Energy Storage Technology Comparison

(1)

Bachelor of Science Thesis

KTH School of Industrial Engineering and Management Energy Technology EGI-2016

SE-100 44 STOCKHOLM

Energy Storage Technology Comparison

From a Swedish perspective

(2)

1

Bachelor of Science Thesis EGI-2016

Energy Storage Technology Comparison From a Swedish perspective

Felix Söderström

Approved Examiner

Viktoria Martin

Supervisor

Justin Chiu

Saman Nimali Gunasekara

A

BSTRACT

Due to increased usage of renewable energy sources a need to store energy, from times of low demand or high production to times of higher demand or lower production, have risen. This report is meant to serve as a comparison between different methods of energy storage from a Swedish point of view. Several technical aspects as well as environmental and social impacts of different energy storage methods have been compared.

The conclusion reached is that PHES is still the most favourable way of storing energy due to the good performance and reliability it offers. If found possible, Sweden should therefore continuously expand the usage of PHES as well as continuing to improve the turbine efficiency. If further expansion of PHES is not possible, CAES could serve as a replacement due to similar performance.

(3)

2

S

AMMANFATTNING

På grund av en ökad användning av förnyelsebara energikällor har även behovet av att kunna lagra energi från tillfällen då mycket energi genereras eller efterfrågan är låg, för att sedan kunna använda energin då efterfrågan är högre, ökat markant. Den här rapporten är menad att jämföra olika metoder av energilagring ur ett svenskt perspektiv. Flera tekniska aspekter samt miljömässiga och sociala påverkningar hos flera energilagringsmetoder har jämförts.

Slutsatsen som nåtts är att PHES ännu är den mest gynnsamma metoden att lagra energi baserat på dess goda prestationer samt dess pålitlighet. I den mån det är möjligt bör Sverige därför fortsatt försöka expander PHES samt fortsatt arbeta med att förbättra turbineffektivitet. Om vidare expansion ej är möjligt längre är möjligt kan CAES användas för långvarig energilagring på grund av dess liknande egenskaper.

För kortvarigare lagring är Li-Ion eller Na-S de mest gångbara alternativen. God effektivitet och hög energidensitet samt låg kostnad är alla åtråvärda egenskaper. I de flesta av dessa aspekter presterar Li-Ion batterier bättre än Na-S. Li-Ion bör därför vara det primära sättet att lagra energi kortvarigt i Sverige, trots dess något större miljöpåverkan.

N

OMENCLATURE

AC – Alternating Current

CAES – Compressed Air Energy System DC – Direct Current

FES – Flywheel Energy Storage LHS – Latent Heat Storage Li-Ion – Lithium-Ion Na-S – Sodium-Sulphur

PHES – Pumped Hydro Energy Storage SHS – Sensible Heat Storage

SMES – Superconducting Magnetic Energy Storage

(4)

3

T

ABLE OF

C

ONTENTS

1 INTRODUCTION ... 4

1.1 PURPOSE &DELIMITATIONS ... 4

1.2 METHODOLOGY ... 4

2 ENERGY STORAGE METHODS ... 5

2.1 MECHANICAL ... 5

2.1.1 Compressed Air Energy Storage (CAES) ... 5

2.1.2 Flywheel Energy Storage (FES) ... 6

2.1.3 Pumped Hydro Energy Storage (PHES) ... 8

2.2 ELECTRICAL ... 9

2.2.1 Superconducting Magnetic Energy Storage (SMES) ... 9

2.3 ELECTROCHEMICAL ... 10

2.3.1 Supercapacitors ... 10

2.3.2 Battery Storage Technologies ... 11

2.4 CHEMICAL ... 17

2.4.1 Power-to-Gas ... 17

2.5 THERMAL ... 18

2.5.1 Sensible Heat Storage (SHS) ... 18

2.5.2 Latent Heat Storage (LHS) ... 19

2.5.3 Thermo-Chemical Storage (TCS) ... 20 3 COMPARISON ... 22 3.1 DISCUSSION ... 24 3.1.1 Technical aspects ... 24 3.1.2 Environmental impact ... 25 3.1.3 Social impact ... 25 3.1.4 Technology maturity ... 27

3.1.5 Need and availability ... 28

3.2 CASE STUDY –WIND FARM AT BIOTESTSJÖN ... 29

3.2.1 Approach ... 30

3.2.2 Case study conclusion ... 31

4 CONCLUSION ... 32

4.1 FUTURE WORK ... 32

5 ACKNOWLEDGEMENTS ... 33

(5)

4

1 I

NTRODUCTION

Renewable sources of energy are becoming responsible for a larger share of electricity produced all over the world. Since it is not possible to regulate when and how much electricity that is generated from sources such as solar and wind power, a way to store the excess energy from times of lower demand or higher production is needed. During the last century several methods have been developed, ranging from the enormous water reservoirs of the pumped hydro energy storage (PHES) to the modern and theoretically optimal superconducting magnetic energy storage. Since applications and conditions vary between techniques, a comparison is necessary to evaluate what type of energy storage is needed. Comparisons have been done before, the intention with this bachelor of science thesis report however, is to have evaluated the results from a “Swedish point of view”.

