,
and the required storage capacity in 2050 is calculated to be significantly lower than for
the other scenarios. The reasons for this are a high level of acceptance for the flexibility
options offered by vehicles and heat pumps in conjunction with a decline in the electricity
consumption of traditional electricity applications which leads to a correspondingly lower
expansion of wind turbines and photovoltaic installations.
28
Another option for balancing load and generation are electrolyzers, which can be used to
produce electricity-based fuels such as hydrogen, methane or liquid fuels. Their integration
into the energy system has several advantages:
■ The energy carriers thus generated can be used in applications where a direct use of
electricity is difficult to implement. Examples are liquid fuels for maritime and air traffic
or hydrocarbons for the chemical industry.
■ Suitable electrolyzers (e.g. PEM: Proton Exchange Membrane; AEL: alkaline electrolyzer)
which allow rapid start-up and shut-down processes can be used as flexible loads, thus
increasing the potential for integrating variable renewable electricity into the energy
system.
■ In times of high electricity feed-in from renewables, these processes can be used to
produce electricity-based energy carriers which can be stored with almost no losses
over a period of several days or months. These energy carriers can also be used in con-
trollable power plants at times of low electricity feed-in from renewable energy sources,
thus securing electricity generation in the long term without the use of fossil fuels.
The extent to which these technologies are installed in each of the considered scenarios is
shown in Figure 9.
140
120
100 Figure 9:
Cumulative installed electro
80
lyzer capacity used for the
generation of electricity-based
60
energy carriers in 2030 and
40 Electrolysis-CH4 2050 for the four scenarios
Electrolysis-Fuel examined (ElectrolysisCH4:
20 for the production of synthetic
Electrolysis-H2 methane; ElectrolysisFuel:
0
for the production of liquid
ce
cy
ce
y
nc
nc
nc
nc
nc
ien
n
cie
re
te
ta
re
te
ta
synthetic energy carriers;

fic
fe
fe
sis
ep
sis
ep
ffi
Re
Re
f
r
r
c
c
Su
Su
Pe
Pe
Ac
Ac
ElectrolysisH2: for the pro-

n-
n-
No
No
duction of hydrogen, as final

2030 2050
energy carrier.)
Although the synthetic production of energy carriers still plays a subordinate role in 2030,
its importance is already apparent in the Persistence scenario with a contribution of 10 GWel.
In 2050 its generation capacity rises to over 120 GWel. The combustion technologies as-
sumed in this scenario for road transport and for the building heat supply are equivalent to
today’s use. These require replacement by renewable synthetic energy carriers.
In the other scenarios, the total installed capacity of electrolyzers in 2050 is between
50 GWel and 75 GWel. Even with strong behavioral changes and the resulting reductions
in consumption as assumed in the Sufficiency scenario, and despite the possibility of
importing renewable synthetic energy carriers from abroad (see Section 3.1.5 below), a
domestic hydrogen economy is shown to be a meaningful part of a fossil-free, renewable-
based energy system.

One argument repeatedly put forward against renewable energy sources is that they do
not provide a base load. Today, however, we are witnessing a fundamental paradigm shift
in energy supply. The demand-based energy supply by large power plants, which was
predominant in the past, is increasingly being replaced by a system in which there is a
continuous balance between energy supply based on renewable sources, which are condi-
tionally controllable and predictable, and the most flexible energy use possible. This results
in a complex interplay between energy supply and time-adjusted energy use (load shifting,
demand response), which includes a stronger coupling of the electricity, heat and transport
sectors, the temporary use of flexible generators and storage facilities of various designs.
As an example, Figure 10 shows the profile of electricity generation and consumption for
one week in April 2050 as calculated in the Reference scenario. The graph shows how the
integration of photovoltaics and wind energy in a future energy system may perform on an
hourly basis.
In the lower graph, the residual load, defined as the difference between non-controllable
power generation, essentially wind energy and photovoltaics, and the non-shiftable base
load (e.g. industrial processes) is shown. Positive residual load means that the current pow-
er is not sufficient due to non-controllable energy generation. Negative residual load means
that there is electricity available which can be used by switchable loads or to charge stor-
age devices. When photovoltaic feed-in peaks during midday (yellow), short-term storage
devices are charged accordingly (red, stationary batteries) and then electricity is converted
into synthetic energy carriers or, if possible, into flexible heat generators, e.g. to charge
heat storage. This operation can be seen, for example, in hours 12 and 36. From hour 50
onwards, the residual load is negative over a longer period of time due to the high electric-
ity input from wind power. The electrolyzer plants operate more continuously than on the
previous days, when operation was only during daytime. The generated hydrogen and any
synthetic energy carriers produced from it, such as methane or liquid energy carriers, can
be stored for longer periods of time and used as required. The addition of hydrogen-elec-
tric technologies (hydrogen-electric vehicles, fuel cells or boilers to provide space or process
heat) reduces the direct demand for electricity. Another option for using peaks in electricity
generation is to raise the temperature level of thermal energy storage (hot water storage)
by converting electrical energy into thermal energy with heating rods (hour 160). If all of
the thermal energy storage systems are full, then the excess electricity is exported. As a last
option, systems are shutdown (hour 156 and following). If the electricity feed-in from re-
newable energy sources is not sufficient, such as in windless nights, the electricity demand
is covered by discharging short-term storage tanks (hours 20, 45, red), importing electricity
(hour 24, purple) or operating thermal power plants (hour 27, CCGT, light green).
30
Electricity supply Other
GWh/h (Power) Gas turbine (H2)
500 Fuel Cell (H2)
Import
400
Combined cycle
300
Pumped storage PP
200 Battery
Combined heat and power
100
Run-of-river
0 PV plants
0 24 48 72 96 120 144 168 Onshore wind
Hours per week
Offshore wind
Electricity use
Other
GWh/h (Power)
Curtailment
500
Grid losses
400 Power-to-Heat
Export
300
Power-to-Fuel
200 Methanization
100 Electrolysis
Pumped storage PP
0
Battery
-100 Heat pumps
Transport
-200
Industry
-300 Traditional Electricity
-400 Residual Load

0 24 48 72 96 120 144 168
Hours per week
In order to guarantee a secure supply even at times when variable renewable energy power Figure 10:
plants provide insufficient electricity over several consecutive days, it is necessary to operate Exemplary profile showing
controllable power plants. Figure 11 shows the electricity generation and consumption for one electricity supply and demand
week in October 2050 as calculated in the Reference scenario. While the peak power reached in April 2050 for the Reference
almost 500 GWel in exemplary April week shown in Figure 10, the maximum generation from scenario. The residual load is
wind and solar in the week shown here is only 180 GWel. the difference between the
hourly electricity generated by
In the first 40 hours of the week shown, the electrical load is balanced mostly by electricity wind and photovoltaics and
generated by wind and photovoltaics. The residual load is close to zero. In addition, com- the non-shiftable base load in
bined heat and power plants and hydrogen-based fuel cells simultaneously generate electrical the power grid.
energy when covering the heating load. Around hours 13 and 24 there is a slight excess of
electricity, so that electric batteries can be charged (red) and electricity can be converted into
heat (light blue).
As a result of a decline in the feed-in from photovoltaic and wind, electricity generation is
no longer sufficient to cover the electrical load from hour 40 onwards, and the residual load
becomes positive. To balance the residual load, stationary batteries are first discharged (red).
Due to the steady decrease in the electricity from wind, the electrical load can be covered
by photovoltaic systems only during the feed-in peaks in the hours from 50 to 60. Once the
photovoltaic feed-in wanes, then other power plants come into play to perform a balancing
function and cover the current load. Pumped storage power plants (dark blue) and gas turbine

Electricity supply Other
GWh/h (Capacity) Gas turbine (H2)
250 Fuel Cell (H2)
Import
200
Combined cycle
150
Pumped storage PP
100 Battery
Combined heat and power
50
Run-of-river
0 PV plants
0 24 48 72 96 120 144 168 Onshore wind
Hours per week
Offshore wind
Electricity use Other

GWh/h (Capacity) Curtailment
250 Grid losses
200 Power-to-Heat
Export
150
Power-to-Fuel
100 Methanization
Electrolysis
50
Pumped storage PP
0 Battery
-50 Heat pumps

Transport
-100
Industry
0 24 48 72 96 120 144 168
Hours per week Traditional Electricity
Residual Load
Figure 11: combined cycle power plants (CCPP) (light green) are used here. During this period, electric-
Exemplary profile for elec ity is also imported from Germany’s neighboring countries (violet). This condition persists up
tricity supply and demand to hour 150, in which the feed-in from wind and solar increases again and from hour 153
in October 2050 for the onwards leads to a negative residual load. Now the stationary batteries (red) and pumped
Reference scenario storage power plants (dark blue) are being recharged.
The time series shown underline the significantly changed operation of controllable power
plants in the future compared to today. While the flexibility requirements for these power
plants increase, the full-load hours – and thus the fuel consumption – decrease.
3.1.5 Fossil and Synthetic Energy Carriers

In addition to the domestic production of synthetic energy carriers by electrolysis, the option
of importing these energy carriers from abroad is taken into account in all of the models. The
price development for these energy carriers was taken from a study carried out for the Federal
Ministry of Transport and Digital Infrastructure (BMVI) [15] 11.
11 The underlying prices were estimated for a production in North Africa and include not only plant costs but
also costs for the capture of carbon from the air (if necessary) and the transport of the respective energy car-
riers to Germany by tankers (including compression and liquefaction, if necessary). Other transaction costs or
costs that could result from the political (domestic) stability of the exporting countries were not considered in
this study. Data on the cost development for the respective energy carriers can be found in the appendix of
the study, which is available on the Internet.
32
Reference Persistence
TWh (Imported energy carriers) TWh (Imported energy carriers)
600 600
500 500
400 400
300 300
200 200
100 100
0 2020 2025 2030 2035 2040 2045 2050 0 2020 2025 2030 2035 2040 2045 2050
Non-Acceptance Sufficiency
TWh (Imported energy carriers) TWh (Imported energy carriers)
600 600
500 500
400 400
300 300
200 200
100 100
0 2020 2025 2030 2035 2040 2045 2050 0 2020 2025 2030 2035 2040 2045 2050
Hydrogen Liquid fuels Hydrogen Liquid fuels
The possibility of importing synthetic energy carriers is used in all scenarios, although Figure 12:
the respective quantities vary greatly. The import of synthetic energy carriers does not The course of the imported
completely replace domestic production, but rather it represents an additional measure. synthetic energy carriers pro-
It therefore makes sense to not only produce synthetic energy carriers domestically from duced abroad by electrolysis
renewable energy sources but also to import them from abroad. The lower full-load oper- and converted to hydrogen
ational hours of the domestic plants are more than compensated by the systemic benefits with renewables from 2020
described in the previous section. The benefit of domestic electrolyzer plants not only lies in to 2050 for the four scenarios
the provision of CO2-neutral energy carriers, but also in the ability to make use of domestic examined
renewable electricity to a greater extent (see Figure 9.)
In the Reference scenario, the amount of imported hydrogen and synthetic liquid fuels
reaches a maximum of 150 TWh per year. In the Sufficiency scenario imports reach only
half at around 75 TWh. In the scenarios Non-Acceptance and Persistence, on the other
hand, the imported quantities are significantly higher at more than 300 TWh and 500 TWh
respectively. Despite the varying import volumes in the different scenarios, the import of
synthetic fuels should be considered in relation to the decline in fossil fuel use. As an
example for the Reference scenario, Figure 13 shows the quantity of fossil fuels used for
the period from 2020 to 2050.
In 2018, around 2700 TWh of fossil fuels were used in Germany [28]. The majority of these
are natural gas, mineral oil, hard coal and lignite. Due to the lack of domestic resources and
the phasing out of hard coal mining, these energy sources are almost exclusively imported
from abroad [29]. In order to achieve the targets for the reduction of energy-related CO2
emissions, fossil fuel imports decrease continuously and reach a value of almost 200 TWh
in 2050 in the Reference scenario, which corresponds to about 7 % of the quantity imported
today. This shows that restructuring the energy system, in addition to meeting the climate
targets, will also significantly reduce Germany’s dependence on energy from other coun-
tries.

