Set3080 Summary of All Lectures

lOMoARcPSD|2686941
SET3080 Summary of all lectures
The Necessity of Storage Technology (Technische Universiteit Delft)
Studeersnel wordt niet gesponsord of ondersteund door een hogeschool of universiteit

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)
lOMoARcPSD|2686941
SET3080 Summary
The Necessity of Energy Storage
Author:
Andrew Spekreijse
Date:
January 27, 2022

lOMoARcPSD|2686941
About this summary
This summary is pretty much finished. I have added more content in the economics section
of the summary. I have also added a bunch of possible exam questions to the summary with
hopefully clear explanations. I doubt I will add much more.
So this is a summary I have made for SET3080 The Necessity of Energy Storage. I haven’t
been able to spend as much time as I’d like on this because of the many assignments the Biomass-
Storage-Economics track has in Q2. I am not completely sure what the exam will be like; the
questions from the sample questions and exam seem more concept questions rather than knowing
how to do calculations (the calculations which do exist are very straightforward with a bachelor
degree behind you).
This summary is spilt into two parts:
• Summary of all covered material, separated by topic
• Worked out questions for the questions asked in the workgroup classes
Like the previous summaries I have made, let me know if anything in this summary is incorrect,
misleading or unclear. I am a student as much as you all are and I have a lot to learn still as well.
Feel free to share this summary to anyone you want, but if you do so, can you please do it
though sharing this link https://l.linklyhq.com/l/iVX9. I am curious how many people get use
out of it. This work is shared under a Creative-Commons BY-SA 4.0 license. So you can share
and adapt, just give credit and share whatever you make under the same terms. This summary is
based off the lecture slides for the lectures given by Prof.dr. F.M. Mulder.
2
lOMoARcPSD|2686941
Contents
1 Topics 5
1.1 Why Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Energy Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Solar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.2 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Energy use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.1 Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.2 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 Demand response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3.4 Energy transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.4 Energy storage methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4.1 Hydroelectric and pumped hydro . . . . . . . . . . . . . . . . . . . . . . . . 17
1.4.2 Biomass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.4.3 Batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.4.4 Artificial fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.4.5 Flywheels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1.4.6 Compressed Air Energy Storage (CAES) . . . . . . . . . . . . . . . . . . . . 33
1.4.7 Heat Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
1.5 Economics of storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
1.5.1 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
1.5.2 Battery Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
1.5.3 Scaling laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
1.5.4 Energy Market time slots . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
1.5.5 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3
lOMoARcPSD|2686941
Page 4 of 66
2 Exam/worked out questions 45

2.1 Solar Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.2 Wind questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.3 Biomass questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.4 General generation questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.5 Hydrogen questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.6 Ammonia questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.7 General storage questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.8 Economic related questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.9 CAES questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.10 Heat storage questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.11 Demand questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.12 Energy transport questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Written by Andrew Spekreijse

lOMoARcPSD|2686941
Chapter 1
Topics
1.1 Why Storage
As you know, the most abundant renewable energy sources are wind and solar. You also know that
these are intermittent energy sources. This naturally requires us to store the energy these sources
produce. That is the entire point of this course. Currently, when too much energy is produced,
negative prices (for electricity) can result on the spot market. This is an example of oversupply
as some generation sites cannot be turned off. This is not good for when companies decide to
invest in renewables because they are getting a very bad return on their investment. So if we can
store this energy instead, and then return it to the grid when prices are higher, MONEY can be
made. Also, it means we can balance the grid and match energy demand with renewable energy
generation.
1.2 Energy Production
Now do not forget, electricity is not equal to energy. While electricity is energy, there are far more
sources of energy used throughout the world than just electricity consumption.
The most common renewable energy discussed is solar and wind. There exist a variety of models
to predict solar and wind energy. The models given in the lectures are based off the USA’s current
energy generation so might not be the most suitable for all regions. Using more first principal
models (such as tracking solar flux or using local wind data) would be more accurate and reliable.
5
lOMoARcPSD|2686941
Page 6 of 66
Figure 1.1: Variation in average solar irridance in Europe
1.2.1 Solar
For solar energy, the impacts of the amount of flux received changes depending on the season
because of Earth’s tilt and because of Earth’s elliptical orbit around the sun. The southern and
northern hemisphere’s experience these effects oppositely (it is summer in the southern hemisphere
when it is winter in the northern). Now the equator and equatorial regions do not experience this
phenomena generally (they mildly do because of the elliptical orbit) because the Earth’s tilt does
not impact them.
As you also know, the higher the absolute value of latitude, the longer the day is during the
summer months in the respective hemisphere. The closer you get to the equator, the more equal
the day length is throughout the year. Additionally, the flux (”strength”) of the sun increases the
closer you get to equator. Combining these results, we find that locations closer to the equator
generally get more energy per year (per unit of area) than locations at higher absolute values of
latitudes. This can be seen for Europe in Figure 1.1.

lOMoARcPSD|2686941
Page 7 of 66
1.2.2 Wind
While wind speeds are linked to solar irradiation, the relationship between the sun shining and
the wind blowing is not as trivial as would be convenient. Wind energy (as you know) comes from
extracting kinetic energy from the wind. Average wind speeds varies with time of year. From
your own experience, you have probably noticed that there is more bad weather (including strong
winds) in winter compared to summer
In general, the larger the diameter of a wind turbines the more energy you can obtain given
a constant wind speed. Because of aerodynamics, once you can taken energy out of the wind
using a wind turbine, the area downstream of the turbines will be a region of ”turbulent” air.
Very long story short, there will be less energy to extract from the wind immediately downstream.
Distance is required for the airflow to ”recover” and become less turbulent and therefore provide
more energy to a subsequent turbine (we are ignoring wake steering here).
Because of their location, offshore wind farms tend to have higher energy yields than onshore
farms (however they are more expensive to build).
Wind energy also depends on geographical location; how close are you to mountains, how
close are you to the sea, what latitude? Mountains and see impact thermal currents in the air
which are directly impact wind speed. Latitude can matter because trade winds (the winds which
consistently blow around the world because of the air’s interaction with ocean currents). However
there is no general ”rule” you can follow for this as all of Earth’s feature impact that the wind
does.
1.3 Energy use
Currently, most energy we use is in the form of storable fuel, namely fossil fuels. If you think of
transport (specifically cars), electrification is easy. However, you might think of other transport
methods (such as planes) or other sectors, such as industry where it is a lot harder to use electricity.
Looking at Figure 1.2 you can see that most of the energy used by industry is heat energy (not
electricity). This makes sense as many industrial processes use high temperatures (such as Haber
Bosch, metal forging etc). This figure does not show information about the type of energy use by
other sectors such as residential or transport, but you probably know from your own experience
that most is either already electricity or can easily be done by electricity.

lOMoARcPSD|2686941
Page 8 of 66
Figure 1.2: Share and breakdown of heat demand in industry
1.3.1 Industry
So now we know that industry is the a big part of the challenge (and I am sure you realise we
need industry because having clothes, phones, cars, planes, cargo ships, ovens, glass etc is all very
useful for everyday living), how we we make industry fossil fuel free? Logically, we know that we
need to produce heat and produce lots of it. Such options are
• Electric heaters (joule heating)
• Solar heat, geothermal
• Burning a renewable fuel (bio-char, bio-oil, hydrogen, ammonia)
• Using heat pumps
Of those options, using a fossil free chemical fuel (carbon based or otherwise) is the most viable
option. This is because most industry doesn’t just need heat, but needs high temperature heat on
a large scale (high temperatures are from 1000 Celsius and up).
As we discussed in SET3013, carbon is needed to create steel from iron. There are projects
being worked on currently (for you duchies, you might know that local steel manufacturer, TATA
steel has one), to produce steel using hydrogen as the heat source (and then I presume add carbon

lOMoARcPSD|2686941
Page 9 of 66
separately from renewable sources). Hydrogen is able to burn hot enough and chemically reduce
iron ore to iron which currently is done by burning coal. Hydrogen, of course, can be produced
from electrolysis of water, and therefore can indirectly be made renewable. Now no matter what,
we still require carbon to make iron into steel. This carbon can come from bio-char (which is
derived from biomass).
Ammonia is another chemical which uses high temperatures to be produced via the Haber
Bosch process. In case you don’t know, ammonia is the world’s most produced chemical as it is
the base feed-stock for the majority of the world’s fertilizers (and if we didn’t have fertilizers, most
of the world’s population would starve). Now ammonia uses the elements hydrogen and nitrogen
when being produced. This hydrogen usually comes from natural gas (methane) because methane
is cheaper than electrolysing water. However, the hydrogen required for ammonia production can
come from renewable electricity and the temperatures required for harbor botch can also come
from burning hydrogen or (as the temperatures required are only 500 C), joule heating can be
used or some sort of heat pump). Fun fact, 1.5% of the worlds energy use is dedicated towards
ammonia production.
Longer chain carbon compounds, such as alkanes, alkenes etc depend on carbon based feed
stocks. Now we can take directly CH4, or we can take CO2 and H2 and use the Fischer-Tropsch
process to create the methane needed as a starting point. In this way, we do not depend on
the extracting of fossil fuels any more. Another option would be to make some biomass undergo
pyrolysis and then gasification to create syngas which is a CO H2 blend which can also create
long chain carbon molecules. Either way, this process can be made renewable but will require
more energy than it does now (and if you want goods at similar prices, the energy will have to be
cheaper than it is now).
Concrete! Did you know concrete is used is almost all infrastructure these days (along with
steel). The main reactions when making concrete is mixing calcium carbonate with heat to produce
Calcium oxide and co2. This calcium oxide is mixed with water to create calcium hydroxide.
Calcium hydroxide is concrete. THe equations for this
CaCO3 + heat −−→ CaO + CO2
CaO + H2 O −−→ Ca(OH)2
Now during this production, you have a concentrated stream of CO2 being produced. This can
be captured and used for creating, for example, alkenes. This is easier than trying to capture CO2

lOMoARcPSD|2686941
Page 10 of 66
Figure 1.3: Energy useage in different transport applications
from the atmosphere because even though we have too much CO2 in the atmosphere, it is only
400ppm. It takes A LOT of air to be sucked in, filtered, the CO2 extracted, to get sufficient CO2
for the reaction, along with a lot of energy.
Now something interesting happens when you leave calcium hydroxide exposed in air. It slowly
reacts with CO2 to form back into Calcium carbonate! This is great because the concrete acts
as a carbon sink over time. This does not mean concrete is carbon neutral, but it is a source of
carbon absorption.
So you might have realised here as well, concrete needs heat to be made. This can again come
from renewable fuels. Currently the heat comes from fossil fuels (which makes even more CO2).
Useful facts: cement production has the 2nd largest share of CO2 emissions from ”industry” at
27%. This means that globally, it represents 6.5% of total energy-related CO2 emissions.
1.3.2 Transport
Now transport. In SET3013 we did a whole week on energy use from transportation of goods and
people. The energy use of this can be seen in Figure 1.3
Now what sort of energy can be used for what sort of transport type? Many options exist,
biofuels from bio mass) are options to all of them. However we don’t sustainability produce enough

lOMoARcPSD|2686941
Page 11 of 66
Table 1.1: Comparison of transportation options with their sustainable energy source
Vehicle type: Energy type:

Light vehicles Electric power with batteries (or hydrogen fuel cells if you are in Japan)
(cars, motor-
bikes etc)
Trucks Electric power with high power batteries. Hydrogen combustion, ammonia
combustion, bio fuel
Trains Electric using overhead wires, big batteries, H2 fuel cells. Renewable fuel
combustion
Airplanes Hydrogen/ammonia combustion. Biofuels also work but don’t eliminate the
other GHG emissions from aviation
Ships Hydrogen/ammonia combustion
biomass (i.e. not 1st generation biomass) for this so alternative methods are also needed. Options
can be seen in Table 1.1.
1.3.3 Demand response
You already know this from SET3060. Demand response is when we shift the demand for electricity
to match the available supply better. If we supply power to an aluminium smelter, we can ask the
smelter to take more power for the next hour while we have a higher generation capacity resulting
from higher wind speeds. In the hour after that when the wind drops, we give them less power.
As long as they get all the power they need in a day to produce aluminium, they’ll be happy. We
are happy. In this way, we can reduce our required installed capacity. Of course, there are limits;
we can’t ask people cooking dinner when the sun is not shining to cook dinner at lunch time when
it is. We also can’t postpone all car production to the summer months when we have lots of sun
and not produce anything in winter. Hence demand response for all industry is not possible. If we
did, we would be produce so many cars during summer that the capacity of the plant we would
need would be significantly larger than what we have now to achieve the same throughput. This
would increase costs massively (and reduce the capacity factor of the plant). IT would not be an
efficient use of resources. However, this is an extreme case; on smaller time shifts (say a week), it
could be doable for some industries. However, the main take away is, we will still need storage.
A common consumer use case for demand response is for charging electric vehicles. The idea
is that consumers can charge their cars when electricity is cheapest, which, coincidentally, is when
the most renewable energy is being produced (throughout the day when the sun is shining and

lOMoARcPSD|2686941
Page 12 of 66
Figure 1.4: The red line refers to the power being produced, the blue dashed line refers to the load
From Henry, S er. a;;., Going Green: Transmission Grids as Enablers of the Transition
to a Low-Carbon European Economy, IEEE Power and Energy Magazine, 12(2), 2014

lOMoARcPSD|2686941
Page 13 of 66
the wind blowing). However making this work is hard because you need some central energy
management system (like a computer) to make sure everything works as it should. However this is
doable and there are now some products coming onto the market which do this for you. Eventually,
smart devices will do it automatically/have the feature built in and it will have a positive financial
benefit for people.
1.3.4 Energy transport
As you remember from your energy model you had to spend hours on for this course; energy has
to be transported from one location to another and you need to do that efficiently. Now there are
a few ways we bulk transport energy, these are
• Power lines (electricity), either AC or DC
• Pipelines (liquid and gas fuels)
• Road/rail/ship (liquid, gas and solid fuels)
And that’s it. Yes you could move batteries around but that isn’t very efficient, nor is wireless
charging. You could beam power to space and back (this is a feasible idea). However, this idea
makes people and governments scared because:
• people think this will microwave their brains
• Governments think this will become a giant energy laser death ray which will impair national
security/sovereignty/freedom/¡insert generic government complaint here¿
Electricity
I will assume you know what are, and what the main differences between AC and DC are as you
covered that in high school physics probably. Now for transmission lines, AC is what we uses in
cities and for transporting energy over short distances, AC is currently cheaper than DC. Over
longer distances, DC becomes cheaper and DC transport results in less losses 1 . A nice graphic of
the cost difference between AC and DC as a function of distance can be seen in Figure 1.5.
1
DC becomes cheaper because the electrons in the wire do not keep changing direction all the time, which not
only creates a magnetic field, but also costs energy

lOMoARcPSD|2686941
Page 14 of 66
Figure 1.5: Cost comparison of AC VS DC transmission over distance with respect to cost
Some of the reasons why there is a cost difference in AC vs DC transmission as shown in

Figure 1.5 are 2 :
• DC has high initial costs because of DC-AC converters (aka inverters), AC-DC converters
(aka rectifiers), filters, DC-DC voltage transformers
• Because of the lack of oscillating of electrons in DC, there is no limit to the length of a DC
3
line. There is a limit for AC lines. The break even distance is around 50-100km
• In AC grids, the electrons oscillate as a certain frequency. If this frequency is not matched
everywhere, then issues occur between out of sync phases. This problem becomes more no-
ticeable the further the distance between different generation facilities. This phase imbalance
occurs because information (and electrons) can only travel as fast as the speed of light4
2
Please tolerate my lack of electrical engineering jargon knowledge
3
If you want to know more, look up ”reactive power”. Basically AC transmission is limited in length because
the longer the line, the more apparent power is going through the line at a given point in time due to the phase
difference between current and voltage in AC. However, this power cannot ”do” anything useful, but it still adds
to the losses in the line and limits what the power through the line can be
4
Well things can travel faster than light but then you need to warp space time, Look up ”Albecurrie Drive”

lOMoARcPSD|2686941
Page 15 of 66
Fuels
When we talk about fuels in this section, we are referring to liquids and gases. Now as mentioned
earlier in the summary, it is possible to go from electricity to fuel. A very common examples
electolyzing water to produce hydrogen and oxygen (details on this process are covered in subsec-
tion 1.4.4. Now you might already know but this was not always an efficient thing to do (older
electrolysis plants had an efficiency of 60%, newer plants around 80%). Now besides the storage
implications, transporting gases of liquids via pipeline is actually very efficient, the professor says
its about 0.5% losses per 1000km. Now reality is more complicated than that number; the losses
will depend on the pressure differential, the viscosity of the fuel, the type of fuel, the phase of the
fuel, the size of the pipeline, the manufacturing method of the pipes etc. If you ever do any sort
of fluid flow in your career, make sure you understand this 5 .
Also note that there are losses going from a fuel to electricity. As you might know, the most
efficient internal combustion engines have an efficiency of around 50%. So going from electricity
to H2 fuel to electricity (via turbines) has an efficiency of 40%. Not great and this is why batteries
are preferred as their discharge and charge efficiency can be about 90% both ways and so a round
trip efficiency of 81%, twice that of H2 (not to mention all the issues with storing, transporting and
using H2 fuel). Now the round trip efficiency can be improved using fuel cells (80% efficiency going
from H2 to electricity) but the overall efficiency (then 64%) could still use a lot of improvement.
1.4 Energy storage methods

This section is all about the name of this course; energy storage. So you might wonder, how do
you characterise the different sizes of different energy storage systems? Well with graphs obviously.
Look at Figure 1.6.
So what does Figure 1.6 tell us? Well it tells us a rough approximation of what different systems
6
exist and their energy capacity as well as the length of time to discharge the system at a given
capacity. Remember rated power is the power a system is designed to provide energy at (although
5
If you want to do a course of flows in pipes, I have heard good things about AE4261 Internal Flows. Be warned,
it is an aerospace course so the work load for the number of ECTS is very high and not what it should be. You
will also need an understanding on thermodynamics and fluid/aerodynamics. If you don’t know what a boundary
layer is, might not be the best course for you
6
Of course you can always have a very small NH3 tank and generator in the 100 kWh range. But the graph is
meant to show large scale practical information

lOMoARcPSD|2686941
Page 16 of 66
Figure 1.6: log-log graph for discharge time and storage capacity. ptL = Power to Liquid. ptG
= Power to Gas. CEAS = Compresses Air Energy Storage. LAES = Liquid Air Energy Stor-
age. SNG = Synthetic Natural Gas. Power to gas/liquid refers to the energy (from electricity)
converted to a gas/liquid (e.g. H2, Ammonia)

lOMoARcPSD|2686941
Page 17 of 66
it can also provide less power depending on the design, like a wind turbine).
This this graph we can see that the current ”standard” of renewable energy, batteries, are quite
limited for long term, large scale energy storage. This is why there is so much focus on chemical
energy storage forms such as synthetic natural gas (SNG), hydrogen and more recently ammonia.
If you think about it, this shouldn’t be that surprising; the current energy storage methods we use
are all fossil fuels, and these are all carbon/hydrocarbon based which is just a chemical energy.
Ok, you might be thinking,
I want to make a storage facility to hold all this energy, and I want it to be 1 MWh in scale. I
want to use batteries because I am think it is the best solution. How should I go about doing this?
So, you need to decide how you are going to scale up. Looking at Figure 1.6, you see that it
is possible to have a battery which is 10 MWh in size. However, you could also have 10 1 MWh
batteries. Which way is correct? This refers to two different ways you can ”scale up: ”scale up”
by creating larger and larger single ”batteries” or scale out, by adding more and more smaller
batteries. This can be graphically shown in Figure 1.7. In practice, this option will come down
to the reduancy you want in your system and the cost of the system. Bigger batteries might
be cheaper but if you have one big 10 MWh battery and it breaks, you storage facility is gone.
However, if one of your more expensive, 1 MWh batteries breaks, you still have 9 MWh of capacity
left. These numbers are of course random but I hope you get the point. Generally speaking, you
can do either ”scale up” or ”scale out” with most energy storage options that are not geologically
constrained such as pumped hydro or CEAS 7 .
1.4.1 Hydroelectric and pumped hydro
The idea of this is simple. If there is a height difference between two bodies of water, you can
run the water from the higher location through a generator to the lower location. The generator
extracts the kinetic energy of the water and turns it into electricity. When you want to store
energy, you pump the water back up from the lower water reservoir to the higher reservoir using
electricity. A simple graphic can be seen in Figure 1.8. The large the height difference (for the
same volume of stored water), the more energy is stored.
7
Compressed Air Energy Storage is currently only done in underground large cavern where you can pressurize
a large volume a small amount. This si because of inefficiencies present when you compresses a gas. However, you
can scale up CEAS by increasing the pressure in a tank for example, however the losses increases quickly unless
you can find a way to prevent the gas heating up while compressing

lOMoARcPSD|2686941
Page 18 of 66
Figure 1.7: Scaling up vs scaling out graphically shown
Now in addition to this, you can also get the upper reservoir filling up automatically. How? No
this is NOT free energy. The upper reservoir is always a lake, and this lake is fed from rivers and
the rivers are filled from the rain area catchment. Therefore the energy for the dam comes from
the sun. Not surprises there. The sun is hot. Next time you go outside and you can see the sun
(i.e. when you are next not in the Netherlands), don’t forget to thank the sun for all the energy
it provides to you. It is the main reason you are able to live on this planet.
Now as I mentioned, the rain is what mostly powers your dam. As such, rain patterns are
important to the fluctuation of energy stored in the dam. If you modeled this in your energy
simulation, you should take taken this seasonal fluctuation into account. Another challenge with
pumped hydro is location, you need a location which can support two large bodies of water with
a significant height difference AND you need this place to not disturb natural wildlife/people
etc. It is very difficult in practice to find such locations. One option the dutch came up with
(which I believe was mentioned in SET3013 as well) is making some dykes in the sea and then
draining the made lake. There are issues with this such as structural rigidity, water tightness, cost,
environmental impact etc. But the idea is there and you can see what I mean by this in Figure 1.9.