1.1 P

URPOSE

&

D

ELIMITATIONS

This thesis has focused on energy storage from a “Swedish point of view”. General properties such as life time, efficiency, capacity, power, energy density and response time is regarded, as are costs, environmental and sustainability aspects, social effects and geographical requirements. The targeted technologies are those that are ready for the market today or in the near future. The aim for this paper is to end in a comprehensive comparison containing necessary information and a recommendation for construction of future energy storages in Sweden.

1.2 M

ETHODOLOGY

(6)

5

2 E

NERGY

S

TORAGE

M

ETHODS

The energy storage methods have been categorized according to their principle of storage; mechanical, electrical, electrochemical and thermal.

2.1 M

ECHANICAL

The most basic way of storing energy is by doing so mechanically. This means converting electrical energy to either potential or kinetic energy via a motor and sometimes also a pump. An electricity generator is later used to revert the process. Mechanical methods discussed are compressed air energy storage, flywheel energy storage and pumped hydro energy storage.

2.1.1 Compressed Air Energy Storage (CAES)

Invented in Germany in 1949, CAES is a technique based on the principle of conventional gas turbine generation. As seen in Figure 1, a motor uses excess energy to pump air is pumped into a container. The air is stored, potentially for months, in order to later be released when the energy is needed. The technique utilizes a combination of the compressed airs potential energy and fuel that is added to the air and ignited as it passes through the turbine [1]. Though using man-made storage containers for the CAES is possible, such containers are more often found in smaller scale projects. The most commonly used method is making use of underground caverns for the container. If a suitable storage is found, for example a depleted oil or gas field, salt caverns or aquifers the compressed air can be stored for up to a year [2].

Figure 1 Compressed Air Energy Storage; 1-motor, 2-compressor, 3-storage container, 4-turbine, 5-generator. Redrawn from [3].

(7)

6

This resulting in environmentally-unfriendly emissions, particularly if the fuel used is fossil-based.

The excess heat from the fumes can be used to reduce losses, but such type of CAES is relatively new and yet not used widely.

As can be seen below in Table 1, CAES potentially offers a long-term storage for a low cost.

Table 1 Compressed Air Energy System properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ie nc y [% ] R esponse tim e Life S pa n C

ost Environmental impact

30 - 350 [2] Gr ows w it h st ora ge unit Up to a ye ar [2]

Varies greatly with pressure, ranging from 0.5 kWh/m3 [4] to 12 kWh/m3 [5] 70 [2] , [5] Min ute s [4] 20 - 40 ye ars [2] , [4 ] 19 – 1 ,325 S EK /kW h [6] 1

Exhausts from fuel impacts the environment to some

extent

2.1.2 Flywheel Energy Storage (FES)

Displayed in Figure 2, FES uses a mass rotating around an axis in order to store kinetic energy. The faster the mass rotates, the more the energy can be stored by the flywheel. In order to minimize losses and increase the efficiency, magnetic bearings are used and the chamber in which the flywheel spins is often filled with helium or drained of air completely to reduce resistances [4].

1_{The price specified in the report was 2-140 €/kWh. The report was received 31 July 2015, therefore the euro to}

(8)

7

Figure 2 Flywheel Energy Storage

Flywheels can be set in motion as well as release their energy in short periods of time. Once in motion, stored energy will be lost at a rate of at least 20 % of the stored capacity per hour [3] of not released. This makes them ideal for peak shaving but not that suitable for long-term storage. As seen in Table 2 below, operating costs are still high for FES-systems.

Table 2 Flywheel Energy Storage properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ienc y [% ] R esponse tim e Life S pa n C

(9)

8

2.1.3 Pumped Hydro Energy Storage (PHES)

First used in Schaffhausen, Switzerland, around 1904, PHES is the most popular technique for storing energy today [7]. When energy is available water is pumped from a lower reservoir to one at a higher position, thus increasing the water’s potential energy. When the energy is needed the water is once again released from its higher elevation to generate electricity.

This technique can store large amounts of energy, but it also needs to occupy a large area. Ideally, two natural lakes can be used as reservoirs. More realistically, a natural lake can be used as the lower reservoir and a manmade structure can replace the upper. Both reservoirs can in need be manmade, though it can increase the cost dramatically.

As can be seen in Table 3 below, PHES offers short response time and a long life span, though the storage density is relatively low compared with other technologies.

Table 3 Pumped Hydro Energy Storage properties

< 3000 [7] Gr ows w it h st ora ge unit - 0.35 - 1.12 kWh/m3_[4] 65 - 85 [5] , [7 ] S ec ond [4] 50 100 ye ars 95 662 S EK /kWh [6] 3

May be harmful for local nature and wildlife due to

size

(10)

9

2.2 E

LECTRICAL

Storing pure electricity is challenging due to the heavy losses that are involved in doing so. The classical example of electrical storage is the capacitor, but as it do not have the capacity or the storage time to be useful large-scale, it have not been included in this report. A variant of the technology, called supercapacitors or sometimes ultracapacitors, is detailed under Electrochemical further down in section 2.3.