TWh (Fossil energy carriers)
3000
2500
2000
1500 Lignite
1000 Hard coal

Oil
500
Natural gas
2015 2020 2025 2030 2035 2040 2045 2050
Figure 13: Figure 14 shows the total quantity and origin of material energy carriers for the years 2030
Fossil fuel use from 2015 to and 2050 in the four scenarios.
2050 for the Reference scena-
rio. The majority is imported. In 2030 the share of fossil fuels will predominate in all four scenarios. The share of synthetic
energy carriers produced using renewable electricity – whether abroad or domestically – is
still very small. In 2050, on the other hand, the differences are much greater for the four
scenarios, both in terms of total quantity and origin. In all scenarios, the share of fossil fuels
is very similar with a value of around 200 TWh, which is largely determined by the specified
Figure 14: CO2 emissions set as a fixed boundary condition in the simulations. In contrast, the quan-
Composition of material ener- tities of synthetically produced energy sources (domestic and imports) range from around
gy carriers and their origin for 230 TWh in the Sufficiency scenario to almost 890 TWh in the Persistence scenario. The
the four scenarios in the years high value in the latter is due to the high stock of internal combustion technologies for the
2030 and 2050. supply of heat and for transport, which are still being used in 2050 in this scenario.
TWh (Energy)
2000
1800
1600
1400
1200
1000
800
600
400
200
0 Fossil Synth. Synth. Total Fossil Synth. Synth. Total

imported domestic imported domestic
2030 2050
34
3.2
Sector-specific Analysis
In this chapter, the consumption sectors of process heat, transport and building heat are
analyzed in more detail for the four scenarios examined.
3.2.1 Industrial Process Heat

In 2018, the supply of process heat accounted for 565 TWhth, which corresponds to around
30 % of Germany’s final energy demand. Accounting for around 90 % of consumption,
industry plays a central role here [5]. Depending on the industry and the processes used,
the required temperature level varies from 100 °C and below to processes requiring several
thousand degrees Celsius [19]. Due to this wide variation, different technologies are possi-
ble for the supply of process heat, depending on the given temperature range. For the cal-
culations and the evaluation of results, the process heat requirement was divided into three
segments. The low-temperature segment extends up to 100 °C and includes processes
such as bleaching, dying and washing in the textile industry as well as pasteurization,
blanching, scalding, cooking and smoking in the food industry. The medium temperature
segment covers the range from 100 °C to 500 °C. Typical processes in this temperature
range are distillation in the chemical industry, pressing in the woodworking industry, baking
in the food industry and drying processes across all industries. The high-temperature
segment includes all industrial processes with temperature levels above 500 °C. Particularly
energy-intensive in this segment - in ascending order - are the chemical industry, the pro-
cessing of glass, ceramics, stone and earth, and metal production and processing.
Figure 15 shows the energy sources used for the process heat supply for the low tempera-
ture segment for the years 2030 and 2050 in each of the four scenarios.
TWhth (Process heat supply) Solar thermal energy

100
Hydrogen
Heat pumps
80
Electricity
60
Biomass
Combined cycle
40
Gas
20
Oil Figure 15:
Process heat supply according
0 Coal
to technology options in the
ce
ce
y
nc
nc
nc
nc
nc
nc
n
n
cie
cie
re
te
ta
re
te
ta
low temperature segment

fe
fe
sis
ep
sis
ep
ffi
ffi
Re
Re
r
r
cc
cc
Su
Su
Pe
Pe
(up to 100 °C) in 2030 and

-A
-A
n
n
No
No
2050 for the four scenarios

2019 2030 2050
examined

The low-temperature segment is responsible for around 18 % of the total process heat
demand in Germany [3], and in 2018 the final energy demand was 102 TWhth. This cor-
responds to a useful heat demand of about 81 TWhth. It is assumed that this will be 75
TWhth in 2030 and will fall to around 70 TWhth by 2050. This reduction is based on the
assumption that annual efficiency gains averaging 0.5 % can be achieved by using the best
available technology (BAT). In the Sufficiency scenario, a slight decline in consumption and
thus production is also assumed, which reduces the demand for process heat by an addi-
tional 0.25 % per year.
Despite the different assumptions in each scenario, a similar trend can be observed in
the generation technologies used for the supply of process heat. While a wide variety of
technologies will still be used for the supply of heat in 2030, technologies with lower CO2
emissions will begin to replace combustion-based technologies from 2030 to 2050. Thus
in 2050, solar thermal, heat pumps, electrode and hydrogen boilers will be used almost
exclusively, with heat pumps accounting for the largest share (see Figure 15).
From 2030 onwards, the share of heat pumps will increase sharply, reaching almost 60 %
in the low-temperature segment in 2050. One reason for this development is their aver-
age coefficient of performance of 2.5, which results in lower final energy consumption
(as compared to a conventional boiler). This factor is becoming increasingly important as
emissions reduction targets become more ambitious. Besides heat pumps, other electric-
ity or hydrogen-based technologies provide low-temperature process heat. Both options
benefit indirectly from an increase in renewable electricity in the system and the associated
reduction of the CO2 factor of electricity (see Table 3). Renewable heat is also provided by
solar thermal systems. Their contribution increases continuously over time with the scenario
average reaching around 11 % in 2050.
Consumption in the medium temperature segment (100 - 500 °C) currently accounts for
18 % of total process heat demand. The technologies for heat supply in 2030 and 2050 for
the four scenarios are shown in Figure 16.
In this temperature segment, too, a displacement of conventional, fuel-based technologies

is evident in the period up to 2050. Instead of oil and gas boilers, variants with lower CO2
emissions are increasingly being used, such as the combustion of biomass. This result repre-
sents a significant difference to heat supply in the sector of space heating and domestic hot
water, where biomass will hardly be used in 2050. At the same time, the use of electricity
to provide process heat via heating elements is increasing. This change concerns, for exam-
ple, the provision of steam and hot water. Today fuel-based heat generators are generally
TWhth (Process heat supply) Solar thermal energy

100
Hydrogen
Heat pumps
80
Electricity
60
Biomass
Combined cycle
40
Gas
20
Oil
Figure 16:
Process heat supply by techno- 0 Coal
ce
cy
ce
ce
y
nc
nc
nc
nc
ien
n
an
cie
re
te
ta
re
te
logy in the medium tempera-

pt
fic
fe
fe
sis
ep
sis
ffi
ce
Re
Re
f
r
r
c
Su
Su
Pe
Pe
Ac
Ac
ture segment (100 - 500 °C)

n-
n-
No
No
in 2030 and 2050 for the four

2019 2030 2050
scenarios examined
36
used for this purpose, yet there is great potential for electrification. Figure 16 shows the
increase in indirect and direct electricity use, particularly through the use of hydrogen,
electrode boilers and heat pumps. In the case of heat pumps for use in industry, it has been
assumed that by 2030 these will be able to provide process heat up to 180 °C, and by 2050
process heat up to 300 °C [30, 31]. These assumptions are based on the developments in
suitable refrigeration circuits.
The third temperature segment describes processes above 500 °C and at 64 % accounts
for the highest proportion of the total process heat requirement [3]. The technology for the
heat supply in this segment is shown in Figure 17.
TWhth (Process heat supply)

350
300
250
200
Hydrogen
150 Electricity
100 Gas
Oil
50
Coal
0
ce
ce
ce
ce
y
nc
nc
nc
nc
n
an
an
cie
cie
re
te
re
te
Figure 17:
pt
pt
fe
fe
sis
sis
ffi
ffi
ce
ce
Re
Re
r
r
Su
Su
Pe
Pe
Ac
Ac
Process heat supply by techno-

n-
n-
No
No
logy in the high-temperature

2019 2030 2050
segment (above 500 °C)
Due to the higher temperature requirements, only fuel-powered boilers can be used in the
high-temperature segment. Heat pumps, solar thermal systems, biomass boilers and waste
heat from combined heat and power generation therefore play no role in this segment.
Accordingly, there will still be a high proportion of gas-fired boilers in 2030 in particular,
although these will be largely replaced by hydrogen boilers and electric furnaces in 2050.
At the same time, coal-fired boilers will decline over time. These will be mainly used for
the production of crude steel. The steel production process is represented in the model by
the primary and secondary route. In the primary, or blast furnace route, iron ore is reduced
using coking coal to produce high-quality steel. The coal requirement for this process is
assumed to be around 90 TWh (around 30 million t CO2) in 2030 and about half of this in
2050 (around 15 million t CO2). In the secondary route, steel is produced from steel scrap
using electric arc furnaces. It is assumed that by 2050 steel produced in this manner will
replace one third of the primary steel. The reason for this limitation is, on the one hand, the
availability of steel scrap and, on the other hand, the fact that certain steel grades cannot
be produced in this process. A further third of the carbon-intensive primary steel produc-
tion process will be substituted by the increasing use of hydrogen [32].
Across all three temperature segments, a shift from conventional technologies to electrici-
ty-based technologies in the process heat sector is emerging in the form of direct use (heat
pumps and electric boilers) or indirect use (hydrogen and methane-based processes). Elec-
trically-based process heat technologies offer further advantages in addition to making use
of renewable electricity. These include the fact that it is comparatively easy to guarantee
high temperature levels and to achieve precise local and temporal heating, thus reducing
losses. At the local level, these technologies also offer the advantage of no noise pollution
and emissions.