lOMoARcPSD|2686941
Page 19 of 66
Figure 1.8: How hydro power works. Note that the ”sea” could also just be another lake
Figure 1.9: Dutch concept of having a giant ”reverse lake” in the sea surrounded by dykes. The
image is in Dutch sorry and I don’t have the time to translate it

lOMoARcPSD|2686941
Page 20 of 66
1.4.2 Biomass
Biomass refers to using food stuff/plants/algae for energy. As you might remember from SET3013,
there are 3 types of biomass:
• 1st generation: Growing food to turn into fuel (such as ethanol). Because of competition
with food supply, considered not ideal and not truly sustainable.
• 2nd generation: Using agricultural waste such as leaves, coconut shells shells, trunks, wood
chippings etc (so the residues of normal agricultural/forestry activities which occur anyway
for food production) for energy.
• 3rd generation: Using algea and other water based plants/organisms as the source of
carbon for bio fuels. Algae are very useful for bio fuels as they can double in mass every 3h
or so.
It goes without saying that different types of biomass have different energy potentials. Now why
bio mass/fuels? Well they can be stored for a long time without losing much energy. Additionally,
biomass can be converted to bio coal/oil which can be stored for long periods of time and have
very similar properties to ”regular” coal/oil and depending on the process employed, can be used
as ”drop-in” replacements for fossil fuels. However, bio-fuels suffer from being more costly than
fossil fuels. Also, because supply is limited to algae and agricultural waste, bio fuels have a ”limit”
to what they can supply, and therefore store, energy wise. Also, as we are dependant on plants
growing (so limited by photosynthesis), available land area for growing, water supply, soil fertility.
There is also an inherent seasonal variation to biomass. If you were to model this, you need to
account for this seasonal variation.
Important note: Because we are dealing with plants, we need to realise the speed at which
plants grow can only be as fast as photosynthesis allows. Photosynthesis is a very inefficient process
for biomass energy. This is because plants are designed to make themselves grow, live and ensure
they are pollinated (to pass on genes). Those goals are different from the goals humans have with
plants. Humans want plants to grow fast, provide calories and nutrients and be easily turned into
fuel. This is why solar panels are significantly more efficient than plants in turning sunlight into
energy. Photosynthesis has an efficiency of 0.2% whereas PV cells are 20%. This is because
plants have evolved with different objectives compared to human designed PV cells. This is one

lOMoARcPSD|2686941
Page 21 of 66
of the biggest limitations to biomass - due to its inherent inefficiency using plant, large amounts
of plants (and therefore land to grow them/get agricultural waste from) is required.
Knowing all this, you might think ”can’t I just use 3rd generation biomass for everything?
It doesn’t take up much land nor competes for food production.” In this you would be right.
However, 3rd generation biomass has limitations. 3rd generation biomass needs lots of CO2 and
fertilizers to have a high production rate and this in practice is hard to achieve. Additionally, the
land required is still high for 3rd gen biomass because you still have to locate this algae farm on
land somewhere. Now we can use vertical farming techniques to reduce the land area required but
we would need an area the size if Portugal to allow for enough land area to produce all the bio
diesel Europe consumes in a year. This is just impractical 8 .
Ok, so you might think, ”why is biomass even considered an option for energy storage?” Well
firstly because it is there; we have available 2nd gen biomass so we might as well use it for
energy. In some countries with high abundance of this resource, it makes sense for them to use
it. Additionally, carbon based fuels (for storage and usage) and very useful due to their chemical
properties, such as jet fuel and steel production.
Side note: If you wanted to store carbon from the atmosphere in the ground, you could take
waste biomass and turn it into bio coal (so coal) and put it back into the ground (in old mines/oil
wells etc). In this way, we actually reduce the carbon levels in out atmosphere. If the cost of
carbon becomes high enough, this might become a profitable thing to do
1.4.3 Batteries
Ok so this section is going to focus on electrochemical energy storage which is reversible, that is,
rechargeable batteries. One time use batteries have more chemical reaction possibilities as the
energy transfer has to occur once in one direction. Rechargeable batteries are harder as the energy
transfer has to go in two directions many times. Over the course of human history, we’ve discovered
only a few types of rechargeable battery that work. These are:
8
I have not mentioned 1st generation biomass. This is intentional as the teacher basically focuses on this in
his slides. I believe he does this because in the USA, most of the biomass energy comes from producing corn to
turn into bio-ethanol. This practice makes corn farmers a lot of money but it is not sustainable as it creates a
dilemma: produce corn for food or produce corn for fuel. The answer to this is usually ”fuel” and that is considered
unsustainable. The teacher has focused a lot on the USA and the energy systems in the USA in this course not
always to the betterment of understanding for students. In the biomass track of SET, 1st generation biomass is
never mentioned as the people involved in biomass energy realise 1st generation biomass is not the path forward

lOMoARcPSD|2686941
Page 22 of 66
• Lead-acid
• Ni-Cd (nickle cadmium)
• Ni-Fe (nickle iron)
• Li-C (Lithium based)
Now most use cases of rechargeable batteries you can think of in every day life involve a portable
device. Think of your phone, laptop, headphones and if you are a rich asf student, electric car.
We can generate a list of requirements for rechargeable batteries as we use them in our everyday
mobile applications because ultimately, we all want more than a days charge on our phone without
out phone weighing 1 kg.
• High energy density (gravametric9 and volumetric)
• High power density (fast recharging)
• High cyclability: 1000 cycles with high fraction of energy use each cycle
• Low cost (cheap battery = more battery)
• Safe (no spontaneous combustion please. Think Samsung)
• Environmentally friendly (we are SET, you know why)
However, if we want large scale grid storage, the requirement list differs. With grid storage we
don’t care as much about low mass as we are not moving our battery pack with us every day. We
also don’t care if our battery pack is x2 bigger because we can put this battery in locations where
land is cheap. We also don’t want to replace this battery every 3 years like we do with our cell
phone, so we want our battery to last for more cycles of charges/discharges.
• Reasonable energy density (gravametric and volumetric)
• Reasonable power density (charging over a day is fine)
• High cyclability: 10000 cycles with high fraction of energy use each cycle
9
gravametric refers to energy per mass unit

lOMoARcPSD|2686941
Page 23 of 66
• Low cost (cheap battery = more battery)
• Safe (no spontaneous combustion please. Think Samsung)
• Environmentally friendly (we are SET, you know why)
Ok so we have our shopping list, what battery is best? Well let’s look at Figure 1.10. We can
see that of the ”standard” rechargeable batteries, Lithium ion and its derivatives are the best.
However we also see that there is something called ”Li metal”. Sadly the information presented
in the lecture doesn’t cover this type of battery, generically known as metal air batteries, in
more detail. TLDR, metal air batteries offer significantly higher energy densities than current
lithium ion batteries. One type of metal air battery which is being developed currently is iron
air batteries. This is basically a battery which runs on rust (which is a reversible electro-chemical
reaction). Some numbers: iron air has an energy density of 1200 Wh/kg where as Lithium polymer
(PLiON) batteries have an energy density of 158 Wh/kg. For comparison Solid state Lithium ion
batteries have an expected energy density of 400 Wh/kg.
So I am going to assume you know how a redox battery works (remember redox reactions from
high school chemistry??). However, if you don’t, I will go over it quickly. Basically a battery
works by a potential difference (i.e. voltage) between 2 different elements/compounds. When an
10
external circuit is connected, electrons can flow through the circuit while an ion can flow from
the annode to the cathode through something called the electrolyte, a substance which facilitates
the transfer of ions. This process can be seen in Figure 1.11. The voltage of a lithium ion cell is
3.5 V. This voltage is around 3 times higher than lead acid of nickle based batteries
So from the above information, let’s delve into the challenges for EV batteries. They are
basically identical to the requirements for portable rechargeable batteries
1. Cost: Cheap materials, cheap manufacturing
2. Energy/power density: Needs to be high because low impacts driving range and charge rate
(”refueling” time)
3. Life time: Need higher number of cycles (higher durability), high depth of discharge (high
use of all stored energy in the battery), long calendar life (in terms of years)
10
for lithium ion batteries, this ion is a lithium ion

lOMoARcPSD|2686941
Page 24 of 66
Figure 1.10: Energy density relation for various rechargeable battery types. PLiON = Lithium
ion polymer
4. Safety and environment: Very polluting or spontaneously combusting batteries would not be
good for TESLA’s image
1.4.4 Artificial fuels
As we’ve seen before, if we have cheap, green hydrogen gas, we can use that hydrogen for energy
11
storage, or convert it to ammonia for higher capacity energy storage, or even e-fuels . So how
11
e-fuels are fuels which take in CO2 from the atmosphere, added with green hydrogen to make synthetic
hydrocarbon fuels. Sounds great but not very practical. To get 1 kg of CO2 from air, you need to process 1508 kg
of air, that is a volume of 1231 cubic meters. This gives you about 22 mols of CO2. Assuming perfect conversion
of CO2 to octane (gasoline) with surplus green hydrogen, then out 1kg of CO2 gives us 0.3135 kg of gasoline. This
is equivalent to 0.445 L of gasoline (treating gasoline as octane for these calculations). We have to draw in 1 231
000 L of air to get 0.445 L of petrol. This already sounds very inefficient and now start realising that it takes a lot
of energy to process more than one million liters of air in a time efficient manner. In the context of a human, the

lOMoARcPSD|2686941
Page 25 of 66
Figure 1.11: Graphic on how a battery works
can we create cheap green hydrogen? Well, electrolysis of water is the best way (we aren’t going
to fall for the blue hydrogen trap here which is just steam reforming of natural gas). There are
few different types of electrolyzers in existence, we’ll cover them now

lOMoARcPSD|2686941
Page 26 of 66
Figure 1.12: Alkaline electrolysis for H2 production. Images taken from ”H2 generation from
alkaline electrolyzer” by Merit Bodner, Astrid Hofer, Viktor Hacker

lOMoARcPSD|2686941
Page 27 of 66
Alkaline Electrolyzers
From Figure 1.12, the process of this process should be clear. We have a basic solution to begin
with and a mebrane which only allows hydroxide ions to pass through separates the positive side
of the electrolyzer (the anode as oxidation occurs) and the negative side of the electrolyzer (place
where reduction occurs). The charectaristics of alkaline electrolysis are as follows:
• Suitable for scales from several MW to 100s of MWs
• Lost cost electrolysis
• Uses nickle based electrodes (relatively cheap) and cheap electrolyte (NaOH, KOH)
• Durable - 10 years of continuous operation fine
• Suitable for scale out
• Issues with intermittency - difficult to switch on and off, so best used for continuous duration
• Efficiency around 65-80%
Polymer Electrolyte Membrane (PEM) Electrolyzers
A simplified diagram of a Polymer electrolyte membrane (PEM) electrolyser as shown in Fig-