2.2.1 Superconducting Magnetic Energy Storage (SMES)

SMES is the theoretically optimal way to store energy. The technology uses a coil made of superconducting material cooled to extremely low temperatures. Electricity that is fed to the coil will remain stored without losses, provided that the coil is kept cold enough [7]. Superconducting materials change their physical properties as they are cooled below its so called transition temperature. When kept at sufficiently cold conditions, the material has zero electrical resistance resulting in zero losses. A simple sketch of a SMES can be seen in Figure 3, showing the three vital components; the cryogenic refrigeration unit that cools the coil, in which the electricity is stored, and the AC/DC-converter. Losses do occur during the conversion from alternating current (AC) to direct current (DC), which the SMES stores, and reverse, resulting in about 90 % efficiency in the end [4].

(11)

10

As the energy is stored simply as electricity, charging and discharging can be done within a moment’s notice, a desirable trait for a technology used to regulate the power grid. However, as the size of the coil has to increase with the needs for higher power and larger quantities of energy to be stored, issues occur when the capacity exceeds 10 MW [4]. For instance, a larger coil results in greater heat generation and thus decreased efficiency.

The techniques required is still expensive, resulting in a high cost as can be seen in Table 4.

Table 4 Superconducting Magnetic Energy Storage properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ienc y [% ] R esponse tim e Life S pa n C ost Environmental impact 10 [7] 0.010 0.030 [7] Up to a da y [7] _{40 - 60 Wh/kg [4]} 0.2 - 2.5 kWh/m3 [6] 90 % [7] Mil li se conds [7] 30 50000 c yc les [4] 596 ,119 – 7 ,096 ,650 S EK /kWh [6] 4 Light environmental impact

2.3 E

LECTROCHEMICAL

Utilizing a combination of the electrical principles and chemical reactions, electrochemical energy storages most notably includes different types of batteries.

2.3.1 Supercapacitors

Supercapacitors are relatively new in the market, but offer great potential due to their high energy density and long life span [4]. The leakage of stored energy while at charged condition, which is a common limitation, will however reduce their long term storage life. They differentiate from regular electrical capacitors by using a fluid such as sulphuric acid or potassium as electrolyte instead of having air or a dielectric material between the plates of the capacitor [7], Figure 4 shows the electrolyte between the two charged electrodes. As can be seen in Table 5, the usage of sulphuric acid may lead to some chemical disposal issues, while

(12)

11

the extremely short response time is a highlight of the technology. A basic sketch of a supercapacitors structure can be seen in Figure 4.

Figure 4 Supercapacitor; Schematic. From [7].

Table 5 Supercapacitor properties

0.001 5 [7] 0.001 0.01 [7] - Up to 20 Wh/kg [7] ₉₅ [7] Mil li se conds [7] 10 4 - 10 6 c yc les [7] 45 ,394 S EK /kW h [7]

5 Chemical disposal issues,

contains sulphuric acid

2.3.2 Battery Storage Technologies

Batteries are most commonly divided into two main categories; primary cells (which can only be used once) and secondary cells, the type that can be recharged several times. The secondary cells are therefore the most interesting type for the purpose of uninterrupted power supply (UPS). Secondary cells can further be divided into standard secondary cells and flow batteries. Standard secondary cells are the more commonly seen type of cell among the two, being popular

5_{The price specified in the book was 7000 $/kWh. The book was available online in March 28, therefore the}

(13)

12

in cars, computers and similar applications. They contain no mechanical parts and all required reactants are held within the battery [7]. Flow batteries are further explained in section 2.3.2.5, Flow Batteries.

2.3.2.1 Lithium-Ion Batteries (Li-Ion)

Due to their relatively high energy density and long life time, Li-Ion batteries have risen in popularity over the recent year, above all in consumer electronic devises. Li-Ion batteries can be made out of many different types of compositions for the ion incorporated, and thus, varying in cost and function as shown in Table 6.

Table 6 Comparison of Lithium-Ion battery compositions [3]

Lithium- Cathode Anode Electrolysis Energy density [Wh/kg] Number of cycles SEK/kWh 2014 iron phosphate (LIP)

LIP graphite Lithium

carbonate 85-105 200 – 2,000 3,850 - 5,950 manganese oxide (LMO)

LMO graphite Lithium

carbonate 140-180 800 – 2,000 3,150 - 4,900 titanium dioxide (LTO)

LMO LTO Lithium

polymer 80-95 2000 – 250,000 6,300 - 15,400 Cobalt oxide (LCO)

LCO graphite Lithium

carbonate 140-200 300 - 800 1,750 - 3,500 nickel-cobalt-aluminium (NCA)

NCA graphite Lithium

carbonate 120-160 800 – 5,000 1,680 - 2,660 Nickel- Manganese-Cobalt (NMC) NMC Graphite, silicon Lithium carbonate 120-140 800 – 2,000 3,850 - 5,250

(14)

13

Note that the characteristics of Li-Ion as seen in Table 7 are derived from the different kind of Li-Ion compositions, listed in Table 6.