3.2.2 Transport
With a final energy demand of 768 TWh in 2017 [33], the transport sector in Germany
emitted about 168 million t CO2 [34]. The model calculations assume that the energy de-
mand of inland waterway, air and fuel-based railway transport will continue to be covered
by liquid fuels due to their higher energy density. The reduction of CO2 emissions in these
application areas is therefore achieved by using an increasing proportion of CO2-neutral
liquid fuels such as biodiesel or synthetic energy carriers. In road transport, a “fuel switch”
is possible by selecting different drivetrain technologies in the model. Motorized individual
transport and road-bound freight transport are each represented by seven different drive-
train concepts. This makes it possible in the course of the optimization process to analyze
both the continuation of using conventional as well as the change towards alternative
drivetrain technologies.
Motorized Private Transport

In order to transform motorized private transport, the German government set the goal of
increasing the number of electric vehicles on Germany’s roads to up to 10 million vehicles
by 2030 [27]. According to data from the Federal Motor Transport Authority, around one
percent of all new vehicle registrations were electric vehicles in 2018 [35]. To achieve the
target of 7 to 10 million electric vehicles in 2030, this share would have to increase to an
average of 23 % to 33 % of all new registrations over the next ten years 12. The results of
the model show that the German federal government’s targets are exceeded in all sce-
narios except the Sufficiency scenario. The number of vehicles per powertrain technology
resulting from the calculations is shown in Figure 18 for the years 2030 and 2050 for the
four scenarios examined.
Mio. (Cars)
60
50
40
30 Battery-electric
20 Hybrid
Hydrogen-electric
10
Internal combustion
0
ce
ce
ce
ce
Figure 18:
nc
nc
nc
nc
en
an
en
an
cie
cie
te
te
er
er
pt
pt
Powertrain technologies used

sis
sis
ffi
ffi
f
f
ce
ce
Re
Re
r
r
Su
Su
Pe
Pe
Ac
Ac
in the passenger car sector in

n-
n-
No
No
2030 and 2050 for the four

2019 2030 2050
scenarios examined
12 Extrapolation assuming an average of 3.4 million new registrations per year
38
Today almost the entire fleet of around 47 million motorized private vehicles is based on
internal combustion engines that run on predominantly on fossil fuels. In the Reference
scenario the number of combustion-based vehicles in 2030 will be significantly smaller
at 33 million vehicles. This decline is mainly compensated by battery electric vehicles, of
which about 15 million will be in operation in 2030. In addition, there are another approx.
3 million hybrid vehicles and 540,000 hydrogen-electric vehicles. These figures correspond
to an annual average of new registrations for the period from 2020 to 2030 of about 1.5
million battery electric vehicles, 300.000 hybrid vehicles and 50.000 hydrogen electric
vehicles. With currently about 3.4 million new registrations per year [36], new registrations
of battery electric vehicles would thus average around 44 % of new vehicles over the next
10 years. The model results indicate that the German federal government’s target of 7 - 10
million battery electric vehicles in 2030 will probably not be sufficient to achieve the climate
protection targets in a cost-optimized manner. The number of battery electric vehicles will
rise to 40 million by 2050. According to the assumptions in the Reference, Persistence and
Non-Acceptance scenarios, the share of battery electric vehicles can be maximum 80 %
of the vehicle fleet (see footnote in Section 2.3), and therefore the remaining share in the
Reference scenario in 2050 is based on hydrogen-electric vehicles. This means that a 95 %
reduction in CO2 emissions compared to 1990 would cause a complete displacement of
internal combustion engines in motorized private transport.
The trend shown in the Reference scenario in 2030 is essentially confirmed in the Non-
Acceptance scenario. In the Persistence scenario, the share of battery electric vehicles is
slightly lower in 2030 and noticeably lower in 2050 due to the assumption that at least
50 % of new registrations are combustion-based vehicles. At the same time, many vehicles
with internal combustion engines will still be in operation in 2050, yet they will be pre-
dominantly powered by CO2-neutral fuels (see Table 4). As explained in Chapter 3.1.4, the
installed capacity of Power-to-Fuel plants and the amount of synthetic fuel imports required
to provide these fuels are by far the largest in the Persistence scenario. In the Sufficiency
scenario, on the other hand, the assumed decline in consumption (reduction in the number
of vehicles and kilometers driven) means that in 2030 it will not yet be necessary to convert
the motorized individual transport to the same extent as in the other scenarios in order to
achieve the CO2 reduction targets. Accordingly, the switch to battery-electric and hydro-
gen-electric powertrains will take place at a later date in the Sufficiency scenario.
In the Reference, Sufficiency and Non-Acceptance scenarios, electric-based powertrain

concepts play a major role in the transformation of the energy system. Battery-electric and
hydrogen-electric powertrains not only have higher efficiencies compared to combustion
engines, but they also offer the potential for flexibility in the energy system. In battery
electric vehicles, this is achieved by charging (grid-to-vehicle) or discharging (vehicle-to-grid)
the vehicle battery as required by the energy system. Hydrogen-electric vehicles open up
the possibility of shifting energy quantities over longer periods of time, even seasonally.
For example, excess renewable electricity can be used for the production of hydrogen in
summer and consumed in the winter months.
Road-bound Freight Transport

Analogous to motorized private transport, the distribution of powertrain technologies in
road-bound freight transport is optimized for all four scenarios. The Sufficiency scenario
assumes a 21 % decline in mileage, or tonne-kilometers driven, in 2050, based on today’s
figures (see Chapter 2.3).

% (Trucks)
100
90
80
70
60
50
Figure 19:
40
Proportion of drive concepts
(normalized to mileage) in 30
Battery-electric*
the truck sector for the four
scenarios examined in 2030 20 Hybrid
and 2050.
Hydrogen-electric
10
* purely battery-electric power Internal combustion
0
trains including overhead-line
ce
ce
ce
ce
y
nc
nc
nc
nc
n
an
an
cie
cie
re
te
re
te
pt
pt
trucks, with the exception of
fe
fe
sis
sis
ffi
ffi
ce
ce
Re
Re
r
r
Su
Su
Pe
Pe
Ac
Ac
the Non-Acceptance scenario in
n-
n-
No
which no overhead-line trucks are

2019 2030 No 2050
considered
Compared to motorized private transport, the conversion of conventional to alternative

powertrain systems takes place to a lesser extent in vehicles for freight transport. In 2030
combustion engines will continue to be the dominant drive technology for road-bound
freight transport in all scenarios, and also in 2030 the proportion of mileage driven by
hydrogen-electric vehicles will still be very low. In 2050 an increase in alternative drive tech-
nologies can be seen in all scenarios, with conventional internal combustion engines still
playing a major role. According to the model results, it makes sense to not convert some of
the combustion-based vehicles and instead to run them on CO2-neutral fuels (see Table 4)
in order to reduce emissions. In the Reference scenario, for example, around 56 % of the
mileage driven in 2050 will be covered by conventional drive technologies. A further 24 %
of the mileage driven will be covered by hydrogen-electric vehicles, followed by trucks run-
ning on overhead contact lines (just under 20 %). In contrast to battery electric vehicles in
motorized private transport, it is assumed that trucks running on overhead lines cannot be
operated in a grid-supportive manner. Instead, their charging behavior is based on stored
driving profiles.
Due to the assumed lack of acceptance to changes in large infrastructures in the Non-
Acceptance scenario, the use of overhead contact-line trucks was excluded here. Pure
battery use was considered for smaller delivery trucks (e.g. for parcel delivery) and city
buses, but not for commercial trucks covering long distances. Although promising devel-
opments are also being tested in this area, the data basis is currently still insufficient, so
that these concepts were not considered. This leads to an intensive use of hydrogen in
heavy-duty traffic in the Non-Acceptance scenario.
The mix of liquid fuels used for all road transport is shown in Table 4. In 2030 fossil fuels
(or fossil fuel contents) dominate in all scenarios. Here it is important to note that no man-
datory blending of biofuels was assumed in the model simulations. Greater differences can
be observed when comparing the scenario results in the year 2050. While fossil fuels still
account for around 50 % in the Reference, Non-Acceptance and Sufficiency scenarios, the
corresponding value is much lower for the Persistence scenario.
40
Overall, it can be seen that in the scenarios with a target of 95 % reduction in energy-
related CO2 emissions, a considerable share of the available fossil fuels is used for the
heavy-duty transport.
Table 4: Overview of liquid fuels used for transport in 2030 and 2050 in the four
scenarios studied (individual and freight transport, air transport, inland waterways
and fuel-based rail transport).
Liquid Fuel Mix in % Fossil Power-to-Fuel Biomass

2030 Reference 100% 0% 0%
Persistence 99% 1% 0%
Non-Acceptance 95% 0% 5%
Sufficiency 94% 0% 6%
2050 Reference 56% 19% 25%
3.2.3 Building Heat

The provision of space heating and domestic hot water accounted for around 30 % of total
final energy consumption in Germany in 2018. A key measure to reduce emissions in this
area is to lower the heating requirement. This can be achieved by renovating the building
envelope of existing buildings to make them more energy efficient and by reducing the
heated living space. This would require changes in behavior, reflected, for example, in a
decrease in the specific living space per inhabitant. The second important measure in the
building sector is the use of heating technologies with lower specific CO2 emissions. The
modeling results based upon the implementation of these measures for the four scenarios
studied are presented below.
Building Energy Renovation

Energy refurbishments in the building stock are represented in the model by three different
degrees of energy renovation. “No refurbishment” is based on the current state of build-
ings from 2011, which have undergone no refurbishment. The second category called “Full
refurbishment” is based on the specifications of the Energy Saving Ordinance (EnEV) and
the third category “Full refurbishment plus” is based on the passive house standard [37,
38]. The number of buildings undergoing energy refurbishment each year is determined by
the optimization algorithm in all of the scenarios except for Persistence scenario. Among
other restrictions, the Persistence scenario assumes that a maximum of one percent of the
building stock is refurbished per year. In the other three scenarios, however, the rate of
refurbishment can be freely determined. The number of non-refurbished or energetically
refurbished buildings directly influences the useful heat demand to be met. Table 5 provides
an overview of the corresponding heating requirements and the proportion of refurbished
buildings in 2030 and 2050 for the four scenarios examined. It also shows the average rate
of refurbishment over the period analyzed.