ure 1.13. These electrolysers were developed to solve some of the issues which alkaline fuel cells
experience. In PEM elecrolyzers, a hydrogen ion is transferred through the membrane, as opposed
to a hydroxide ion. The membrane separating the cathode and anode is also known as a ”proton
exchange membrane” (hydrogen ions are just protons after all)
Some notes on PEM electrolyzers:
• Platinum and Iridium catalysts on cathode/anode (expensive materials)
• Good on-off capabilities
• Higher cost when compared to alkaline electrolysis (because higher material cost)
• Efficiency around 65-75%

average human breaths in 12 kg of air a day. So given we need to process 1508 kg of air, it would take a person
125 days to process this much air. All for not even a liter of fuel. This hopefully also explains why carbon capture
from the air (i.e. not directly from a emission source) is an energy intensive and inefficient process...rant over

lOMoARcPSD|2686941
Page 28 of 66
Figure 1.13: PEM electrolyzer. Image credit to US Department of Energy, Office of Energy
Efficiency & Renewable Energy

lOMoARcPSD|2686941
Page 29 of 66
Figure 1.14: Solid Oxide Elecrolyzer Cell (SOEC). Image credit Wikimedia
• Lifetime: 10 years
• Flexible with fuel (ammonia can also work for example)
Solid Oxide Electrolyzers
There is another type of electrolyzer, Solid oxide electrolysers. However, it isn’t treated in the
course so feel free to skip this section. I will cover it here though if you want to learn more (and
for completness sake of this summary).
Solid oxide electrolyzers, also known as Solid Oxide Electrolyzer Cell (SOEC), operate at
medium high temperatures (500 to 850 Celsius). The difference in this electrolyser is that an oxide
ion is transferred through the electrolyte, not a hydrogen ion or hydroxide ion. See Figure 1.14
The reaction at the anode and cathode
2 O2 – −−→ O2 + 4 e – anode
2 H2 O + 2 e – −−→ H2 + O2 – cathode
Here are some characteristics of such SOECs

lOMoARcPSD|2686941
Page 30 of 66
• High operating temperatures - long time to start up times, material issues with high tem-
perature and oxidation level compatibility
• Not limited by carnot efficiency, therefore able to achieve very high efficiencies
• No expensive catalysts
• Low operating cost
Battolyzer
Ok so lets do some Frankenstein thinking; what if we added a really robust battery (say Ni-Fe
cell) and the most robust electrolyzer (alkaline) together? Well then we’d get something called the
12
battolyser. Read the footnote .
Figure 1.15: Battolyzer chemistry
Figure 1.15 shows the chemistry of the battolyzer. Firstly, the battery part of the cell is charged.
The produced Fe and NiOOH which are formed are then used as catalysts which electrolize the
12
The battolyser was researched at TU Delft and this research led to a spin off Dutch company, Battolyser BV.
Mr. Mulder is involved in this company and therefore a potential conflict of interest exists. When a conflict of
interest exists, it is generally good practise for any claim made to be verified by an independent source, and until
this is done, any claim should always be taken with a grain of salt

lOMoARcPSD|2686941
Page 31 of 66
water. There are no noble metals required (which were for PEM electrolyzers, although they were
not for alkaline electrolysers, which this technology is based off). The concept for the battolyzer
comes from Edison’s Fe-NI battery which produced hydrogen gas as a bi-product. Therefore this
technology is able to store energy in a battery and produce hydrogen simultaneously.
Adtantages of this device:
• Efficiency: 80-92% (unclear if this is hydrogen generation from energy or energy storage in
the battery or both)
• Robust (can be abused in its operation)
• Can turn on-off quickly
• Economically competitive with other batteries and electrolysers
IF this is true, the battolyzer has great potential. As discussed with the previous electrolyzer
designs, the most expensive one (PEM) is also the only one which can switch on/off easily. Low
lost, rapid response electrolysis makes an energy storage future much more feasible.
Ammonia Energy
In this course, ammonia was brought up as a fuel multiple times. It has a lot of potential. Here
is a pro/con list of using ammonia as energy. It is worth noting that the professor in his past has
worked on ammonia energy but because of the negative reasons, people seemed dismissive of it.
Maybe that will change one day.
Table 1.2: Pros con list of ammonia energy
Pros of ammonia energy Cons of ammonia energy

Can be created from renewable sources using Toxic to humans
renewable electricity
Higher content of hydrogen per unit volume Smells bad (people don’t like)
than liquid hydrogen
Liquid at 10 bar
100 years of industry knowledge working
with the compound
World’s most produced chemical. Transport
technology already developed
Global Warming Potential of 0

lOMoARcPSD|2686941
Page 32 of 66
Ammonia can be created from the synthesis of hydrogen and nitrogen. Nitrogen of course is
readily available in the air.
1.4.5 Flywheels
One of the oldest methods for storing energy is by spinning something really heavy really fast.
You might know this as a flywheel. Flywheels have been in operations since the dawn of the
Industrial Revolution and are still in use 200-150 years later. We covered flywheels extensively in
SET3013 summary so refer to that if you want more details on deriving the equations I am about
to share.
The equation for rotational kinetic energy is given by:
1
E= · I · ω2 (1.1)
2
Where E is energy, I is mass moment of inertia (kg·m2 ), ω is rotational speed (rad/s). However,
we tend to never completely stop a flywheel spinning because starting it up again is a pain and
can be very power intensive (as we have to deal with the high inertia). Therefore we will often
have a minimum angular speed which we don’t let the flywheel drop below. Therefore the energy
we can extract is
1 2 2
Eextract = · I · (ωmax − ωmin ) (1.2)
2
Now what is the mass moment of inertia for a hollow cylinder? Either you can derive it yourself
like Arnot made us do in SET3013 or you can be smarter and look at someone else’s work.
1
I= · m · (b2 − a2 ) (1.3)
2
1
I = · π · ρ · l · (b2 − a2 ) (1.4)
2
Where m is mass, b is the outer radius of the cylinder,a is the inner radius of the cylinder, ρ is
the material density of the cylinder, l is the length of the cylinder.
Now you may be looking at the previous equations and wonder ”how fast can I spin a flywheel?”.
If you think back to SET3013 summary, you might remember me taking about ”stress” , how
different materials have different ”maximum stress” and using a spoon as an example of all this.

lOMoARcPSD|2686941
Page 33 of 66
So this is the answer in practice; the faster we spin a flywheel, the more stress it is put under and
eventually the material can’t take anymore and will the flywheel will break. Now we can use a
combination of different materials and different cross sectional shapes for our flywheel to allow us
to spin faster, but we aren’t really going to go into much detail on that here.
Now the more mechanically minded of you might be wanting to know, ”why don’t we use fly
wheels more than batteries?” Here are the issues with flywheels.
• Fly wheels self discharge due to bearing friction, air drag (even if in a near vacuum), eddy
currents in electrical components
• For mobile applications, flywheels are by their nature heavy. This is not very desirable for
transport applications
Overall, not that bad but flywheels are also less energy dense than batteries. However, flywheels
are really good at rapid charge/discharge in the order of a few kilo-watts in a few seconds (10-100
13
seconds). . The efficiency of fly wheels can be as high as 95% and can have a lifetime of 20 years.
Mostly, due to their limited stored energy, they can be used as grid stabilisation services (taking
up and giving energy
1.4.6 Compressed Air Energy Storage (CAES)
This is another topic we covered in SET3013! So much for learning new and exciting things.
Compressed air storage works the way it implies in the name; it compressed air for later energy
use. Now the big advantage is that the working fluid (air) is very cheap, it costs exactly $0 to
use. However, the infrastructure around the concept isn’t as cheap. Compressors, turbines and
very large closed volumes such as massive storage tanks or caves are not cheap. The equation for
energy stored inside a compressed gas is as follows
Pb
E = Pb · Vb · ln( ) (1.5)
Pa
Where Pb is the pressure of the gas, Vb is the volume of the gas which is compressed and Pa is
the ambient pressure (where the compressed gas can be vented to)
13
Of course, this rapid loading/unloading does impact the lifetime of the flywheel. This leads to material fatigue
and eventually failure. For a relatable (but not quite the identical) example, bend a paperclip back and forth a
small amount many times, eventually it breaks

lOMoARcPSD|2686941
Page 34 of 66
Unfortunately, compressing gas is inherently inefficient. When you compress a gas, due to
the nature of compression the gas you compress will heat up. This heat energy comes from the
compression and therefore lowers the efficiency dramatically. Also when expanding the gas to get
the energy back, the gas cools down which lowers the pressure (therefore potential energy) and
can lead to ice build up.
How do you mitigate these issues? There are two main ways
• Use large volumes and low pressure differences. Compressing a large volume of air by, say
1 bar, compared to compressing a small volume of air by 100 bar, there is significantly less
heat build up with compression and significantly less cooling when expanding.
• Cool down the gas while you compress it and heat up the gas when you expand it. There is
naturally an efficiency to this process but in principal, you can use the same heat generated
with compression as the heat to heat up the gas as it expands, if you have heat storage
For the first option, you need a BIG underground cave. Thankfully geology provides numerous
locations where this is possible. Think old gas fields, salt mines etc. However in saying this, there
are currently only two commercial CAES plants are in operation today as the exact geological
conditions are not everywhere. The second option is basically adiabatic CAES.
Adiabatic CAES
So while adding in worked out question into the summary, I realised I had to add something in on
adiabatic CAES.
An adiabatic process is a thermodynamic process in which there is no heat transfer from the
14
thermodynamic system to the surrounding environment. Entropy may increase
So as you read in the first part of this section on CAES, when you compress a gas, the gas
heats up (we will assume high pressures here). Now if you take this heat from the gas and store
it somewhere, you can use the heat later before expansion of the gas (when, you know, you want
to make electricity). This is useful as the heat from the heat storage can be used to increase the
pressure of the gas before expansion, allowing you to recover the ”wasted” energy you lost to heat
during compression. This can massively increase the performance of CAES plants (because we are
basically removing the main inefficiency of the system).
14
In an earlier version, I said entropy could stay the same, that is incorrect. If the process is also reversible (i.e.
entropy does not increase, the process may be called isentropic

lOMoARcPSD|2686941
Page 35 of 66
So you might wonder, why not just keep the heat in the gas? The issue with this is that a
gas cools down very fast. We haven’t covered radiation in this MSc program (well not the courses
I have done) so I will assume most of you don’t know much about radiation as a form of energy
transfer. If you do, feel free to skip.
The Stefan-boltzmann law says that infra-red radiation is proportional to the temperature of
an object to the 4th order (so T 4 where T is in kelvin). Additionally, the energy lost from this
infra red radiation is linearly proportional to the sensitivity of the object (for air it is 0.8) and
linearly proportional to the surface area of the object (all this integrated with respect to time). In
15
equation form
Z
E= A · ϵ · σ · T 4 · dt (1.6)
time
Now for a gas, it’s surface area is equal to the surface area of all the individual gas particles,
which for calculation purposes, is close to infinity (remember that the energy of a gas is contained
with the vibrations, linear movements and rotations of the particles). Therefore a gas can lose its
temperature very quickly. However, this issue can be minimised if the heat is stored in a solid as
the solid can be better managed from a temperature perspective.
1.4.7 Heat Storage
Think of things you do every day, and think of how much heat those activities produce. In summer,
there is usually too much heat that we don’t want it anymore. But what if we could store this
heat as energy? In principal, there are four main ways to use waste heat energy
• Use as such
• Store as such
• Upgrade
• Convert to electricity
15
A is surface area, ϵ is the emissivity coefficient of the material, σ is the Stefan–Boltzmann constant, T is
temperature in kelvin, t is time