Table 7 Lithium-Ion Batteries properties

0.001 0.1 [4] - Da ys [7] 80 - 200 Wh/kg [5] [7] 200 - 500 kWh/m3 [6] 80 -95 [4] - 200 250 000 c yc les [4] 1 ,680 [3] 18 ,886 S EK /kWh [6] 6

Chemical disposal issues, contains lithium

2.3.2.2 Lead-Acid Batteries

Lead-acid batteries were the first type of rechargeable batteries to be developed, based on the reaction between lead-oxide and sulphuric acid. Their efficiency, life time, response time and number of discharge cycles (summarized in Table 8) vary with application, number of used cycles and temperature [7]. This tend to give lead-acid batteries a rather short life time, 1200-2700 cycles or 5-17 years of operation [4], [3] as can be seen in Table 8.

They are however cheap and easily recycled [7]. Due to their high efficiency (of up to 90 % as shown in Table 8), low requirement of maintenance and low energy leakage, they make good long-term storage [3].

(15)

14

Table 8 Lead-Acid Batteries properties

- - - 30 Wh/kg, 180 W/kg [3] 50 - 80 kWh/m3 [6] 75 90 [3] , [7 ] - 5 -17 ye ars, 1 ,200 2 ,7 00 c yc les [4] , [3] 944 1 ,700 S EK /kWh [6 ] 7

Chemical disposal issues, contains lead

2.3.2.3 Sodium-Sulphur Batteries (Na-S)

In sodium-sulphur batteries, sodium and sulphur are heated to around 300 °C. At this point, the sodium melts and is used as the electrodes of the battery, as can be seen in Figure 5. Once the process is running, the heat from the reaction is enough to keep the temperature up. Modern cells are mostly reliable but the high temperature of the battery makes ignition a possible risk. However, sodium-sulphur batteries have the advantages of being made out of inexpensive and non-toxic materials and being highly efficient [3], and having a very fast response time, as can be seen in Table 9.

(16)

15

Figure 5 Sodium-Sulphur Battery; 1-Sulphur electrode, 2-Solid beta alumina ceramic electrolyte, 3-Sodium electrode. Redrawn from [3].

Table 9 Sodium-Sulphur Batteries properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ienc y [%] R esponse tim e Life S pa n C

0.05 [7] 0.4 [7] _- 60 Wh/kg [4] 156 - 255 kWh/m3 [6] ₈₀ 85 [4] Mil li se conds [7] 15 ye ars or 4 ,500 c yc les [4 ] 4 ,800 S EK /kWh [4] Light environmental impact 2.3.2.4 Nickel-based Batteries

(17)

16

have the advantage of a very low discharge rate and therefore also a rather long storage period, as can be seen in Table 10.

Table 10 Nickel-based Batteries properties

0.5 50 [4] , [7 ] - Mont hs [7] 50 - 80 Wh/kg [3] 60 - 150 kWh/m3 [6] ₇0 - 90 [4] , [7 ] - 10 -20 ye ars, < 300 c y cles [7] , [3] 4 ,249 16 ,997 S EK /kWh [6] 8

Chemical disposal issues, for the types containing cadmium

2.3.2.5 Flow Batteries

Flow batteries differentiate from other battery technologies in term of construction. The cell of the battery does not carry the reactants internally, but instead they are stored in external reservoirs and pumped through the cell when needed. The advantage of this is that the reservoirs can be increased in size, and by that capacity, at a relatively low cost [7]. As summarized in Table 11, their life time and response time are satisfactory and neither do they suffer from self-discharge [4], [5].

(18)

17

Table 11 Flow Batteries properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ienc y [% ] R esponse tim e Life S pa n C ost Environmental impact 0.05 15 [4] , [5 ] 120 [5] 16 - 33 kWh/m 3 [6] 75 [5] 85 [4 ] Mil li se conds [4] - 1 ,039 - 9 ,443 S EK /kWh [6] 9 Chemical disposal issues

2.4 C

HEMICAL

Storing energy chemically is another rather new method with great potential, most notably including the so called Power-to-Gas method. Issues regarding production and storage of the gases still pose issues for commercial usage.

2.4.1 Power-to-Gas

The technique of using excess electricity to produce gases such as hydrogen or methane is generally called power-to-gas. Hydrogen can, for example, be produced by the electrolysis of water. Hydrogen is attractive as an energy storage medium because it is cheap and has a decent energy storage density, as can be seen in Table 12. Hydrogen can be transported used as the fuel in either thermal power plants or fuel cells. Either way the rest product is only water [7] and the environmental impact is therefore very small.