Table 5: Net energy demand for space heating and domestic hot water for the four scenarios
considered. In addition to the degree of refurbishment, the energy demand depends on the
weather data set which is taken into account (weather year).
Useful heat demand Share of Rate of

in TWhth refurbished buildings refurbishment
in % in %
2030 2050 2030 2050 2020-2050
Reference 725 604 51 79 1.4
Persistence 720 631 52 74 1.0
Non-Acceptance 726 593 51 83 1.5
Sufficiency 759 576 43 88 1.7
The results show that building energy refurbishment plays a role in the energy system
transformation in all scenarios. According to the results, the percentage of refurbished
buildings in the building stock will be between 43 % and 52 % in 2030 (today’s share:
40 %). Despite this increase, the useful heat demand can only be reduced slightly by 2030.
This can be attributed in part to the increasing number of buildings over the years. In
addition, the simulations were carried out using different data sets (weather years), and
the data set used for the year 2030 has a comparatively low mean outdoor temperature
and solar radiation, which corresponds to a comparatively high space heating requirement.
Due to the assumed decrease in consumption, the results from the Sufficiency scenario
show that the CO2 targets can also be achieved for the building sector with much less
effort compared to other scenarios. This is reflected in the lower percentage of refurbished
buildings calculated in the model for the year 2030.
A different picture emerges for the year 2050, when the Sufficiency scenario has the high-
est proportion of refurbished buildings. This follows the scenario’s assumption that sup-
portive efforts are made in politics and the population to lower emissions by reducing ener-
gy consumption. For this purpose, it was assumed that the minimum rate of refurbishment
increases to two percent per year between 2030 and 2040. Because of this, the Sufficiency
scenario shows the greatest reduction in useful heat demand among the four scenarios. In
the Reference scenario, on the other hand, the optimization specifies an annual refurbish-
ment rate that is up to twice the current level, so that the average over the period under
consideration is 1.4 %. The Non-Acceptance scenario shows a similar development. Here a
slightly higher rate of refurbishment is achieved, due to the restrictions set in other sectors.
In the Persistence scenario, the rate of refurbishment remains fixed at a value of 1.0 % even
after 2030. As a result, the useful energy demand is almost 30 TWhth higher in 2050 com-
pared to the Reference scenario.
Across the four scenarios, a consistent picture emerges, which emphasizes the role of
building energy refurbishment. The results show that a large part of the building stock
should be refurbished in order to achieve the emission reduction target of 95 % compared
to 1990 at optimal costs. The annual rate of refurbishment varies between about 1.0 % and
2.1 %, averaging about 1.6 % in the Reference, Sufficiency and Non-Acceptance scenarios
respectively.
42
Technologies for Heat Supply
The second important measure to reduce CO2 emissions in buildings is to replace gas or
oil-fired boilers with technologies having lower specific emissions (in relation to the amount
of heat provided). Heat pumps are especially suitable here, since a large percentage of the
required energy is taken up from the surrounding environment (e.g. outside air, soil, build-
ing exhaust air). Other possibilities are the use of biomass in pellet boilers, solar thermal
systems or even the use of fuel cells or combined heat and power plants, which make more
efficient use of the energy carriers than a combustion boiler.
In the model, the building’s heat demand is covered by a heat generator or a heat grid.
A variety of technologies are used to supply heat in heat grids. These include large heat
pumps, combined cycle gas turbine plants (CCGT) with heat recovery (combined heat and
power generation, driven either by heat or electricity depending on the current require-
ments), condensing boilers, solar thermal systems and electrode boilers. Another possibility
is to install large storage tanks (hot water tanks), which open up the flexibility options in
the supply. The technologies for the supply of heat in buildings are shown for 2030 and
2050 for the four scenarios in Figure 20.
% (Heating systems )
100
90
Gas boiler
80
Wood
70
Fuel cell CH4
60 Fuel cell H2
50 CHP
Hybrid HP
40
Elec. HP (air)
30
Elec. HP (ground) Figure 20:
20 Gas HP Share of heating systems that
supply space heating and do-
Oil
10 mestic hot water in buildings
Heat grids in 2030 and 2050 for the four
0 scenarios examined. Solar
e
ce
ce
ce
y
nc
nc
nc
nc
nc
en
an
an
cie
cie
re
re
thermal energy and heating

t
pt
ist
pt
fe
fe
sis
ffi
ffi
ce
ce
s
Re
Re
r
r
Su
Su
Pe
Pe
rods can be used additionally,

Ac
Ac
n-
n-
No
No
but are not shown here

2019 2030 2050 (HP: heat pump).
Today, space heating and domestic hot water are primarily provided by oil and gas boilers,
which account for around three quarters of all heat generators in Germany. Their share
decreases with time across all scenarios, yet still will be about 61 % in 2030 in the Refer-
ence scenario. This decline is largely compensated by electric heat pumps and connections
to heat grids, which will cover 12 % and 20 % of the heat demand respectively in 2030. In
the Persistence scenario, the share of heat pumps and connections to heat grids is lower, at
10 % and 15 %, respectively. This is due to the boundary conditions for this scenario which
are explained in Chapter 2.3.

In 2050, the heat supply in the Reference, Sufficiency and Non-Acceptance scenarios will
be provided in about 38 % of all buildings by heat grids. Electric heat pumps will be in-
stalled in 42 % to 54 % of the buildings. In total, these technologies thus cover a large part
of the demand for space heating and domestic hot water. The dominance of these technol-
ogies in the three scenarios is mainly due to the high conversion efficiency of heat pumps,
a key technology also in heat grids. The average annual performance factor of heat pumps
of around 3.4 indicates that a large amount of environmental heat is used, which has an
impact on the resulting emissions. In addition, heat pumps can be operated in combination
with thermal energy storage systems to support the grid and thus contribute to load bal-
ancing in the electricity system. Thus, on days with plenty of sun or wind, electricity from
renewable energy sources can be used to raise the temperature level in the thermal energy
storage by means of heat pumps. The heat stored in this way can be used during periods of
low availability of renewable energy and thus reduce the electricity load. In heat grids, com-
bined cycle gas turbine (CCGT) plants are operated during low renewable availability, thus
contributing to load balancing. If required, decoupled heat can be used directly or rather
to charge large heat storage units. All in all, multi-modal heat grids, i.e. grids fed from
different generators in combination with large heat storage facilities, have proven to be an
effective measure for providing flexibility and load management in dense urban areas.
The use of biomass for space heating and domestic hot water will play almost no role in
2050. From a systemic point of view, it appears to be more cost-effective to use biomass
for the provision of process heat and for the production of liquid fuels or biogas (see Sec-
tion 3.2.1 and Table 4 and Table 6) and to rely instead on electricity-based technologies and
district heating for space heating.
The Non-Acceptance scenario shows that a restriction of wind energy expansion and
the associated higher expansion of photovoltaic systems leads to a larger share of hydro-
gen-based technologies in the consumption sectors. For example, in 2050 almost 15 % of
households will be supplied by a fuel cell heating system to cover their heat requirements
(CH4 or H2-based), which equates to 105 TWhel of electricity generation in 2050.
Based on the boundary conditions in the Persistence scenario, the share of conventional
heat generators will be at least 50 % in 2050, i.e. half of all buildings will be heated with a
gas boiler. Thus, the proportion connected to the heat grid in 2030 in the other scenarios,
is first reached in 2050 in this scenario. The share of air-source heat pumps is also only
18.5 %. Due to the boundary conditions set, it is much more difficult in this scenario to
achieve the emission targets by means of reducing the useful heat demand and by chang-
ing the composition of the heating technologies. Accordingly, CO2-neutral synthetic energy
carriers generated from renewables must be used to a greater extent (similar to the trans-
port sector). See Table 6.
The part of the heat demand not covered by heat pumps and heat grids is supplied by a
mix of different heat generators, with gas boilers and fuel cells making up the largest share.
In 2050 only a small percentage of the gas used will still come from fossil sources. Also,
the emissions from the gases used for this purpose will decrease due to the reduced use of
fossil natural gas and the simultaneous increase in biogas, synthetically produced methane
and hydrogen. The composition of gaseous energy carriers in the grid is shown in Table 6
for the years 2030 and 2050.
44
Table 6: Mix of gaseous energy carriers in the grid in the four scenarios for the years
2030 and 2050.
Gas Mix in % Natural Gas Biomass Power-to-Gas

2030 Reference 90% 6% 4%
2050 Reference 16% 67% 17%
Across all scenarios, it can be stated that in combination with energy refurbishments, an
extensive changeover from the use of conventional heat generators to heat pumps and
heat grids is an essential cornerstone of a cost-optimized transformation of the heating
sector.
Energy Mix for Building Heat Supply

For the provision of space heating, the model distinguishes between two systems, which
are characterized by different temperature levels: Radiator and panel heating systems (e.g.
radiant floor heating). The latter requires a lower supply temperature because the heat is
transferred to the room over a larger surface area. The conversion from one type of heating
system to another is handled in the optimization. Depending on the type of system, such a
refurbishment can have a greater or lesser effect on the conversion efficiency of the entire
heating system.
With a share of around three-quarters, radiator heating systems account for the majority
of the heating systems in German buildings today. By 2050, their share will decline contin-
uously – and the more ambitious the boundary conditions are in the respective scenarios,
the faster this will happen. As a result, the heat supply will become more efficient, especial-
ly in heat pump systems. Figure 21 shows the quantities of heat provided per technology
for the years 2030 and 2050. In addition, technologies that supplement the heat supply,
such as solar thermal systems or heating rods, are also listed.
TWhth
800 CHP
Fuel cell
700
Gas boiler
600
Heating rod
500
Wood boiler
400
Oil boiler
300
Solar thermal energy
200
Heat grids
100
Heat pumps Figure 21:
0 Storage losses Heat supply in the building
sector according to hea-
ting technology for the four
ce
ce
ce
ce
ce
ce
y
nc
nc
n
en
an
en
an
cie
cie
re
re
scenarios for 2030 and 2050.

t
pt
pt
fe
fe
sis
sis
ffi
ffi
ce
ce
Re
Re
r
r
Su
Su
Pe
Pe
Ac
Ac
Storage losses are shown ne-

n-
n-
No
No
gatively (CHP: combined heat

2030 2050
and power unit).