lOMoARcPSD|2686941
Page 36 of 66
Use as such
Use as such: With this we mean take the waste heat and reuse it for something else. If from
a steel plant there there is waste gas which is above 100 degrees, this heat energy could actually
be used for other processes, such as boiling water, molding plastics, synthesising chemicals etc.
However, a challenge with this is getting the waste heat from where it is produced to where it
needs to be used and these distances could be vast (10s to 100s of km) and of course there will
be high heat losses over those sorts of distances if this heat is being transported as a fluid in a
pipeline.
Store as such
What if we just store the heat in a giant ”container” of sorts, so we can use the heat energy later.
If we accept this idea as plausible, what ”containers” can we use?
• Sensible Heat: Energy stored in the vibrational modes of molecules (water, sand, concrete,
molten salts etc). Basically store in a material and let it get hot.
• Latent Heat: Energy stored in Media as it changes phase (solid to liquid, liquid to gas).
Basically, phase change a substance (often water because abundance). When you want energy
back, phase change back
• Thermochemical Energy: Energy stored in chemical bonds of molecules (reversible oxi-

dation, metal to metal oxide). Basically using chemical bonds (formed by heat) which can
later be reversed. If you could electrolyse water by heating it up, then you’d create hydrogen
and this would be perfect.
Sensible Heat For sensible heat, how do you calculate the energy stored? Think back to what
we did in SET3013 and you might remember this equation
∆E = m · CP · ∆T (1.7)
Where ∆E is energy which can be extracted (theoretically), Cp is the specific heat capacity (at
constant pressure) of the substance and ∆T is the temperature difference between the material

lOMoARcPSD|2686941
Page 37 of 66
and ambient. Please note that in practice, a large hunk of concrete will never have a uniform
temperature (because thermal conductivity) so this equation is very much an upper limit. Also
note the material you are using needs to not vaporise easily (because then its latent heat storage).
Concrete and salts are often used. Now if you store this heat underground, you minimise losses
to the surrounding environment. So you can use warm water generated in the summer to heat a
building in winter and cold water made cold by the winter to cool a building in summer. Fun fact,
this is exactly how the library at TU Delft works with its air conditioning system. The efficiency
of a system like this is good if you require heat energy. However if you want to go from this heat
energy to electricity, the methods available are not great efficiency wise (thermoelectric generator,
turbines) are not known for their high efficiency.
Latent Heat Latent heat; remember this is just making a material store energy by undergoing
a phase change. Figure 1.16. Thinking back to high school science class; when a substance, say
water, undergoes a phase change, the energy which was previously increasing the temperature of
the water now is going into breaking the inter molecular bonds between water molecules (which is
keeping the H2O as a liquid).
How might this be advantageous? Before we get to that, I want to bring your attention to
these things, Figure 1.16. This is what is also known as a hand warmer. If you lived in a cold
country, you might know what these do: You ”break” them and they will release heat for a period
of time. Then, you can put them in warm/hot water and they will ”reform” such that you can
reuse them. These products work by a phase change occurring.
So then, how is this advantageous? Well, if we can choose a material whose phase change
temperature is equal to the application of heat we want (say heating a room to 22 Celsius),
then we can use a Phase Change Material (PCM) which has its phase change occurring at that
temperature. This can be very useful for applications where specific heat control is needed (and
we don’t want to pay for expensive thermocouples, heaters and control systems)
Thermochemical energy This is where energy is stored via a reversible chemical reaction. In
general, an equation will have the form
A+B−−⇀
↽−− C + ∆Hr

lOMoARcPSD|2686941
Page 38 of 66
Figure 1.16: Diagram of energy stored in a molecule and the temperature change of said molecule
while a phase change occurs
Figure 1.17: Example of a Phase Change Material (PCM). This was one of the first images I found
on google, no affiliation with this brand

lOMoARcPSD|2686941
Page 39 of 66
Where C is a new compound or a joining of A and B together (packed together). The key thing
about this sort of reaction is that it is fully reversible, unlike the burning of hydrocarbons in air.
One example can be the following:
Ma Xb (NH3 )n + (m-n) NH3 −−⇀
↽−− Ma Xb (NH3 )m + ∆Hr (m > n, n ≤ 0)
Where M is a positively charged ion (usually a metal). X is a negatively charged ion (a halogen
gas ion usually, so F, Cl, Br etc). So this might look scary, but it is saying something quite simple.
NH3 (AMMONIA) is a gas under standard conditions. This reaction though says that ammonia
gas can come out of a solid which just reduces the size of this solid. Further more, this is a
reversible reaction! This is useful because now we have a very convenient reversible reaction with
16
abundant elements as a form of chemical storage .
Now you might want to know more about the physical state of the system during this reaction,
well thankfully we are given an equation called the Clausius-Clapeyron or Van ’t Hoff equation
PN H3 δH ◦ ∆S ◦
ln( )= − (1.8)
P0 R·T R
Where R is the specific gas constant, T is the temperature in kelvin and PN H3 is the pressure
of NH3 and P0 is the ambient pressure. The exact use of this I am not sure, but the lecture puts
a lot of attention on this (uncharacteristically I think) so I think it is good to add.
Upgrade
This wasn’t really covered in the lectures. Good for you! quick section.
Convert
This might be the most logical. If we wave waste heat, we can make electricity with it. This can
be done via the most common method of electrical energy generation
• Turbines
Steam turbines work by using pressurized steam being expanded over a gas turbine (an example
of a gas turbine is an aircraft engine). This wasn’t part of the course so I won’t go into more detail.
If you have the right design, you could also use a Stirling engine to generate electricity. Stirling
engines are actually really cool and this sort of engine (which using the Striling thermodynamic
16
I had no idea this was a thing. This reaction is very interesting to learn about I think

lOMoARcPSD|2686941
Page 40 of 66
cycle) has some advantages over its Brayton cycle or Otto cycle engine cousins (an example of a
17
Brayton cycle engine is a gas turbine, an example of an Otto cycle engine is an ICE car engine)
Another method of creating energy from heat is with thermoelectric generators. These work
off the same principal as thermocouples; a different in temperature between two different metals
causes an electric current to flow. However this is a very inefficient device and not practical for
our scales (they are used on spacecraft though).
1.5 Economics of storage

Now we will focus on money. After all, we need to pay for this energy transition and sadly money
18
doesn’t grow on trees
1.5.1 Transport
Over the last few years, the cost for both solar and wind have fallen dramatically. For offshore
wind, the prices have fallen to a point where subsidy is not longer required to build and operate
wind farms profitably. This is great news!
However, as you know, renewable energies are more dependant on local weather events such
as clouds, wind speeds, rain etc. One way to help solve this intermittentcy is by transport energy
from where there is wind to a place where this isn’t. For this, pipelines and/or power lines can be
used but either way, this will have an inherent cost.
1.5.2 Battery Storage
Energy storage, via batteries of chemical energy, also has a cost. The cost of Li-ion batteries has
fallen a lot over the past 30 years. Quotes costs for different cells by GM and Tesla are shown
below (information for 2017)
• 128 $/kWh for TESLA
• 145 $/kWh for GM (General Motors)

17
If you have time and an interest in thermodynamics, I can recommend reading the Wikipedia pages on different
thermodynamic cycles. I’ll be honest and say I don’t fully ”get” all of them but it’s useful information and
terminology to learn if you are going to be an engineer who works with energy in my opinion
18
Remember money is not really paper anymore. Also, if money did grown on trees and grow like leaves, we’d
have runaway hyperinflation and be screwed in more ways than just the climate

lOMoARcPSD|2686941
Page 41 of 66
1.5.3 Scaling laws
Now if you have read the news before, did anything with economics in high school, did history etc,
you would have heard of the term ”mass production” and ”economies of scale”. What these refer
to is the fact when you make one thing, many many times, you are able to build processes, tools
and supply chains which allows savings on materials, energy usage and fixed costs allowing each
produced product to be cheaper than if the scale was 10 times less. This is the reason why salt
19
costs 0.5-1.5 euros in Albert Heijn rather than 20 euros . This teacher gives us an equation for
estimating the cost reductions with increasing scale. This is very general and each industry has
more specific cost reductions for increasing scale.
cb 0.6
Eb = Ea · ( ) (1.9)
ca
Where: ca is the capacity of equipment a, cb is the capacity of equipment b, Ea is the purchase
cost of equipment a, Eb is the purchase cost of equipment b. How to use: If you know the cost
and size of capacity (i.e. number of machines) at state a, and you know what capacity you want
to get to (state b), then Eb will be the total cost to implement capacity cb .
What does this look like graphically?, well on a ”normal” plot we’d see a logarithmically
increasing line. On a long-long plot, we’d see a straight line increase with a slope of 0.6. This can
be seen in Figure 1.18. These curves are also known as the ”learning curve” (apparently) because
the idea is that the more we scale something up, the better we learn how to do whatever it is we
are doing and therefore the cheaper it will become.
19
don’t believe me? The reason Munich (Germany) go so wealthy was because of the nearby salt mines. Salt
throughout human history has been very expensive because of it’s use in the human body, for it’s ability to preserve
food (in the days before refrigeration) and because it can make things which normally tasty mediocre taste amazing

lOMoARcPSD|2686941
Page 42 of 66
Linear plot of "Scaling Law" Log-Log plot of "Scaling Law"