Table 12 Power-to-Gas properties

(19)

18 1 100 [9] Gr ows w it h st ora ge unit Da ys [3] 2.7 - 160 kWh/m3 at 1 - 700 bar [6] 62 82 [4] S ec min [4] - 19 142 S EK /kWh [6] 10 Light environmental impact

2.5 T

HERMAL

Thermal energy storage (TES) stores either heat or cold to use at a later time. Heat from industries can be stored during summer to provide warmth during winter, cold can be saved winter time by e.g. hockey rinks to cool during the summer. Alternatively the heat from solar power plants can also be converted to electricity. TES can be further divided into sensible heat storage, latent heat storage and thermochemical storage [10].

2.5.1 Sensible Heat Storage (SHS)

With SHS, energy is stored by heating a mass, either solid or fluid, that does not change state during the process [5]. If used for electricity storage purposes, the heat is recovered via the production of water vapour which drives a turbo-alternator system [6]. It is today the most commonly used type of thermal storage, using either a liquid like water, synthetic oils or molten salts as the storage medium, or a solid medium such as rocks or ceramics [10]. Operating temperature is most often between 200-300 °C for low temperature storages and up to 1400 °C for hot storages [6]. Molten salt, which as of now, is considered the best thermal storage medium for solar power plant applications and operates at temperatures of up to 850 °C [11].

SHS tends to suffer from low energy density, where they also struggle with losses due to the required temperature difference for the heat transfer driving force [6]. If paired with thermal solar plants, the cost of the stored energy can be pushed extremely low, as seen in Table 13.

(20)

19

Table 13 Sensible Heat Storage properties

P owe r [ MW ] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ie nc y [% ] R esponse tim e Life S pa n C ost Environmental impact 50 [10] Gr ows w it h st ora ge unit Varies with temperature and material - - - 1 S EK /kWh [12]

2.5.2 Latent Heat Storage (LHS)

LHS makes use of phase change materials (PCMs) to store the thermal energy, in terms of latent heat of fusion. Therefore the considerable amounts of energy that is used or released when a material changes its phase is stored. LHS uses almost exclusively the solid to liquid phase change due to being more efficient and easier to handle than other phase changes [6]. The world’s most common PCM is water, more specifically the phase change of ice to water. An example of a facility using water for LHS is the hospital in Sundsvall, Sweden. The hospital collects snow in the winter, to later in the year use the cold for cooling [13]. A more detailed description of the process can be seen in Figure 6.

(21)

20

Another popular PCM for LHS purposes is sodium hydroxide. The material is highly corrosive, complicating the recycling process, but the high fusion temperature of the material, high-temperature stability and low steam pressure makes it very attractive as a thermal medium [5]. LHS generally has a higher storage density when compared to SHS [11], especially LHS using PCMs with high fusion temperatures as the energy density of the PCM generally increases with its temperature [6].

Table 14 Latent Heat Storage properties

P owe r [ M W] C apa cit y [MW h] S tora ge P eriod Storage Density Ef fic ienc y [% ] R esponse tim e Life S pa n C ost Environmental impact Varies with temperature and PCM Highly dependent on the PCM 2.5.3 Thermo-Chemical Storage (TCS)

TCS uses chemicals that absorbs and releases large amounts of thermal energy when reacting. [6], [11]. If the reaction is hindered from occurring, long time can pass from charging at low demand to discharging at higher demand without significant losses [6].

(22)

21

Table 15 Thermo-Chemical Storage properties

(23)

22

3 C

OMPARISON

(24)

23

Table 16 Comparison table

Power [MW]

Capacity [MWh]

Storage

Period Storage Density

Efficiency [%]

Response

time Life Span

Cost

[SEK/kWh] Environmental impact

CAES 30 - 350

[2]

Grows with

storage unit Months [2]

Varies greatly with pressure, ranging

from 0.5 kWh/m3_{[4] to 12 kWh/m}3_[5] _{70 [2], [5]} _{Minutes [4]} _{20 - 40 years [2], [4]} _{19 - 1,325 [6]} Exhausts from fuel combustion impacts

the environment to some extent

FES 0.002 - 20

[7] 0.001 - 0.1 [7] Hours [8]

100 [3] - 200 Wh/kg [4]

20 - 80 kWh/m3_[6] 90 - 95 [4] Seconds [7] 20 years [3], [8]

946,220 -

3,784,880 [6] Light environmental impact

PHES <3000 [7] Grows with

storage unit - 0.35 - 1.12 kWh/m

3_[4] 65 - 85 [5],

[7] Sec-min [4] 50 - 100 years 95 - 662 [6]

May be harmful for local nature and wildlife due to size

SMES 10 [7] 0.01 - 0.030 [7] Hours - Days [7] 40 - 60 Wh/kg [4] 0.2 - 2.5 kWh/m3_[6] 90 % [7] Milliseconds [7] 30 - 50,000 cycles [4] 596,119 -

7,096,650 [6] Light environmental impact

Supercapacitors 0.001 - 5

[7] 0.001 - 0.01 [7] - Up to 20 Wh/kg [7] 95 [7]