In the Reference scenario in 2030, a large part of the heat will still be transmitted by radi-
ator heating. Gas boilers will provide 50 % and oil boilers 11 % respectively. At the same
time, the simulation results show an increase in panel heating systems and in heat gen-
erators with lower CO2 emissions such as wood boilers, heat grids, heat pumps and solar
thermal systems. Heat pumps in particular benefit from the lower supply temperature of
the panel heating systems and thus the higher conversion efficiency.
The further conversion from radiator to panel heating becomes apparent after 2030, with
the share of panel heating rising to almost two thirds by 2050 in the Reference scenario. As
shown in Figure 21, a large part of the heat demand is covered by heat pumps and heating
grids. In addition, other technologies that can be used to balance the residual load are also
gaining in importance. These include, for example, heating rods (used on days with high
feed-in from variable renewable energy) or fuel cells (used more on days with low feed-in
from variable renewable energy). This is evidenced, for example, in the Non-Acceptance
scenario, in which the electricity generation by fuel cells partially compensates for the
reduced expansion of wind energy.
In the Persistence scenario it was specified that, in addition to the continued use of con
ventional technologies, no technologies that promote flexibility such as heating rods are
to be used. Therefore, these do not represent an option for balancing the residual load.
This restriction is partly compensated by using solar thermal systems to a greater extent.
For example, the heat generated from solar thermal systems in 2050 will be just under
80 TWhth compared with around 45 TWhth in the other scenarios. This corresponds to an
installed capacity of around 110 GWth for solar thermal systems in the Persistence scenario
in 2050, of which a good two-thirds of the systems are in buildings and almost one-third
feed into heat grids; this ratio is similar in all scenarios. In addition, all scenarios include
solar thermal energy systems for processes in trade and industry with an installed capacity
of around 15 GWth.
3.2.4 CO2 Reduction by Sector

In order to make comparisons with sector-specific observations from other studies, the
emissions from the areas of application electricity, transport, building heating and process
heating (as broken down in this study) were converted into the transport, building, industry
and energy sectors [27].
The energy sector describes the emissions resulting from the combustion of fossil fuels in
power plants serving the public electricity and heat supply and from pipeline compressors
and refineries as well as other fugitive emissions from the energy economy. This means that
the figure also includes emissions caused by electricity consumption in private households,
transport, industry and in the tertiary (TCS) sector. Accordingly, all emissions for electrici-
ty generation calculated with the model are allocated to this sector. The transport sector
describes emissions caused by domestic road, rail and air transport as well as inland and
coastal shipping. The buildings sector includes emissions caused by the use of fuel for space
heating and hot water in households, in the tertiary sector, industrial and public buildings.
Finally, the industrial sector includes all emissions caused by the provision of process heat
(including steel production). Process-related emissions, on the other hand, are not shown,
in line with the balance sheet limit used as a basis for the model. Figure 22 shows the re-
sulting CO2 emissions per sector from 2020 to 2050 for the Reference scenario.
46
Mio. t (Energy-related CO2 emissions)
1000
800
Transport
600 Figure 22:
Buildings Energy-related CO2 emissions
400 for the Reference scenario in
Industry million t CO2 after breakdown
200
into the sectors of transport,
Energy sector
0 1990 2020 2030 2040 2050 buildings, industry and energy
industry
According to calculations with this model, the energy sector shows the largest reductions
up to 2030, while emissions from the transport sector will decline comparatively less by
2030.
The development of the specific CO2 factors of the various energy carriers is also shown for
the Reference scenario in Figure 23.
gCO2/kWh (Specific emission values)

400
350
300
250
200
150
100
50
Figure 23:
0 Development of the CO2
20
25
30
35
40
45
50
factors for electricity, hydro-

20
20
20
20
20
20
20
gen, grid-bound gaseous

Gridbound gaseous energy carriers Liquid Fuels Electricity Hard coal Lignite energy carriers, liquid fuels,
lignite and hard coal
The development of the CO2 factors shown in the results of the model underlines the en-
ergy sector’s strong contribution to emission reductions, since the CO2 factor for electricity
in particular already shows significant reductions by 2030. In 2050, electricity generation
is about 3 gCO2 /kWhel and thus almost completely free of fossil fuels. The CO2 factors for
gaseous energy carriers in the grid and for liquid fuels also change over time, but with a
significant delay compared to electricity. From 2020 to 2050, the CO2 factor of liquid fuels
decreases from 242 gCO2 /kWh to 140 gCO2 /kWh and of gaseous energy carriers in the
grid from 199 gCO2 /kWh to 29 gCO2 /kWh in the scenario described here.
This evaluation underlines the major role that sector coupling plays in the energy transition.
Successive reductions in the use of fossil fuels in the application areas of heat (buildings,
processes) and transport, combined with a greater use of electricity, and an accelerated re-
duction of specific emissions in electricity generation, appear to be the most cost-effective
way to achieve the desired reduction targets for CO2 emissions in energy supply.

3.3
Complete Reduction of Energy-related CO2 Emissions
In addition to the previous scenarios, this section presents the further steps necessary
to move from a far-reaching (minus 95 % relative to 1990) to a complete (minus 100 %)
reduction of energy-related CO2 emissions. This means that the energy consumption of all
sectors is based 100 % on renewable energy sources used either domestically or abroad 13.
In order to analyze the technical feasibility and the effects of a complete reduction of ener-
gy-related CO2 emissions, two model calculations are carried out, which are derived from
the previously considered scenarios. The first model investigates a complete reduction of
energy-related CO2 emissions by 2050; otherwise, the boundary conditions of the Refer-
ence scenario are essentially used as a basis (hereinafter referred to as “Reference100”). The
second model calculation is based on the boundary conditions of the Sufficiency scenario.
It considers a strongly accelerated, complete reduction of energy-related CO2 emissions by
2035 (hereafter referred to as “Sufficiency2035”).
3.3.1 Complete Reduction of Energy-related CO2 Emissions by 2050

A comparison of the scenarios Reference and Reference100 shows that the primary energy
demand in 2050 hardly differs between the two cases. Accordingly, the use of natural gas,
oil and hard coal in 2050, which in the Reference scenario totals 190 TWh, must be re-
placed. Imports of electricity and biomass remain at a similar level in both scenarios due to
the set boundary conditions (biomass availability and grid expansion to neighboring coun-
tries). Around 80 % of the energy that is no longer available in the form of fossil fuels in
Reference100 is compensated by an increased electricity generation from solar and onshore
and offshore wind. Figure 24 shows a comparison of the installed power plant capacities
for the generation of electricity from variable renewable energy sources (wind, sun) in the
years 2030 and 2050.
GWel (Installed power)

800
700
600
500 PV 2019
PV roof, east-west
400
PV roof, south
300
PV ground, south
200
Onshore wind
Figure 24: Installed power
100
plant capacity for electricity Offshore wind
generation from variable 0
ce
ce
0
10
10
renewable energy sources

en
en
ce
ce
er
er
n
n
f
(wind, sun) in 2030 and

Re
Re
re
re
fe
fe
Re
Re
2050 for the Reference and 2019 2050

2030
Reference100 scenarios
13 The capture and storage of CO2 underground (Carbon Capture and Storage, CCS) is not examined in this
study, as there is currently no legal basis for its implementation in Germany.
48
The differences shown in the installed capacities for electricity generation from variable
renewable energy sources are still negligible in 2030. This is because the reduction targets
of the two scenarios do not differ significantly up to this year. In 2050, the Reference100
scenario has a total installed capacity of 734 GWel, which is around 70 GWel greater than
the Reference scenario. About half of this additional capacity is accounted for by wind and
the other half by photovoltaic plants.
The required higher capacity of renewable power plants for electricity generation leads
to a correspondingly greater need for flexibility in order to effectively incorporate the
fluctuating electricity generation into the energy system. This is evidenced, for example,
by the installed capacity of short-term energy storage facilities and electrolysis plants. At
240 GWhel, the capacity of short-term energy storage facilities in 2050 is around 30 GWhel
higher in the Reference100 than in the Reference scenario. The installed capacity of elec-
trolysis plants is around 50 GWel, or about 12 GWel higher, respectively. In 2050, therefore,
hydrogen production in Reference100 is about 30 TWhH2 greater than in the Reference
scenario.
However, this additional plant capacity and the domestic production of hydrogen are
not sufficient to fully compensate for the loss of fossil fuels. For example, the imports of
synthetic liquid fuel will almost double, reaching 55 TWh. In the Reference100 scenario, the
amount of imported hydrogen will increase by around 30 TWhH2 to 146 TWhH2 in 2050.
The sectors in which the additional hydrogen (60 TWhH2 in total) is used and how the
hydrogen is provided are shown in Figure 25 for the year 2050.
TWhH2 (Final energy)

Green hydrogen
400
Steam reforming
350 Electrolysis
300 Import
Transport
250
Building heat
200
Process heat
150 H2 reconversion
100 Gas grid
Storage surplus
50
Storage losses
0
Figure 25:
ce
00
ce
00
n
n
1
1
re
re
ce
ce
Hydrogen supply and

fe
fe
n
n
Re
Re
re
re
fe
fe
use for the Reference and

Re
Re
Supply Use Reference 100 scenarios for

2050
The results show that most of the produced hydrogen is used in the transport sector and
in industry. Thus, the complete conversion in motorized private transport from conventional
combustion engines to electricity-based alternatives takes place in the Reference100
scenario two years earlier than in the Reference scenario. This means that more hydrogen-
electric vehicles are on the road, increasing the hydrogen demand by 56 TWhH2 in 2050.
The main difference lies in the conversion of the freight transport, however. For exam-
ple, the share of hydrogen-based heavy good vehicles increases from around 40 % in the
Reference scenario to twice that in the Reference100 scenario. (The remaining 20 % of
the transport fleet is made up of heavy good vehicles running on overhead contact-lines).
This changeover shows that in order to achieve a complete reduction of energy-related
CO2 emissions in all sectors, it is more cost-effective to replace the remaining internal
combustion engines in freight transport with hydrogen-electric powertrains and to use

domestically produced or imported hydrogen than to operate internal combustion engines
with synthetic liquid fuel. This statement for the year 2050 naturally has a high degree of
uncertainty, since it depends decisively on the import prices for synthetic energy carriers
produced abroad from renewable electricity.
A further percentage of the produced hydrogen is used in the industrial sector. Since coal
is substituted completely by hydrogen in steel production, more hydrogen is used in the
Reference100 than in the Reference scenario. At the same time, hydrogen fuel cells are
used to provide space heating and domestic hot water, which is the third highest demand
in both scenarios, regardless of the CO2 emissions target value (see Figure 25). A compara-
tively small proportion of the hydrogen is fed into the natural gas grid and then converted
back into electricity in gas turbines. The remaining quantity represents storage losses or the
stored carryover to the following year.
While the absolute value of the hydrogen supply differs in the two scenarios presented, the
proportion from the different sources of production differs only slightly. For example, elec-
trolysis in Germany accounts for about 60 % in each of the two scenarios and the import
of synthetic hydrogen for almost 40 %. Steam reforming and biohydrogen will only play a
minor role in 2050.
In summary, the comparison of the Reference and Reference100 scenarios shows that in
addition to a more extensive expansion of renewable power plants for electricity genera-
tion, the increased use of hydrogen plays a central role in reaching a 100 % reduction of
energy-related CO2 emissions by 2050.
3.3.2 Complete Reduction of Energy-related CO2 Emissions by 2035