200 scaling law loglog scaling law
linear
175
150
101
125
log(C1)
100
C1
75
50
25
0
0 25 50 75 100 125 150 175 200 101 102
S1 log(S1)
Figure 1.18: On the left figure you can see a linear plot of the scaling law (solid blue line). The
dashed orange line is the cost is a linear relationship between scale and cost is used. The left plot
is a log-log plot of the scaling law (the sloe is equal to the exponent in the scaling law, so 0.6)
Example
Say you have need an ammonia production facility for 20 GW and you have data for an
ammonia production facility of 2 GW (cost, $ 1 000 000). If you assume a linear increase in
price, the price will be ten times as higher
2 GW = 1 M $
20 GW = 10 M $
If we use the scaling law provided above
Ea = 1 M $
ca = 2 GW
cb
Eb = Ea · ( )0.6
ca
20 0.6
Eb = 1 · ( ) = 3.98 M $
2
Instead of the larger plant costing 10x more, it costs 4x more Written by Andrew Spekreijse

lOMoARcPSD|2686941
Page 43 of 66
You might realise something; this rule can not go on forever as there are material cost limits.
You can’t make a car for a cost less than it costs to buy the material in the first place. So we can
use a modified equation
0.6
(( ccab Mb 0.6
+ 4 · f ccab · ( M a
) )
Eb = Ea · (1.10)
1+4·f
In this equation, the variables have the same meaning as the previous equation. Ma is the
market size in the known case, Mb is the market size in the higher scale case. f is the fraction of
material cost (for the total cost) at scale.
Ok so what does this mean? Well this equation is working on the idea that the total industry
(for raw materials) could expand with a large enough increase in production. The idea is that
if the demand for a raw material increases enough, then new mines will have to open/improved
efficiency in existing mines will occur driving down prices of the raw materials. Of course, if the
industry can’t expand (say that all mining locations of nickle are in use and at capacity), then no
benefits can result from increasing demand for a raw material. This law also then assume no cost
reduction (or increase) will result from the increased demand though.
1.5.4 Energy Market time slots
You will remember from SET3060 that electricity markets are a thing. In that course, we dealt
with the day ahead markets with hourly time slots. However, there are more than just this market
and they serve different purposes:
• Long term contracts/futures market: This market deals with long term bulk energy
supply. Time scale can be weeks or months. Often brought and sold on stock exchanges.
These contacts exist to ensure supply for a user at a set price
• Day ahead markets: This market deals with hourly time-slots where prices and volumes
of electricity are predicted the day before. Used for the whole sale prices. Large volumes of
electricity are traded here.
• Intra-day market: This market deals with power ever 15 minutes inside a given day. This
is to manage the real demand with real supply. This is required from a technical perspective

lOMoARcPSD|2686941
Page 44 of 66
to keep the grid in balance load wise and frequency wise. Small amounts of energy are traded
here.
1.5.5 Costs
There are a few definitions you need to learn which relate to costs. These have come up in exams
in the past although I do not distinctly remember seeing them presented in the slides beyond
presenter notes:
Definition
CAPEX: This is short for CAPital EXpenditures. This is the capital cost to buy the ma-
chine/asset (i.e. the amount of money you paid to someone)
Definition
LCOS: This is short for Levelized Cost Of Storage. The definition of this is CAPEX of
the asset, plus the cost of energy for the lifetime of the asset, divided by the total energy
delivered by the asset. In an equation form,
CAP EX + Ecost
LCOS = (1.11)
Eout
Where LCOS is levelized cost of storage, CAP EX is the capital expenditure of the storage
asset, Ecost is the cost of the energy of the storage asset over it’s lifetime and Eout is the
energy supplied by the storage asset over its lifetime

lOMoARcPSD|2686941
Chapter 2
Exam/worked out questions
2.1 Solar Questions
Question
In the Netherlands the insolation equals about 1100 kWh/m2/yr. Assuming a maximum
insolation of 1000 W/m2, that does that mean for a difference between the maximum power
of a typical solar panel and the average output (assume the panel is horizontal). And for
the average capacity factor?
How to solve this? Firstly, we do not have an area, so everything will be done in terms of
per m2. Say that an average solar panel has an efficiency of 20%. That means an average
solar panel in the Netherlands has MAXIMUM
P∗ = ∇ · I
P ∗ = 0.2 · 1000
W
P ∗ = 200 maximum
m2
45
lOMoARcPSD|2686941
Page 46 of 66
Now we also know the average irradiance. It gives it to us in kWh/yr so actually we have a
unit of power, not energy. So, determining the average power
Eavg, year · ∇ · (3600 · 1000)

P∗ =
tyear
1100 · 0.2 · 3.6e6
P∗ =
365 · 24 · 60 · 60
W
P ∗ = 24 2 average
m
The 3600 · 1000 on the numerator is converting kWh (per meter sq) to J (per meter sq) and
tyear is in seconds as that will result in getting watts.
Now if you recall from SET3013 - Renewable energy (for those of you who just took the
resit, I hope it went well), we know capacity factor is the ratio between actual energy/power
produced over a year divided by the theoretical maximum. For renewable, because of their
indeterminacy, this number if always less than one. For our case, we have,
P ∗actual
Cf =
P ∗max
25
Cf =
200
Cf = 0.125
Question
Describe the systematic variation of the solar isolation throughout the year at the latitude
of the Netherlands.
Generally speaking, the higher the latitude, the longer the day in summer and the shorter
the day in winter. Additionally, the level of solar flux received will be higher in summer
than in winter (more W/m2). Therefore more isolation will be seen around the middle of
the year (June, July, August) than in the ends of the year (December, January, February)

lOMoARcPSD|2686941
Page 47 of 66
2.2 Wind questions

Question
What systematic can be found in wind power throughout the days and year in NL?
Generally speaking, winter months will be more wind and higher wind speeds than in the
a
summer months. Additionally, more wind will occur in the afternoon
a
THis is the offcial answer. In reality, the wind characteristics of a random location will be uniquely
characterised to that location. This is because geography has a very complex relationship with wind speed
2.3 Biomass questions

Question
What are the limits of the production of biomass? With what approximate energy efficiency
is energy produced?
There are 2 parts to this question: Limiting factors and efficiency. Starting with the first
one
The limitations are whatever limits the production of biomass, i.e. plants and algae. Some
limitations are (considering only 2nd and 3rd generation biomas)
• Available land area
• Fresh water supply
• Soil fertility
• Fertilizer availability
• CO2 availability (for 3rd generation biomass/algae specifically)
Next we have efficiency. Because we are dealing with plants, we need to realise the speed at
which plants grow can only be as fast as photosynthesis allows. Photosynthesis is very inef-
ficient for turning sunlight into energy, approximately 0.2% efficient. Efficiencies for turning
the biomass into human useable fuel (via pyrolisis/torrefaction/liquifaction and then via
gasification/gasification+syngas upgrading) will lower this effective efficiency even further

lOMoARcPSD|2686941
Page 48 of 66
Question
What would the concequences for energy storage be if biomass was worldwide abundently
avalible?
The key thing to realise in this question is the word abundantly. If biomass is super
abundant (i.e. we can have as much of it as we want without causing any issues), then we
have sufficient quantity of biomass energy to overcome its inherent inefficiencies. Then we
would be able to create as many bio fuels as we want and because these fuels are inherently
storable, we would never need to worry about storage every again.
Question
large scale CO2 withdrawal from the atmosphere as a result of biomass growth could lead
to negative CO2emissions. Under which circumstances is this the case?
Ok, firstly realise this questions says ”negative emissions”. THis means we are taking in
more CO2 than we are putting back into the atmosphere. Remember, biomass energy is
carbon based and this carbon comes from the atmosphere. Therefore, for negative CO2
emissions, we would have to be producing more bio fuels and we use (i.e. burn). Or, all the
emissions from burnt bio fuels are captured and then stored/used immediately to create for
bio fuels
Question
Can you think of ways living nature performs long term carbon sequestration? How long
term is this?
Not sure why ”living nature” is used in the question but oh well. So how does nature se-
quester carbon? Well buy buying it underground by plants/animals dying and their remains
being forced underground by geological forcesa . As you know from general knowledge, this
stored carbon can remained stored for millions of years.
a
geological forces such as earthquakes, volcano eruptions, glacier forming, floods, tsunamis. All in all,
all the natural disasters the Netherlands doesn’t experience, except for floods. NL has had a lot of floods.
Hopefully never again though

lOMoARcPSD|2686941
Page 49 of 66
2.4 General generation questions
Question
Describe the intermittency of the energy sources; biomass, solar, hydro, geothermal and
wind energy. What is the associated need for storage?
Biomass is using plants for energy. Plants grow in different seasons (usually in spring and
summer). Therefore, biomass has a seasonal dependence and therefore is not constant with
its energy production (3rd generation biomass like algae suffers less from this, however
temperature still is a factor and that often comes from the seasons). The energy output of
biomass is chemical bonds (with carbon) and therefore it is easy to store and the biomass
energy naturally is long term storage.
Solar is intermittent because solar flux changes with season and the sun doesn’t shine at
night. This energy is in the form of heat or electricity. Because of day-night fluctuations,
short term storage is needed. Because of seasonal variance, long term storage is needed.
Wind is intermittent because sometimes the wind doesn’t blow/blow strong enough/blows
too strong. As wind speed is related to flux, there is also a season variance in wind speed.
Because the wind can blow one hour and not the next, there is short term storage needed
and because of the seasonal variance, long term storage is needed.
Hydro has a seasonal dependence depending on the reservoir type. If it is a big lake, then
rainfall has an impact on storage levels, just as much as droughts. However hydro is not
a very intermittent form of energy storage due to the large volumes of water. With large
damns, the type of storage is inherently long term storage. Smaller ”pumped storage” damns
can be considered short term storage.
Geothermal energy is energy which comes from the Earth’s crust. This depends on location
and not season or the what the sun is doing. The energy which can be extracted has an
upper limit but no lower limit, and therefore no storage is required.
Question
Describe the information presented in the image below, in relation to intermittency of the
energy sources: biomass, solar, hydro and wind energy, and the need for energy storage

lOMoARcPSD|2686941
Page 50 of 66
Of So this graph shows a few things:
• The minimum and maximum energy estimates provided by various renewable energy
sources shown by the vertical blue boxes
• The Energy required for various demands
So let’s start with biomass: Biomass has a energy estimates between 50 EJ/yr and 492
EJ/yr. Note that this graph says nothing about indeterminacy. However, we know that
biomass energy production is season dependant and therefore there is indeterminacy. How-
ever, biomass has inherent energy storage built into the energy source and therefore other
forms of energy storage would not be needed.
Solar: This energy source has a large variation in energy estimated. However, all of the
energy estimated can provide enough energy for the world. However, we know solar it
intermittent as sun doesn’t shine at night therefore storage is needed.
Hydro: Hydro has a basically negligible (relatively) variation in energy estimates. It does
not provide enough energy for the wholer world. Hydro is inherently a stored form of energy
therefore no extra energy storage is required.
Wind: Wind has a large variability in the energy it is estimated to be able to provide. it can
just (at the upper estimates) provide enough energy for the world. We also know wind it
intermittent (sometimes there is no wind, although in Delft this is rarely the case), therefore
storage is required for wind energy.