Milliseconds [7] 10

4_{- 10}6_{cycles [7]} _{45,394 [7]} _{Chemical disposal issues}

Li-Ion 0.001 - 0.1

[4] - Days [7]

80 - 200 Wh/kg [5] [7]

200 - 500 kWh/m3_[6] 80 - 95 [4] - 200 - 250,000 [4]

1,680 [3] - 18,86

[6] Chemical disposal issues

Lead-Acid - - - 30 Wh/kg, 180 W/kg [3] 50 - 80 kWh/m3_[6] 75 - 90 [3], [7] - 5 - 17 years, 1200 - 2700 cycles [4] [3] 944 - 1,700 [6]

Light environmental impact if recycled properly, contains lead-acid

Na-S 0.05 [7] 0.4 [7] - 60 Wh/kg [4] 156 - 255 kWh/m3_[6] 80 - 85 [4] Milliseconds [7] 15 years 4,500 cycles [4] 4800 [4]

Light environmental impact

Nickel-based 0.5 - 50 [4], [7] - Months [7] 50 - 80 Wh/kg [3] 60-150 kWh/m3_[6] 70 - 90 [4], [7] - 10 - 20 years, <300 cycles [7], [3] 4,249 - 16,997

[6] Chemical disposal issues

Flow batteries 0.05 - 15

[4], [5] 120 [5] - 16 - 33 kWh/m

3_[6] 75 [5] - 85

[4]

Milliseconds

[4] - 1,039 - 9,443 [6] Chemical disposal issues

Power-to-Gas 1 - 100 [9] Grows with

storage unit Days [3] 2.7 - 160 kWh/m

3_{at 1 - 700 bar [6]} _{62 - 82 [4]} _{Sec-min [4]} _- _{19 - 142 [6]} _{Light environmental impact}

SHS 50 [10] Grows with storage unit

Months

[12] Varies with temperature and material - - -

1 [12]

LHS - - - Varies with temperature and material - - - -

(25)

24

3.1 D

ISCUSSION

Different techniques differentiate in areas such as technical aspects, environmental impact, social impact and technology maturity level. All of these factors, as well as the application area, weigh in when an energy storage system is chosen as there is none storage option that covers every need.

3.1.1 Technical aspects

Although there are a lot of technical differences between the different storage technologies, they are not always the most important factor when deciding for an energy storage option. Sometimes, the technical aspects have little importance by themselves, but can affect for example the price of the stored energy.

The power describes how much energy that can be released at a time. Having the energy storage match its source in terms of power is therefore a good idea for UPS purposes. But even if the differences between e.g. Li-Ion batteries and PHES, as can be seen in Table 16, can be seemingly huge, smaller installations such as batteries and FES can be connected in series in order to raise their combined power. The power outage of the energy storage is therefore important, but not an impossible obstacle.

(26)

25

3.1.2 Environmental impact

No energy storage methods comes without environmental impact, even if the method does not affect the environment while operating chemicals and components always need to be produced and recycled to some environmental cost as either raw materials or pollution from the production. More notable are those that directly impact the environment as they operate. For example, the most commonly used energy storage in Sweden, PHES, occupies large parts of the surrounding area, disturbing wildlife and the connected eco-systems. Hydroelectric power plants have been built in Sweden to the extent that the streams suitable for PHES installations still remaining with large capacities are protected as nature reserves and due to social protests [14]. It is therefore believed that efficiency improvements will have a larger impact on Swedish PHES than further extension.

Similar in occupying large areas is CAES. Unlike PHES the method makes use of already existing underground caves, reducing the visible impact. For efficiency reasons, CAES requires using fuel to mix with the air, releasing fumes contributing to air pollution and the greenhouse effect. For a more environmentally friendly CAES-system, fossil-based fuels could be replaced by biofuel of some kind, hopefully reducing the amounts of pollutions released. Systems could also be added to make better use of the heat from the combusted fuel, which otherwise will be lost.

An attribute shared among several types of battery energy storages is chemical disposal issues as they all rely on chemical components to react. The exceptions of this mainly being Na-S batteries as these are made out of non-toxic materials that are easy to recycle. Lead-acid batteries can be considered easily recycled to some extent, but the lead in them poses a threat to eco-systems at leaks and ultimate disposal.

Depending on the thermal medium chosen, all of the TES technologies may or may not impact the environment. Several thermal mediums leaves little to none environmental impact if handled properly, such as water that can be released if needed once cooled down or liquid salt and solids such as concrete that are easily recycled. Other mediums, such as synthetic oils and chemicals for TCS may be more difficult to recycle.

3.1.3 Social impact

(27)

26

of cities, some draws greater attention. The most notable one being PHES, as noted above in 3.1.2, which is hindered by social protests due to their large social and environmental impact on the local area. Another exception is CAES which tends to utilize caves and mines on the countryside.

As the ground serves as an insulator, TES with SHS can favour greatly from being buried, making thermal energy storages naturally hidden. High temperature SHS does however pose a risk in case that an accident would occur, as some materials are heated above 1000 °C. A leakage could potentially cause substantial damage to the employees, and to the nearby area as well as the environment.