For a development of the German energy system that is compatible with a target of 1.5 °C,
it would be necessary to completely reduce Germany’s energy-related CO2 emissions by
2035, based on Germany’s remaining CO2 budget [39]. The share of the remaining total
budget available for the energy sector would then amount to about
6 billion tonnes. The following analysis examines a transformation path leading to a com-
plete reduction of energy-related CO2 emissions by 2035.
It is therefore assumed that the achievement of a climate neutral energy system within 15
years would only be feasible if the transformation path is supported by broad social con-
sensus. Aside from this time scale, the Sufficiency2035 scenario is essentially based on the
assumptions of the Sufficiency scenario. Since these assumptions alone do not ensure the
technical feasibility of climate neutrality by 2035, three further assumptions are made:
1. the use of coal for steel production will be substituted by hydrogen by 2035
2. the minimum load at which coal-fired power plants must operate will be reduced
by half by 2035
3. the import of synthetic fuels will be possible from 2025, i.e. five years earlier than
in the other scenarios
In carrying out the model calculations, the annual growth rates and expansion potentials
of the technologies were not changed, nor were the rates of system replacements; for ex-
ample, there was no obligation assumed to replace conventional heating boilers before the
end of their service life. In this respect, the results can only provide a first indication of the
implications that such an accelerated phase-out of fossil fuels would have.
50
Based on the given assumptions, it can be seen that a complete reduction of energy-relat-
ed CO2 emissions by 2035 is only possible if considerable quantities of CO2-neutral energy
carriers are available for import. This makes it possible to completely displace fossil fuels
and reduce CO2 emissions to zero by 2035. Compared to all other scenarios, the required
import volume of synthetic energy carriers produced carbon-neutral is very high. This
amount will then decline again by 2050 (see Figure 26). The maximum import volume is
570 TWh and occurs in 2035.
TWh (Imported synthetic fuels)

600
500
400
300
200 Liquid fuels

Methane
100
Hydrogen
2020 2025 2030 2035 2040 2045 2050
After the initial increase in synthetic fuel imports, which are necessary to displace fossil fu- Figure 26:
els, a continuous reduction is evident, so that only about 70 TWh will be imported in 2050. Time series of imported
This roughly corresponds to the import volume of the other scenarios in 2050, showing synthetic energy carriers
that the import of synthetic fuels in this scenario takes place despite the high energy prices, from 2020 to 2050 in the
but is only a transitional solution on the way to CO2 neutrality. Under the given assump- Sufficiency2035 scenario
tions it is more favorable to gradually reduce the high import volume in 2035 by installing
renewable power plants to generate electricity and also plants to further convert it to other
forms (e.g. electrolysis). An example of this development is the continued growth in the
installed capacity of wind, photovoltaic and electrolysis plants after 2035 (Table 7).
Table 7: Installed capacity of power plants for the generation of electricity from variable
renewable sources and electrolysis plants in the Sufficiency2035 scenario.
Installed Plant Capacity in GWel 2020 2025 2030 2035 2040 2045 2050
Wind 72 104 144 185 226 230 232
Photovoltaics 60 105 181 264 327 362 384
Elektrolysis 0 4 11 32 61 75 83
The correlations presented show that a complete reduction of energy-related CO2 emis-
sions by 2035 will require considerable efforts. Corresponding cost aspects are discussed in
the following subchapter.

3.4
Cost Analysis
3.4.1 Total Cumulative Expenditure of the Energy Transformation

In the previous chapters, various possible transformation paths for the German energy
system were presented. These consider distinct and broadly effective attitudes of society
that are accordingly characterized by specifically defined boundary conditions. A cost opti-
mization was carried out under the given boundary conditions. As explained in Section 2.1,
the cumulative expenditure for conversion, expansion and operation of the entire energy
system for the period 2020 to 2050, which consists of investments (CAPEX) and all costs
(OPEX: energy carriers, operation, maintenance, transaction costs), is determined and min-
imized by the optimization calculations for each scenario. Depending on the scenario, this
value can vary considerably. Figure 27 shows the differences between the scenarios and
a Business-as-Usual (BAU) scenario. In the BAU scenario there is no system optimization,
i.e. the current state of the energy system is maintained 14 and emissions are only slightly
reduced by assumed efficiency increases (technology learning curves). Investments are
made exclusively to replace systems at the end of their service life with similar systems.
Bn. € (Differential expenditures)

4500
4000
3500 Bn. € (Diffe
3000
4000
Net additional expenditure
2500
VRE
2000 3000
Building refurbishments
1500
Heat supply 2000
1000
Transport
Figure 27: 500 Infrastructure, 1000
Differential expenditures for Storage, Converters
the various scenarios compa- 0 Imports: electricity, biomass,
red to the business-as-usual synthetic energy carriers 0
-500 Fossil energy carriers and
(BAU) scenario. The net additi-
power plants
onal expenditures per scenario
-1000 -1000
ce
ce
cy
5
nc
10
03
(diamonds) are calculated by

en
n
an
re
te
ce
2
ci
pt
cy
fe
sis
n
ffi
ce
summing the positive and ne-

ien
Re
re
r
Su
Pe
Ac
fe
c
Re
ffi
n-
gative expenses (VRE: variable

Su
No
renewable energy).
Figure 27 shows a cost comparison for all the scenarios examined. It shows the additional
expenses in comparison with the BAU scenario, differentiated according to the areas in
which these additional expenses are incurred. Negative values describe reduced expendi-
ture, in particular due to the lower costs of the import of fossil fuels in all climate protec-
tion scenarios. The sum of additional and reduced expenditures, represented by a hash
mark, describes the net additional expenditures for the scenarios examined.
14 The legal decision to phase-out of nuclear power in Germany in 2022 is also taken into account in the BAU
scenario.
52
These net additional expenditures amount to around € 1580 billion for the Reference
scenario, € 2330 billion for the Persistence scenario, € 1590 billion for the Non-Acceptance
scenario and € 440 billion for the Sufficiency scenario. These four scenarios are character-
ized by a reduction of energy-related CO2 emissions by 95 % relative to the 1990 value.
The diagram also shows the values for the scenarios which reach a complete reduction
of energy-related CO2 emissions. The corresponding value for the Reference100 scenario,
in which the complete reduction of energy-related CO2 emissions is achieved by the year
2050, amounts to 2100 billion €. The value for the Sufficiency2035 scenario, in which
the complete reduction of energy-related CO2 emissions is achieved already in 2035, is
3330 billion €. The average annual net additional expenditure over the entire period from
2020 to 2050 is € 14 billion annually for the Sufficiency scenario and € 107 billion for the
Sufficiency2035 scenario, respectively. The corresponding average values for the other
scenarios with a 95 % reduction in energy-related CO2 emissions by 2050 are around
€ 51 billion for the Reference and Non-Acceptance scenarios and around € 72 billion for
the Persistence scenario. In relation to Germany’s gross domestic product in 2019,15 the
net additional expenditures range from 0.4 % (Sufficiency scenario) to 1.5 % (Reference
and Non-Acceptance) and around 2 % (Persistence and Reference100) and up to 3.1 %
(Sufficiency2035). To give another impression, the annual amounts mentioned above are
compared to the business turnover during Christmas time, which for 2019 in Germany
was 16 just under € 102 billion, i.e. about twice the average net annual expenditure for the
energy system transformation in the Reference and Non-Acceptance scenarios.
When compared to the BAU scenario, the Reference scenario is used to illustrate the main
contributions leading to additional expenditures. At € 804 billion, the highest additional
expenditure is used for investments in power plants that generate electricity from variable
renewable energy sources. The additional expenditures for the expansion of infrastructure
(e.g. grids), storage technologies and converters for the production of synthetic chemical
energy carriers from renewable electricity (Power-to-Gas, Power-to-Fuel) total about € 500
billion. The third-largest additional expenditure of around € 255 billion is attributable to the
energy refurbishment of the building sector. This includes the expenditures for converting
radiator heating systems to low-temperature (panel) heating systems, which will enable an
even more efficient use of heating technologies, especially heat pumps. Furthermore, in the
scenarios in which an emission reduction is taken as a basis, the import of synthetic fuels
from abroad causes additional expenditures. This expenditure amounts to € 260 billion in
the Reference scenario. Finally, an additional expenditure of € 196 billion is attributable
to conversion technologies in the consumption sectors for the provision of heat (building
heating and industrial process heat) and € 170 billion in transport. In terms of the under-
lying cost trend, battery electric vehicles will achieve cost parity with conventional com-
bustion vehicles as early as 2028, which is why the additional expenditure in the transport
sector is primarily attributable to freight transport.
15 Germany’s gross domestic product in 2019 was € 3436 billion. See e.g.:
https://de.statista.com/statistik/daten/studie/1251/umfrage/
entwicklung-des-bruttoinlandsprodukts-seit-dem-jahr-1991/
16 See https://de.statista.com/statistik/daten/studie/2750/umfrage/weihnachtsumsaetze-des-einzelhandels/

The differences in the composition of costs in the Non-Acceptance scenario and the Ref-
erence scenario differ only slightly, although the costs of importing synthetic fuels are also
slightly higher in the Non-Acceptance scenario. The Sufficiency scenario has the smallest
overall difference. At just under € 440 billion, this is around 30 % of that of the Reference
scenario and 20 % of the Persistence scenario. The decrease in consumption assumed in
the Sufficiency scenario leads to a reduction of all cost components (see Figure 27), which
makes the savings in fossil fuels more significant. For example, the savings in the Reference
scenario amount to around 30 % of the additional costs shown, whereas in the Sufficiency
scenario these savings amount to 40 %.
In the Reference100 scenario, a complete reduction in energy-related CO2 emissions by

2050 was set as the boundary condition. When looking at the differences, it becomes
clear that the shares of the individual cost items in the cumulative expenditures are almost
the same as in the Reference scenario. This means that in the Reference100 scenario the
main trends of the Reference scenario are adopted and reinforced in order to displace the
remaining fossil energy sources. Despite the more ambitious CO2 reduction target, the net
additional expenditures of the Reference100 scenario are lower than those of the Persis-
tence scenario. This result makes it clear that, from a cost perspective, the insistence of
conventional technologies (especially combustion technologies in the transport and build-
ing heating sectors) means a considerable additional expenditure is needed to achieve the
climate protection targets.
Finally, a complete reduction of energy-related CO2 emissions by 2035 was considered in

the scenario Sufficiency2035. Here the decrease in energy consumption, assumed in the
Sufficiency scenario, was adopted. The net additional expenditure of € 3330 billion is more
than double that of the Reference scenario and indicates that a complete reduction of
energy-related CO2 emissions by 2035 is associated with considerable additional expendi-
ture, also from a cost perspective. This can be explained by using the example of imports
of synthetic energy carriers. These represent the largest block of costs in this scenario (see
Figure 27). For example, around 570 TWh of synthetic fuels are already imported in 2035
in this scenario, while the import volumes in the other scenarios are still comparatively
small in 2035 and increase continuously until 2050 (see Figure 12 and Figure 26). Thus, the
volume of synthetic fuels imported in the Sufficiency2035 scenario in 2035 is similar to that
in the Persistence scenario in 2050. A comparison of the import volumes over the entire
period under consideration shows that they are almost a third higher in the Sufficiency2035
scenario than in the Persistence scenario. This leads to correspondingly higher costs, since
the assumed import prices of synthetic fuels in 2035 are on average 45 % higher than in
2050. Corresponding cost reductions over time also apply to many technical components,
especially those that are still in the early stages of market introduction and where signifi-
cant cost reductions can still be expected due to further technological development and
industrial production. Since the switch to a complete reduction of energy-related CO2
emissions by 2035 occurs earlier than in the other scenarios, the overall costs are higher
over the entire period.
54
Bn. € (Differential expenditures)
3500
3000
2500
2000
1500
1000
Net additional
expenditure
500
Figure 28:
Energy carrier costs
Cumulative expenditures
0
Operation and over the entire period of
maintenance costs consideration (2020 to 2050)
-500
Investments for all scenarios examined,
-1000 separated into investments
(including replacement invest-
ce
35
nc
nc
nc
10
en
ments and capital costs), other

20
cie
te
ta
e
er
nc
sis
ep
y
ffi
f
nc
Re
re
operating and maintenance

r
Su
Pe
Ac
cie
fe
Re
n-
ffi
costs and costs of the energy

No
Su
carriers (in € billion).
Figure 28 shows an overview of the expenditures incurred from 2020 to 2050 for all energy
carriers (both fossil and imports produced with renewable energy), operating and mainte-
nance costs, and investments (including replacement investments and capital costs). Here,
too, positive shares show the additional expenditure and negative shares show reduced
expenditure, compared to the BAU scenario. The net additional expenses are marked by
the hashes and the values are identical to those in Figure 27.
The results show that the additional expenditure for investments is between 63 % and
75 % in almost all scenarios. These investments are necessary to restructure the energy
system. When the transformation is (largely) completed in 2050, the investments will de-
crease significantly, since from then on only replacement investments have to be made.
The cost analysis carried out does not provide a complete picture of total social (or eco-
nomic) costs for two reasons:
■ No external cost analyses have been carried out for any of the developments – neither
for the BAU scenario nor for the climate protection scenarios, e.g. environmental or
health costs arising as a result of one development or another.
■ No macro-economic analysis was carried out that took value added and employment
issues into account.