lOMoARcPSD|2686941
Page 51 of 66
2.5 Hydrogen questions
Question
How much volume is required to store 1 TWh (1e9 kWh) of energy in gaseous hydrogen at
1 bar?
Ok so for this you need to know one of the following
• energy density per unit mass and its mass density of H2 at one bar
• Energy density per unit volume H2 at 1 bar
If you were using the first method, you’d need to find the Higher Heating Value of
ammonia and the density of gaseous hydrogen at 1 bar. Typically, these would be the
values you’d find in a table of properties of ammonia. However, the teacher uses the 2nd
way based of the energy density graph using energy per volume method
The value the teacher gets from the energy density graph is 3.5 kWh/m3. Therefore
Estore
V =
ρenergy
1e9
V =
3500
V = 285e6 m3
If you pretend the storage method is putting this ammonia in a box, the box would have
side lengths of 658 m, which is not very realistic. This is the reason why hydrogen storage
a
usually pressurized hydrogen to obtain greater volumetric energy densities
a
Higher volumetric density is obtained from a higher mass density of hydrogen. The HHV of H2 does
not change with increasing pressure

lOMoARcPSD|2686941
Page 52 of 66
2.6 Ammonia questions
Question
A crude estimation of the scaling cost of technologies can be indicated with the equation
S1 0.6
C1 = C0 · ( )
S0
Where Cx is the cost of the technology at scale Sx . Use this equation to estimate the potential
benefit for realising grid scale ammonia storage for all sectors. Assume the following
• Ammonia industry cost 0.05 currently
• Ammonia industry currently uses 1.5% of total energy use. All industry uses 9% for
feedstock and 16% for energy
Compare scaling up just the ammonia industry and scaling up all industries
Alright, so this is not the most clear question in my option. What it wants is for us to
determine how much scaling up the ammonia industry would cost and what scaling out all
industries would cost. Why? Well because if we don’t have good storage, then all industry
would need to scale up to account for a lower capacity factor (if no energy storage, how do
you make cars at night with solar energy?). Ok, so let’s first find the cost to scale up just
the ammonia industry
So currently, ammonia provided 1.5% of all energy use and feed stocks in general use a total
of 9% of energy use. Therefore, the amount of energy (for feed stocks) which is not supplied
by ammonia is 9% − 1.5% = 7.5%. This means we need to scale up the ammonia industry
7.5
(on an energy basis) by 9
= 0.8333 = 83.3%.a
Using the scaling estimation
S1 0.6
C1 = C0 · ( )
S0
83.3 0.6
C1 = 0.05 · ( ) = 0.708
1

lOMoARcPSD|2686941
Page 53 of 66
Now this is not done yet! Before we scaled up ammonia, other industry cost 95% (as ammonia
cost 5% of all industry beforehand). So we need to add this 95% (0.95) on to the new cost
0.708 + 0.95 = 1.658
Ok now for scale out. This is where we scale up all industries equally. Of course in this
situation we aren’t using the full capacity of all industry at all times (as we don’t have
storage). We will assume a capacity factor for industry of 0.2. This means to produce the
1
same amount of stuff, we need, 0.2
= 5 times bigger plants. Now the total cost for all
industry is logically 1 (100%). Therefore applying the scaling law;
S1 0.6
C1 = C0 · ( )
S0
5
C1 = 1 · ( )0.06 = 2.6
1
The cost would be 2.6x higher

a
I do not know why this 83 is then used for how much the ammonia industry scales up. This is what he
does in the slides but his explanation is not the best imo. If you know, please let me know so I can update it
Question
How much volume is required to store 1 TWh (1e9 kWh) of energy in liquid ammonia?
Ok so for this you need to know one of the following
• energy density per unit mass and its mass density of NH3
• energy density per unit volume NH3
If you were using the first method, you’d need to find the Higher Heating Value of
ammonia and the density of liquid ammonia. Typically, these would be the values you’d
find in a table of properties of ammonia. However, the teacher uses the 2nd way based of
a
the energy density graph using energy per volume method
The value the teacher gets from the energy density graph is 4000 kWh/m3. Therefore

lOMoARcPSD|2686941
Page 54 of 66
Estore
V =
ρenergy
1e9
V =
4000
V = 250000 m3
If you pretend the storage method is putting this ammonia in a box, the box would have
side lengths of 63 m, which is very feasible
a
Issue with this method is that for substances like hydrogen (which are gas at most pressures), the mass
density will increase with increasing pressure
2.7 General storage questions

Question
Assume one would need to store 1 TWh in the Netherlands using pumped hydropower. How
much water would the dam dontain and how high would the dikes need to be?
Ok so this is a simple calculation using gravitational potential energy. If you don’t remember
the equations, refer back to the summary for SET3013. If you didn’t do that course, think
back to high school or ask me/someone you know to send you the link for the summary
So this problem has a challenge, namely how do we choose an area and a height? Well the
answers use an area of 35 000 km2 so we will use this too (for reference, the area of the
Netherlands is 41 000 km2).
So the equation we will use and the working is as follows. We will pretend the sea is our
other reservoir and the Netherlands is at sea level (it is below but anyway). Also, remember
that the average water height is what we have to deal with.
E = m · g · ∆h
(h − 0)
E = ρwater · A · hwater · g ·
2
1e12 h
1· = 1000 · 35000e6 · hwater · 9.81 ·
3600 2
h = 4.6 m

lOMoARcPSD|2686941
Page 55 of 66
So our answer is 4.6m. So given most of NL is below sea level, the sea could actually be our
upper reservoir!
Useful information: The Daily energy use of NL is about 2.4 TWh, so the energy in our
reservoir is not even half of the daily energy use in NL.
Question
More and more new houses are built in NL without connection to a gas grid, with solar
panels on the roof and an electricity grid connection; one then aims to realise so called ”zero
on the meter houses” for their energy use throughout the year.
1. Explain what types of energy storage solutions are effectively made use of now in such
cases to enable this
2. What solutions would one have have to choose in a 100% renewable powered future in
2050 to enable such houses
So I find this question unclear. What I believe it is trying to ask is ”currently, how can zero
on the meter houses be a thing and ”in 2050, how can houses be zero on the meter”
I find this odd as what works for now will also work for 2050. Anyway.
For the first question, zero on the meter houses can be achieved by storing the energy
produced by the solar panels in either a battery, or by in storing hot/cold water underground
for use in heating (heat storage).
For the second question, the same as the first applies but now, the house can also send energy
to the grid when surplus energy is made. This surplus energy (probably in summer) can
be turned into artificial fuels and when there is no surplus energy produced, these artificial
fuels (like ammonia) can be converted back into energy by power plants to power the home.
The house would have supplied more energy than what it receives therefore is ”zero on the
meter”
This is really dependant on the house, the house’s energy use and other factors. You can
also design houses which use passive heating in winter and cooling in summer. If you have
so much roof area that you always produce surplus energy over an average week, you might

lOMoARcPSD|2686941
Page 56 of 66
just need a 50 kWh home battery pack for all your energy needs.
Note
A ‘battolyser’ integrates an alkaline Ni-Fe battery and alkaline electrolyser in one device.
What could be an advantage? Could one also see a disadvantage under some market cir-
cumstances?
An advantage of a battloyser is that it’s claimed to be quickly able to turn on/off depending
on surplus energy demand. A disadvantage under current market conditions is that the
technology has a TLR (Technology Readiness Level) of around 5 and therefore not suitable
for commercial use. Additionally, as the technology is derived from a low efficiency battery
type which only produced H2 as a side effect, it is likely the separate efficiency of energy
storage (in the battery) and H2 production are lower separately than other technologies.
In the even cheaper methods of electrolysis eventuate (extreme case, significant increase in
noble metal supply through Luxembourg’s space mining intuitive comes to fruition, or more
realistic case, catalysts with comparable activity as noble metals are fully commercialised
for PEM electrolysis), the economic advantages of the battolyser will be lost.
2.8 Economic related questions
Note
Looking at Denmark in the figure below (it is in Dutch, not going to translate it sorry),
answer the following:
1. What is different?
2. Why not more negative prices?
3. How can one make a business with low prices?

lOMoARcPSD|2686941
Page 57 of 66
1. The biggest difference visible is that Denmark has very low prices significantly more
of the time than other countries. This comes from the large amount of wind energy
in Denmark compared to other European countries. Remember from the energy mar-
ket game from Q1, you should bid your energy at the marginal cost of running that
production facility. Wind energy is (basically) free once you have a turbine.
2. Negative prices come from an oversupply of generation while these is not enough de-
mand. If you know a situation like this will occur, you will shut down the surplus
generation sites to maintain positive prices. You always have the option of just dis-
connecting your solar farms from the grid temporarily. After all, paying someone to
take your good is not a way in which you can make money. However, it is not always
possible to shut down all your plants as the ramping times or technical constraints of
the plant do not allow shut down to occur (hence negative prices).
3. Making a business with low prices is easy or hard depending on who you are. If you
are producing the energy, it is hard because you aren’t making enough to pay for the
costs of producing the energy, maintaining facilities, building new facilities etc. If you
have a large storage facility though, you can buy all this energy at low or negative
prices and then sell it back to the grid at high prices! Perfect way to make money and

lOMoARcPSD|2686941
Page 58 of 66
actually help solve energy issues
Question
For a flywheel used for storage, the target is to have a Levelized Cost Of Storage (LCOS)
under 5 cents/kWh. Explain under what conditions this is possible for a flywheel which
costs 10000 $/kWh. Is this practically feasible?
So first thing to notice is that our LOCS is given in terms of money per unit energy, as is
our CAPital EXpenditure (CAPEX) for the flywheel. So therefore all values we derive will
be in this form.
So our basic equation for this is as follows:
CAP EX · Ecost
LCOS =
Eout
Where LCOS is levelized cost of strage, CAP EX is the capital expenditure of the storage
asset, Ecost is the cost of the energy of the storage asset over it’s lifetime and Eout is the
energy supplied by the storage asset over its lifetime
Now we need to keep LOCS/kWh below 5 ct/kwh. Therefore
1000 + n · EkW h
0.05 >
∇·n
Where EkW h is the energy cost per kWh for one ”cycle” (i.e. charge-discharge) of the flywheel
and n is the number of cycles of the flywheel over it’s entire lifetime. ∇ is the efficiency of
the flywheel. We will assume it to be 80% (even though it is likely closer to 90%). Now lets
go ahead and rearrange this equation such that we separate out two unknown variables
1000 + n · EkW h
0.05 >
0.8 · n
0.05 · (0.8n) > 1000 + n · EkW h
0.05 · (0.8n) − n · EkW h > 1000

1000
0.04 − EkW h >
n

lOMoARcPSD|2686941
Page 59 of 66
Ok so this might look like we’ve achieved nothing of note but let’s have a look at it. From a
practical perspective, we know n will be greater than 1, that is n ≤ 1. Given our requirement
that the LHS of the equation is greater than the RHS of the equation, then we know EkW h
cannot be less than 4 cents/kWh. We have answered the first part of the question.
Now for the second part of the question:
Working with the above and using 1 cent increments, how many cycles are required when
the electricity price is 3 cents per kWh (remember, we have to have a price per kWh of less
than 4 cents):
1000
0.04 − EkW h >
n
1000
0.04 − 0.03 >
n
1000
n>
0.01
n >100000 cycles
So at 3 cents per kWh, the flywheel must go through 100 000 cycles in its lifetime. Given
flywheel systems are very robust and can last 30 years easily, how many cycles do we have
per day?
nlif etime
nday =
tlif e,years · 365
100000
nday = = 9.13 cylesperday
30 · 365
As you should remember on the flywheel section from this summary, flywheels are used
mainly for short term energy supply for grid balancing. Hence they often charge and dis-
charge often per day. Therefore 9/10 cycles per day is well within the realm of feasibility
Question
The electricity market system is organized on small timescales for frequency control, 15
minutes for the off balance market, and the day ahead market operates on 1h timeslots. Long
term contracts are also present (months/years in advance). Indicate on which market the

lOMoARcPSD|2686941
Page 60 of 66
following technologies can provide services and why: Flywheels, batteries, CAES, pumped
hydropower, heat storage, power2gas.
For this question, remember the length of time each storage technology can store energy for.
Also remember how fast can this storage type react to changes in energy demand as well as
how long can energy be stored.
• Flywheels are good for intra day markets (15 minutes) because they provide low energy
volumes but have quick response times and high capacity for rapid cycling making them
ideal for grid balancing
• Batteries are good for intra day and and day ahead markets. Batteries have quick
response times yet can storage energy in large amounts for days at a time which is the
time scale of day ahead
• CAES (Compressed Air Energy Storage) can have a very fast response time and there-
fore is more suitable for intra-day markets and day ahead markets. Depending on
storage location (some huge underground old oil/gas fields in the middle east for ex-
ample), compressed air could hold vast amounts of energy for a long time and therefore
in certain locations, can be useful for long term constructs.
• Pumped hydro can be used for for all market types, depending on the size of the
facility. In the UK, pumped hydro are used as quick response batteries ( 15 minutes)
and in other parts of the world (Colombia), hydro is used to store energy for inter
day trading and store energy for future months as well. Therefore this system can be
suitable for all markets.
• Heat storage on large scale for electricity is not very efficient as going from heat to
electricity is hard where as electricity to heat is easy (blame entropy). So heat storage
can be very good at converting electricity on short notice and therefore is good for
intra day markets. Long term heat energy storage is not really good because the heat
is disappointed exponentially faster with higher temperature (boltz law) so long term
markets is not suitable. However there might be applications in short term markets
(inter day) as temperatures won’t decrease as much over a few days.