Some batteries, such as Li-Ion and Na-S, includes reactants which may behave violent. Most modern Li-Ion cells keeps the lithium bound at all times keeping it from igniting. In Japan, 2012, a fire was reportedly caused by a Na-S battery [7].

Aside from the physical form of the energy storage, the factor with the greatest effect on the public is the cost of the energy stored. As mentioned in section 3.1.1, Technical aspects, the price of the stored energy is decided by many different factors. Table 17 lists the technologies from potentially cheapest to most expensive.

Table 17 Cost of energy stored in SEK/kWh

Technology Cost [SEK/kWh]

1 Sensible Heat Storage (SHS) 1

2 Power-to-Gas 19 - 142

3 Compressed Air Energy Storage (CAES) 19 – 1,325

4 Thermo-Chemical Storage (TCS) 76 - 944

5 Pumped Hydro Energy Storage (PHES) 95 - 662

6 Lead-Acid 944 - 1,700

7 Flow batteries 1,039 - 9,443

8 Lithium-Ion batteries (Li-Ion) 1,680 - 18,886

9 Nickel-based batteries 4,249 - 16,997

10 Sodium-Sulphur batteries (Na-S) 4,800

11 Supercapacitors 45,394

12 Superconducting Magnetic Energy Storage (SMES) 596,119 - 7,096,650

(28)

27

When studying Table 17, it stands clear that storing thermal energy is highly cost efficient, regrettably no details regarding the price of LHS have been found. One can also see that it is cheaper storing energy using long term options such as CAES and PHES, the exception of this being Power-to-gas which has the potential to store gas at a very low cost despite not being as mature as several other techniques on the list.

The medium range price class in Table 17 is occupied with electro-chemical options, with SMES and FES being the most expensive by some margin.

3.1.4 Technology maturity

Most technologies listed are to some degree under development. Below in Table 18 is a rough listing of the most promising technologies in term of technology maturity. PHES is the most commonly used as well as the most mature technique.

Table 18 Technology maturity level listing. Based on [4].

Technology maturity

1 Pumped Hydro Energy Storage (PHES)

2 Lithium-Ion Batteries (Li-Ion)

3 Lead-Acid Batteries

4 Compressed Air Energy Storage (CAES)

5 Sodium Sulphur batteries (Na-S)

6 Flywheel energy storage (FES)

7 Flow Batteries

8 Supercapacitors

9 Superconducting Magnetic Energy Storage (SMES)

10 Power-to-Gas

Similar to PHES, but not quite as mature, is CAES. Although CAES have been operating on the market since 1978, there are very few CAES-facilities operating in the world [2]. Despite reliability at a satisfactory-level, it fills a similar role in energy storage as the more favoured PHES. PHES have long been considered an easier method when compared to CAES, which may be a reason for the humble usage of CAES to this point.

(29)

28

scale energy production, such as domestic solar or wind power. Due to their unreliable nature, short life span and shallow discharge depth, they are not yet optimal for UPS [4]. Li-Ion batteries are popular due to their high energy density and their area efficiency, as large storage capacities can be kept in smaller facilities [4]. The technique has during the recent years been introduced to the market, a nearby example is a facility in the UK’s Okney Islands, outside the coast of Scotland, where a 2 MW Li-Ion battery is installed and was taken into operation in 2013 [7]. There is also a smaller facility already installed in Sweden, Falköping, a 75 kW Li-Ion is used to regulate the locally produced energy from wind turbines [15]. Na-S offers similar qualities as Li-Ion, although not as mature and therefore tend to underperform in comparison with Li-Ion in several categories. The technique is to some extent commercialised in Asia [7] but further development may improve safety.

A mechanical contestant to the various batteries is FES, a short-term energy storage in development offering similar qualities such as short reaction time and high energy density which is ideal for peak shaving. The technique has to some degree been introduced to the market with several operating facilities in both Europe and America [8]. High energy density and good performance are highlights for the method, but high storage costs makes the technique quite expensive.

In terms of thermal energy storage, a widely used method is thermal SHS with Spain and U.S.A. being the two main users [11]. SHS is commonly paired with thermal solar plants, the climate in Sweden does not, however, favour this method for UPS. Larger facilities, such as industries and residential areas, can also use SHS as seasonal thermal storage, saving heat in the summer to release in the winter and vice versa. Notable users already operating in Sweden includes Stockholm Arlanda Airport [16] and the Anneberg residential area [12]. Similarly, LHS is also being used by facilities such as hospitals, as mentioned in section 2.5.2, Latent Heat Storage (LHS).

Thermal storage in Sweden does not however seem to be fit for UPS applications and should be kept as storages for cooling or heating.

3.1.5 Need and availability

(30)

29

favourable when trying to find a cavern suitable for acting as storage for a CAES-facility, though I have not found any studies or investigations of the subject. Lead-acid batteries can also serve as long-term storage, but they need more development before commercial use is an option.