3.4.2 CO2 Avoidance Costs
CO2 avoidance costs can be calculated based on the results of the calculations performed.
For this purpose, the net additional costs resulting for a scenario are divided by the CO2
emissions avoided, compared to the BAU scenario. Figure 29 shows the corresponding
values for all six scenarios, each shown as mean values for the years 2021 to 2030, 2031
to 2040 and 2041 to 2050, as well as the mean values for the entire period considered,
from 2021 to 2050.
€/tCO2 (CO2 avoidance costs)

350
300
250
200
150
Figure 29:
100
CO2 avoidance costs of all six
scenarios examined. In each 50
case, the mean values for the
years 2021-2030, 2031-2040 0 2021-2030 2031-2040 2041-2050 2021-2050
and 2041-2050 are shown as
well as the mean value for the
entire period of consideration Reference Persistence Non-Acceptance Sufficiency Reference100 Sufficiency2035
from 2021 to 2050.
With the exception of the Sufficiency2035 scenario, the CO2 avoidance costs in all other
scenarios rise continuously over the three decades considered. This reflects the principle
of mathematical optimization, i.e. first taking the most cost-effective avoidance measures
and associated investments and then later implementing the more costly measures. In the
Sufficiency2035 scenario, the high quantity of imported synthetic energy carriers produced
abroad with renewable electricity has a strong impact in the years from 2031 to 2040. The
highest value of more than 300 €/tCO2 occurs in this decade.
For the Sufficiency scenario, the CO2 avoidance costs are by far the lowest. They average
50 € per tonne of CO2, since the assumed reduction in energy demand due to behavioral
changes is not associated with costs. For the Reference scenario, costs average around
150 €/tCO2 and they rise over the three decades from around 50 €/tCO2 in 2021-2030, to
142 €/tCO2 in 2031-2040, to just over 180 €/tCO2 in 2041-2050. Both the development
and the values in the Non-Acceptance scenario have a similar order of magnitude as the
Reference scenario. Over the entire period under consideration, the CO2 avoidance costs
average about 162 €/tCO2 and are thus around 7 % higher than for the Reference scenario.
Based on the model results, a complete reduction of energy-related CO2 emissions by

2050, as in the Reference100 scenario, leads to CO2 avoidance costs which are about
40 €/tCO2 higher than in the comparable case yet with only 95 % emissions reduction by
2050. The highest costs are caused by a strong insistence on continuing to use today’s
technologies, such as combustion technologies for heat provision and for the transport
sector. In the Persistence scenario, the average CO2 avoidance costs are 233 €/tCO2, which
are on the same order of magnitude as for the Sufficiency2035 scenario.
56
The CO2 avoidance costs given here demonstrate that changes in behavior in particular can
have a considerable impact on the costs of energy transition. Collective societal behavior
that saves energy and reduces emissions in large sections of society, e.g. induced by a
change in values, would not only require less financial expenditure and less investment. It
would also result in less expansion of renewable energy installations and other technical
equipment in the future energy system, and thus presumably lead to greater acceptance
of the changes accompanying the energy system transformation.

4
Conclusion
This study analyzed the influence that key societal behaviors and attitudes have on the
progress of the energy transition and how this affects the necessary investments and costs
of restructuring the energy system. The results show that it is technically and systemically
feasible to reach the climate protection targets for the energy supply based on renewable
energy sources. The model calculations, carried out on an hourly basis for the next thirty
years, show that a secure energy supply throughout all consumption sectors is guaranteed
for each hour, despite the high share of variable renewable energy sources in the electricity
supply.
The different scenarios also illustrate that the differences in the expenditure and costs
required to achieve the goals strongly depend on the framework conditions that are largely
determined by societal behavior and attitudes. For example, if large parts of society were
to change their behavior towards a more economical use of energy, this would have a con-
siderable impact. As a result, the necessary amount of facilities for the conversion, storage,
distribution and use of energy and the associated costs would be substantially lower than
in all other scenarios considered. In contrast, adhering to the use of combustion-based
technologies for the heat supply and transport would lead to substantially higher capacity
requirements for renewable energy power plants and other related technical facilities. Also,
the import of synthetic, chemical energy carriers, produced abroad on the basis of renew-
able electricity, would increase. Such persistent behavior would make the energy transfor-
mation more expensive. Strong resistance to the expansion of large infrastructures such as
wind turbines and grids can be partially compensated by a modified path, albeit at slightly
higher costs than in the case of cost-optimal development. Greater installments of photo-
voltaic systems and battery storage are elements of such a path.
An essential prerequisite for a comparatively cost-effective achievement of the climate tar-

gets is the continuous development of all relevant technologies for the conversion, storage,
distribution, use and system integration of renewable energy. Only then will it be possible
to achieve the projected reductions in costs and the increases in performance and service
life. The study has also made it clear that the use of thermal and electrical storage systems
in Germany is just as meaningful as the establishment of domestic production, processing
and use of hydrogen in a wide range of applications. The development of domestic markets
is important for all technologies and furthermore contributes significantly to successful
technology development. On the one hand, markets for local manufacturers are created,
and on the other hand, experience can be gained with the corresponding plants and their
system integration. These are also essential prerequisites for achieving the desired improve-
ments in costs and performance. At the same time, knowledge of these technologies and
confidence in their reliability grows. This is also an important factor for export.
However, applied research and development does not end with component development;
it can also make important contributions to increasing market integration. The concept of
integrated photovoltaics is an example of this. It is foreseeable that a massive expansion of
large open space photovoltaic systems could lead to conflicts and possibly to acceptance
problems. On the other hand, the integration of photovoltaics into building envelopes,
vehicle bodies and roads as well as agricultural land and water surfaces, will open up huge
areas already used for other purposes for the dual use of solar power generation. Creating
the appropriate products and solutions for these applications will be an important task in
the further development of photovoltaics and will present new opportunities for domes-
tic production. Electric heat pumps are another example. The investigations carried out
underscore the important role of heat pumps in achieving a cost-effective transition in the
heat sector. However, especially in urban areas regulations that limit the exploitation of
heat sources are in place; noise emissions can also cause acceptance problems. Here, too,
applied research and development in close cooperation with manufacturers and users is
58
needed to address the arising problems and to develop new solutions. These are just two
concrete examples of the many questions that need to be answered in order to develop
customized solutions that are highly relevant for the successful implementation of the
energy transition.
The future energy system will be characterized by a large number of interconnected sys-
tems. These will interact with each other and operate so as to provide as much dynamic
support to the system as possible. Photovoltaic systems, heat pumps, stationary battery
storage systems and charging stations for electric vehicles, but also technical systems,
many in the small power range, are particularly important here. In the coming decades, the
number of these systems is expected to reach the double-digit millions. Against this back-
ground, the development of solutions for efficient, stable and reliable system integration
and operation of these many components plays a decisive role. This can be achieved only
by using modern concepts from the information and communications technology (ICT).
Application-oriented system research is just as relevant for the development of feasible
solutions and business models as for energy system analysis, which serves as a compass for
the successful development of the total energy system towards a climate-neutral energy
supply.
With this study, we hope to contribute valuable input to the discussion on the feasibility of
achieving a climate-neutral energy system, which is essentially based on two main pillars:
renewable energy use and high efficiency in energy conversion and utilization.