lOMoARcPSD|2686941
Page 61 of 66
• power to gas (gas being hydrogen) can respond within hours. So this storage can be
used for all market types. Gas can be stored on a long term basis.
Question
One often hears the remark ‘in a renewable future there will be many moments when elec-
tricity will be available for free’. Comment on that, is it for free, what does that mean for
the implementation of the generation, and would storage influence this?
Currently ”free energy” means negative prices, meaning the generation company is paying
others to buy energy, hurting their profitability and reducing their ability to finance future
renewable energy developments (assuming a socially responsible company here). If renewable
energy is more often for ”free for consumers”, new generations plants will become increasing
uneconomically viable and therefore new plants will be more difficult to finance due to a
decrease in profits. This will lead to potential energy shortages in the future by inherent
under-capacity and then significantly more price variability. Storage can help alleviate this
issue. Excess energy can be stored at pre-agreed, economically advantages prices (future’s
market) which will help stabilise prices, ensure future renewable energy projects remain
economically viable.
2.9 CAES questions

Question
CAES can profit from making it an adiabatic technique. Explain why and how one intends
to realise that
What this question is asking is how can adiabatic compression can make CAES more prof-
itable. Remember that an adiabatic process is a process in which energy is not transfers to
the environment, but kept inside the system a . In reality, this is never possible but for our
case, it means that we have to use all (or almost all) the heat generated from the compres-
sion of the gas without losing any of the heat energy (compression naturally increase the
temperature of the fluid which is being compressed. Sadly there is no way around this)

lOMoARcPSD|2686941
Page 62 of 66
If a process occurs quickly, there is little time for the heat generated to be transferred outside
the system (say the system is the entire CAES plant). To then later use the ”generated”
heat, we can store the generated heat in a heat sink. Then, when we expand the gas (to
produce electricity), we can heat up the gas before it is expanded and use the temperature
increase to increase the gas pressure even more (ideal gas law), getting the energy we ”lost”
during compression back.
a
In an earlier version, I incorrectly said that entropy does not increase. This is wrong. Entropy may
change during an adiabatic process (such as a chemical reaction). If entropy does not increase and the
process is adiabatic, then the process is called isentropic. If you don’t know what entropy is, I suggest you
look it up; it is a fundamental thermodynamic concept so videos on it will be easy to find
Question
A cost level for CAES that is quoted is 400 – 700 $/kWhe installed capacity. Indicate how
such cost level compares to other solutions that can operate in the same market. What can
you tell about the cycle life of CAES?
Such costs are higher than those of batteries for static storage, and much higher than for
pumped hydro. The cycle life of CAES will be very large since it is essentially mechanical
storage. That will mean that the investment costs are divided over many many cycles which
makes the costs levelised over the operational lifetime lower.
Question
Can CAES be applied everywhere or are there specific conditions required?
IN theory yes; CAES can be applied anywhere you have a large tank which can hold high
pressures. However, practically CAES requires MASSIVE volumes which nature provides as
caves. Whats more, these cases need to be pressure tight and have suitable geology which
will not be brittle when exposed to high pressures. These geological conditions are not
present everywhere in the world

lOMoARcPSD|2686941
Page 63 of 66
2.10 Heat storage questions

Question
In a renewable future a steel company will operate it’s blast furnaces with hydrogen, electric
heating, and it will use heat storage and exchange in so called ‘Cooper Stoves’. Explain
these two choices of the steel manufacturer.
The reduction of iron ore to iron requires high temperatures. Hydrogen can serve this
purpose. However do note that carbon will still be required to create steel because of the
material characteristics of steel (this carbon can come from bio-char) a . The ”Copper stoves”
and electric heaters can be used to pre-heat the hydrogen and air before combustion, thereby
increasing the flame temperature of the combustion reaction. The high temperature exhaust
gasses can be used to heat the these copper stoves. Precise temperature control is required
when producing steel, the amount of energy the stoves provide to the incoming flow can
help smooth out any intermitancy The stoves can help reduce any intermitancy in the power
available for the electric heaters.
a
The official answer misses this. Steel just includes carbon and therefore it is needed
2.11 Demand questions

Question
Demand response is considered an important manner to acco,,odate varying renewables,
that can also limit the necessity of energy storage. Shortly explain with two examples where
demand response can play such a role, and two examples where demand response will be
insuffcient.
Ok, so we know demand response can only be a solution for when the time shift for the
demand is small.
One case where demand response can work is the charging of electric vehicles; if the vehicle
is not being used much over the day, the car can be charged when electricity being produced
is the cleanest. Another case where demand response can be used is with a dishwasher; if
the dishes on;y have to be done by the evening, then the dish washer can be turned on when
there is more renewable power being produced.

lOMoARcPSD|2686941
Page 64 of 66
One case where demand response can’t be used is in the production of cars. Because of the
significantly large increase in plant capacities, all car production cannot be postponed to
summer because the cost of the plant sitting around in winter would be unreasonably high.
Another example of demand repsonse being insufficient is the washing of clothes, not every
person has enough clean clothes to last them half a year. a .
a
If you do have enough clean clothes the demand response on your washing is viable, then I believe you
have too many clothes and you aren’t actually helping anyone owning that many. But that’s my opinion
Question
In a renewable future powered by wind and solar, industry will still need power. Answer the
following:
1. Indicate what may happen with the capacity factor of industries that rely on renewable
energy when a storage and conversion of electricity ot fuel is absent
2. What could be a benifit of realising a grid scale energy storage and hydrogen production
for fuel for the goods industryprocuses?
Firstly, i charged the wording of the question in the slides as it did not read well.
For 1:
The capacity factor would decrease. In the absence of stored energy, industry can only
produce while there is energy,. that is, during the day while the wind is blowing and the sun
is shining. Assuming the same number of goods have to be produced, then the capacity of
an industry plant has to be increased to allow for all the goods to be produced in a shorter
period of time. Now the theoretical total output of the industry has increased (by having a
bigger factory), yet the amount of items produced is the same, meaning the capacity factor
is lower.
For 2:
If there is a gird scale energy storage method, the issue of indeterminacy for industry is
removed. This means the same sized plants can produce the same amount of good, keeping
capacity factors of industry the same.

lOMoARcPSD|2686941
Page 65 of 66
2.12 Energy transport questions

Question
Explain what the role of long distance transport of electricity will be in a renewable future
Long distance transport can smooth the energy production over a large region, meaning
energy security becomes less dependant on local weather. Low production in one location
can be compensated by another location
Question
What ways, other than DC electricity, can energy be transported over long, intercontinental
distances? What efficiencies do these methods have?
Remember, AC transmission has limitations on distance. Therefore it is not an option for

intercontinental distances. The only other options would be pipelines, tankers and rail.
Assuming we are working with only a liquid fuel (say ammonia because liquid hydrogen is
horrible to work with), then we need to create hydrogen gas from water, this process is about
80% efficient. Then we need to convert the hydrogen (and nitrogen) to create ammonia. The
Haber-Bosch process has an efficiency of about 50%. Ammonia combustion for a power plant
might also be 50% efficient. For the creating and usage of ammonia is about 20% efficient.
Transport is efficient though at 0.5% per 1000km in a pipeline and rail and ship transport is
comparable (remember how efficient rail and ship freight was back from week 2 of SET3013?)
Question
Why is DC electricity transport promoted for long distance grid connections? Include in
your answer the phase of an AC network
In general, DC transmission becomes cheaper over a long enough distance. However, AC

has a limitation owning to the fact AC has a phase.
AC electricity in most countries oscillates at 50Hz; an electron changes direction ever 0.02
seconds.
The speed of light is about 300 000 km/s. Electrons transmit information at about this speed,
therefore the time taken for an electron to know to change direction over a distance of, say
3000km
3000 km, is: 300000km/s
= 0.01 seconds. Remember that an electron’s direction is changed

lOMoARcPSD|2686941
Page 66 of 66
by a reverse in the potential between 2 points and this reverse in potential is communicated
by electromagnetic waves.
Therefore, we can see that the time taken for the electron to know it’s direction (phase) has
changed is about half of the time taken the electron has to change direction anyway to stay
in the same phase (50Hz) as the electricity in the grid. This ratio is too high and causes the
electrons to be out of phase and have issues with providing energy.

Set3080 Summary of All Lectures

Uploaded by

Copyright:

Available Formats

Set3080 Summary of All Lectures

Uploaded by

Document Information

Original Description:

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

Set3080 Summary of All Lectures

Uploaded by

Copyright:

Available Formats

lOMoARcPSD|2686941

SET3080 Summary of all lectures

The Necessity of Storage Technology (Technische Universiteit Delft)

Studeersnel wordt niet gesponsord of ondersteund door een hogeschool of universiteit

The Necessity of Energy Storage

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

About this summary

• Summary of all covered material, separated by topic

2 Exam/worked out questions 45

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

1.1 Why Storage

1.2 Energy Production

Figure 1.1: Variation in average solar irridance in Europe

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

1.3 Energy use

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Figure 1.2: Share and breakdown of heat demand in industry

• Electric heaters (joule heating)

• Solar heat, geothermal

• Burning a renewable fuel (bio-char, bio-oil, hydrogen, ammonia)

• Using heat pumps

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Figure 1.3: Energy useage in different transport applications

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Vehicle type: Energy type:

1.3.3 Demand response

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

1.3.4 Energy transport

• Power lines (electricity), either AC or DC

• Pipelines (liquid and gas fuels)

• Road/rail/ship (liquid, gas and solid fuels)

• people think this will microwave their brains

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Some of the reasons why there is a cost difference in AC vs DC transmission as shown in

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

1.4 Energy storage methods

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

1.4.1 Hydroelectric and pumped hydro

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)

Figure 1.7: Scaling up vs scaling out graphically shown

Written by Andrew Spekreijse

Gedownload door Floor Borstlap (floorborstlap@kpnmail.nl)