Bibliography
[1] EUROPÄISCHE KOMMISSION: Ein europäischer Grüner Deal: Erster klimaneutraler
Kontinent werden.
URL: https://ec.europa.eu/info/strategy/priorities-2019-2024/european-green-deal_de
[2] PALZER, Andreas: Sektorübergreifende Modellierung und Optimierung eines zukünf-
tigen deutschen Energiesystems unter Berücksichtigung von Energieeffizienzmaßnah-
men im Gebäudesektor. Karlsruhe, Karlsruher Institut für Technologie. Dissertation.
2016-04-28. URL: http://publica.fraunhofer.de/documents/N-408742.html
[3] STERCHELE, Philip: Analysis of Technology Options to Balance Power Generation
from Variable Renewable Energy: Case Study for the German Energy System with the
Sector Coupling Model REMod. 1. Auflage. Düren: Shaker, 2019 (Schriftenreihe der
Reiner Lemoine-Stiftung)
[4] DEUTSCHES INSTITUT FÜR NORMUNG E.V.: Energieeffizienz von Gebäuden –
Berechnung des Energiebedarfs für Heizung und Kühlung (ISO 3790:2008);
Deutsche Fassung EN ISO 13790:2008 (in Kraft getr. am 09.2008) (09.2008)
[5] BMWI: Gesamtausgabe der Energiedaten - Datensammlung des BMWi. 2019.
URL: https://www.bmwi.de/Redaktion/DE/Binaer/Energiedaten/
energiedaten-gesamt-xls.html
[6] TROST, Tobias; HORN, Sönke; JENTSCH, Mareike; STERNER, Michael: Erneuerbares
Methan: Analyse der CO2-Potenziale für Power-to-Gas Anlagen in Deutschland.
In: Zeitschrift für Energiewirtschaft 36 (2012), Nr. 3, S. 173–190 – Überprüfungsdatum
2015-03-17
[7] HANSEN, Nikolaus: The CMA Evolution Strategy: A Tutorial. 4/4/2016
[8] ECKERT, Werner: CO2-Ausstoß überraschend stark gesunken.
URL: https://www.tagesschau.de/wirtschaft/co2-ausstoss-deutschland-101.html –
Überprüfungsdatum 2020-02-04
[9] BMWI, BMU: Energiekonzept: für eine umweltschonende, zuverlässige und
bezahlbare Energieversorgung. München, 2010
[10] SCHULER, Johannes; KRÄMER, Christine; HILDEBRANDT, Silvio; STEINHÄUßER,
Reimund; STARICK, Anja; REUTTER, Michaela: Kumulative Wirkungen des Ausbaus
erneuerbarer Energien auf Natur und Landschaft.
Bonn: Bundesamt für Naturschutz, 2017 (BfN-Skripten 463)
[11] BMVI: Räumlich differenzierte Flächenpotenziale für erneuerbare Energien in
Deutschland. Berlin, 2015
[12] FATH, Karoline: Technical and economic potential for photovoltaic systems on
buildings. 2018
[13] MASUROWSKI, Frank; DRECHSLER, Martin; FRANK, Karin: A spatially explicit
assessment of the wind energy potential in response to an increased distance
between wind turbines and settlements in Germany, 2016
[14] DEUTSCHER BUNDESTAG: Artikel 1 - Dreizehntes Gesetz zur Änderung des Atomge-
setzes (in Kraft getr. am 2.7.2011) (2011-07-02).
URL: https://www.bmu.de/gesetz/13-gesetz-zur-aenderung-des-atomgesetzes/ –
Date of verification” 2020-02-04
[15] BBH; LBST; Fraunhofer ISE; IKEM:
Integriertes Energiekonzept 2050: Strom Wärme Verkehr Industrie. 2018
[16] SCHUBERT, Markus; KLUTH, Tobias; NEBAUER, Gregor; RATZENBERGER, Ralf;
KOTZAGIORGIS, Stefanos; BUTZ, Bernd; SCHNEIDER, Walter; LEIBLE, Markus:
Verkehrsverflechtungsprognose 2030. 11.2014
60
[17] SENKPIEL, C.; BERNEISER, J.; BAUMANN, D.: Open-source simulation of the long-term
diffusion of alternative passenger cars on the basis of investment decisions of private
persons: Transportation Research Part A - Policy and Practice (publication planned).
2020
[18] BÜRGER, Veit; HESSE, Tilmann; PALZER, Andreas; HERKEL, Sebastian; ENGELMANN,
Peter: Klimaneutraler Gebäudebestand 2050. Dessau - Roßlau, 2016
[19] BLESL, Markus ; KESSLER, Alois: Energieeffizienz in der Industrie.
Berlin: Springer Verlag, 2013
[20] ZELL-ZIEGLER, C.; FÖRSTER, H.: Mit Suffizienz mehr Klimaschutz modellieren:
Zwischenbericht. Juli 2018
[21] BÜRGER, Veit; BERN, Mari Roald (Mitarb.): Identifikation, Quantifizierung und Sys-
tematisierung technischer und verhaltensbedingter Stromeinsparpotenziale privater
Haushalte. 3. Aufl. 2009
[22] UMWELTBUNDESAMT: Energiebedingte Emissionen.
URL: https://www.umweltbundesamt.de/daten/energie/
energiebedingte-emissionen#textpart-1
[23] BMWI: Strommarkt der Zukunft: Stromerzeugung: Zahlen und Fakten.
URL: https://www.bmwi.de/Redaktion/DE/Dossier/strommarkt-der-zukunft.html. –
Aktualisierungsdatum: 2019-03-06 – Überprüfungsdatum 2020-01-23
[24] BMWI: Kohleausstieg und Strukturwandel.
URL: https://www.bmwi.de/Redaktion/DE/Artikel/Wirtschaft/
kohleausstieg-und-strukturwandel.html – Überprüfungsdatum 2020-02-04
[25] FRAUNHOFER-INSTITUT FÜR SOLARE ENERGIESYSTEME ISE: Energy Charts.
URL: https://www.energy-charts.de/ – Überprüfungsdatum 2020-02-04
[26] ICHA, Petra: Entwicklung der spezifischen Kohlendioxid-Emissionen des deutschen
Strommix in den Jahren 1990 - 2018. 2019
[27] BMU (Hrsg.): Klimaschutzprogramm 2030 der Bundesregierung zur Umsetzung des
Klimaschutzplans 2050. 2019
[28] UMWELTBUNDESAMT: Primärenergieverbrauch.
primaerenergieverbrauch#textpart-1 – Überprüfungsdatum 2020-01-23
[29] UMWELTBUNDESAMT: Primärenergiegewinnung und -importe.
primaerenergiegewinnung-importe – Überprüfungsdatum 2020-01-23
[30] ARPAGAUS, Cordin; BLESS, Frédéric; BERTSCH, Stefan; SCHIFFMANN, Jürg:
Wärmepumpen für die Industrie: Eine aktuelle Übersicht. Tagung des
BFE-Forschungsprogramms, 2019
[31] ARPAGAUS, Cordin: Hochtemperatur-Wärmepumpen: Marktübersicht, Stand der
Technik und Anwendungspotenziale. Berlin, Offenbach: VDE Verlag GmbH, 2019
[32] SALZGITTER: Low CO2 Steel Making.
URL: https://salcos.salzgitter-ag.com/ – Überprüfungsdatum 2020
[33] UMWELTBUNDESAMT: Energieverbrauch nach Energieträgern, Sektoren und
Anwendungen.
energieverbrauch-nach-energietraegern-sektoren
[34] UMWELTBUNDESAMT: Indicator: Greenhouse gas emissions.
URL: https://www.umweltbundesamt.de/en/
indicator-greenhouse-gas-emissions#textpart-1

[35] KRAFTFAHRT-BUNDESAMT: Jahresbilanz der Neuzulassungen 2018.
URL: https://www.kba.de/DE/Statistik/Fahrzeuge/Neuzulassungen/neuzulassungen_
node.html;jsessionid=C6762B12F285B12F931A0366B034D2EF.live11292
[36] STATISTA: Anzahl der Neuzulassungen von Pkw in Deutschland von 1955 bis 2019.
URL: https://de.statista.com/statistik/daten/studie/74433/umfrage/neuzulassungen
-von-pkw-in-deutschland/
[37] DEUTSCHER BUNDESTAG: Energieeinsparverordnung (in Kraft getr. am 1.5.2014)
(2014-05-01).
URL: www.enev-online.de – Überprüfungsdatum 2015-02-09
[38] FEIST, Wolfgang: EnerPHit und EnerPHio+i – Zertifizierungskriterien für die Modernisi-
erung mit Passivhaus-Komponenten.
URL: http://www.passiv.de/de/03_zertifizierung/02_zertifizierung_gebaeude/
04_enerphit/04_enerphit.htm – Überprüfungsdatum 2015-03-19
[39] ROGELJ, J.; SHINDELL, D.; JIANG, K.; FIFITA, S.; FORSTER, P.; GINZBURG, V.;
HANDA, C.; KHESHGI, H.; KOBAYASHI, S.; KRIEGLER, E.; MUNDACA, L.:
Mitigation Pathways Compatible with 1.5°C in the Context of Sustainable
Development: Chapter 2, Table 2.2.
URL: https://www.ipcc.ch/site/assets/uploads/sites/2/2019/05/
SR15_Chapter2_Low_Res.pdf – Überprüfungsdatum 2020-01-23
[40] HENNING, H-M.; AGERT, C.; VOGT, T.; ROHRIG, K.; NIEPELT, R.; FICHTNER, W.;
GAWEL, E.; FISCHEDICK, M.: Digitalisierung und Energiesystemtransformation –
Chancen und Herausforderungen. In: FVEE-Themen “Die Energiewende – smart und
digital. Beiträge zur FVEE-Jahrestagung 2018”. Berlin 2019, ISSN · 0939-7582
62
FRAUNHOFER INSTITUTE FOR SOLAR ENERGY SYSTEMS ISE
Contact:
Dr. Christoph Kost
[email protected]
Fraunhofer Institute for Solar Energy Systems ISE

Heidenhofstraße 2
79110 Freiburg / Germany
www.ise.fraunhofer.de
Institute Directors:
Prof. Dr. Hans-Martin Henning
Prof. Dr. Andreas Bett
Titelfoto: Simon Dannhauer iStock / Grafik und Gestaltung: triolog

Fraunhofer ISE Study Paths To A Climate Neutral Energy System

Uploaded by

Copyright:

Available Formats

Fraunhofer ISE Study Paths To A Climate Neutral Energy System

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

Fraunhofer ISE Study Paths To A Climate Neutral Energy System

Uploaded by

Copyright:

Available Formats

FRAUNHOFER INSTITUTE FOR SOLAR ENERGY SYSTEMS ISE

Paths to a Climate-Neutral Energy System

Philip Sterchele, Julian Brandes, Judith Heilig,

Fraunhofer Institute for Solar Energy Systems ISE, Freiburg

Freiburg, February 2020

Hans-Martin Henning Andreas Bett

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 3

The Persistence scenario is characterized, in particular, by strong resistance to the use of

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 5

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 7

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 9

It is becoming increasingly obvious that, in addition to questions of technical feasibility

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 11

2.1 The Renewable Energy Model

Energy sources Energy conversion/storage Consumption

Wind Wind turbine

Sun Run-of-River PP Vehicle battery

Environmental Geothermics Stationary

Electricity Process heat

Coal and oil Gas

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 13

Hydrogen can be used in transport, industry (materially or energetically to generate process

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 15

In addition to domestic production, synthetic energy carriers produced from renewable

The mathematical optimization identifies cost-optimized transformation paths that com-

Mio. t (Energy-related CO2 emissions)

Energy-related CO2 emissions

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 17

Table 1: Description of key boundary conditions in the scenarios examined6 7

Industrial Process Heat: Moderate development. Due to technological innovations and

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 19

Transport: Conservative development. Conventional internal combustion engines based

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 21

3.1.1 Primary Energy and Final Energy

3.1.2 Development of Electricity Demand

The decline of conventional fossil-fuel powered technologies in favor of electricity-based

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 23

Electricity Demand in TWhel Reference Persistence Non-Acceptance Sufficiency

TWhel (Electricity supply)

1250 Photovoltaics Gas turbine PP

Onshore wind CHP (Heat grid)

500 Electricity imports Nuclear PP

Reference Persistence Non-Acceptance Sufficiency

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 25

GWel (Installed capacity)

of photovoltaic systems and

wind turbines in 2030 and

2050 for the four scenarios

examined 2019 2030 2050

3.1.4 Balancing Variable Electricity Generation

GWel (Installed capacity)

or controllable power plants

in 2030 and 2050 in the

four scenarios examined

Fraunhofer ISE | Paths to a Climate-Neutral Energy System 27

GWhel (Battery capacity)

GWel (Installed capacity)

synthetic energy carriers;

ElectrolysisH2: for the pro-

renewable energy sources