TES is interesting for Sweden, not mainly for UPS purposes but as temperatures in Sweden ranges greatly from summer to winter a great amount of energy can be stored and thus saved when used as a seasonal thermal storage. The Arlanda Airport Aquifer Thermal Energy Storage reportedly offer 10 GWh per year in terms of district heating [17].

Regarding short-term energy storage, two types of batteries are to some degree introduced to the market, namely Li-Ion and Na-S batteries. A Li-Ion battery is (as noted in 3.1.4) already installed in Sweden since 2013 and can therefore be seen as a proven concept. Na-S batteries are used all over the world, either combined with wind turbines or used for UPS services. In Germany, 2013, Younicos and Vattenfall opened up a combined Na-S and Li-Ion battery with a capacity of 1 MW in a joint pilot project. The battery was the first large-scale battery to be integrated in the European electricity balancing market [18].

FES is today mostly used in a smaller scale, such as hospitals and factories. There are exceptions though, like the Mainstream Renewable Power flywheel storage in Ireland, a 2 MW FES storing electricity when the demand on the local wind turbines are low [19].

3.2 C

ASE STUDY

–

W

IND FARM AT

B

IOTESTSJÖN

In 2007 Vattenfall applied for permission to build a wind farm close to the nuclear power plant Forsmark kraftverk, at the Biotestsjön Lake, about 15 km north west from Öregrund. The wind farm would have a combined power of 2.3 MW and could generate a total of 105 GWh per year [20]. With the nuclear power plant nearby, the wind farm would be a great way to supply extra power at times of high

demand on the power grid. To ensure that the wind farm can supply power whenever needed, combining it with an energy storage system would be a good idea.

(31)

30

The main application for the energy storage would be to store energy from night time, when the demand is low, to day time when demand gets higher. Therefore, a storage period ranging from a few hours to at most a few days would be enough.

As the location, Biotestsjön, is a manmade lake formed at the outlet of the power plant’s cooling canal the temperature in the lake is six to eight degrees above the temperature of the surrounding sea, making it a unique opportunity to study the correlations between water temperature and biological processes [21]. As the area serves for research purposes as well as a home for wild life such as eagles, keeping the environmental impacts of the energy storage would be highly important.

3.2.1 Approach

As both PHES and CAES affect their immediate surrounding area, and as none of the optimal conditions for these technologies seems to be met, none of them are to be considered as an option for the wind farm’s energy storage. As mentioned in 3.1.4, Technology maturity, the viable options remaining are Li-Ion batteries, lead-acid batteries, Na-S batteries and FES. Below in Table 19 the information regarding said technologies from Table 16 are repeated for easier comparison.

Table 19 Case study energy storage options

Li-Ion Lead-acid Na-S FES

Power [MW] 0.001 - 0.1 - 0.05 0.002 - 20

Capacity [MWh] - - 0.4 0.001 - 0.1

Storage period Days - - Hours

Storage density 80 - 200 Wh/kg, 200 - 500 kWh/m3 30 Wh/kg, 50 - 80 kWh/m3 60 Wh/kg, 156 - 255 kWh/m3 100 - 200 Wh/kg, 20 -80 kWh/m3 Efficiency [% ] 80 - 95 75 - 90 80 - 85 90 - 95

Response time - - Milliseconds Seconds

Life span 200 - 250,000 cycles 5 - 17 years, 1,200 - 2,700 cycles 15 years, 4500 cycles 20 years Cost [SEK/kWh] 1,680 - 18,886 944 - 1’700 4,800 946,220 - 3,784,880 Environmental impact Chemical disposal issues Chemical disposal issues Light impact if recycled properly Light impact

(32)

31

lead-acid option. The benefits of lead acid batteries in this case would be their favourable price and light local environmental impact, provided that no leakage of lead to the nearby nature occur. Final disposal of the lead would in the end affect the environment, but probably in another location. The wind farms’ crucial need for reliability and the technology’s weak performance in general does however outweigh these benefits and make lead acid poor option. Out of the remaining three, Li-Ion is by far the more commonly used. It can potentially match FES in terms of energy density and efficiency, as well as Na-S in terms of costs. Li-Ion is however the option with the largest environmental impact, also including the threat of ignition which might bring large risks to the nearby power plant as well as environment. Any investigations regarding the nuclear plants’ safety concerns in case of fires has not been done with this case study.

Na-S has less environmental impact than Li-Ion as the components are easier to recycle. In terms of life span, efficiency and energy density, Na-S falls off compared with Li-Ion and FES, aside from having similar safety concerns as Li-Ion. Pricewise, Na-S is comparable with Li-Ion or cheaper and substantially less expensive than FES.

FES offers high energy density and efficiency, and it doesn’t affect the environment much once it is up and running. Despite being a relatively mature technology, as mentioned in 3.1.4 Technology maturity, it still comes at a much higher price than other technologies. The self-discharge rate is also quite high with efficiency dropping to about 78 % after five hours [5].

3.2.2 Case study conclusion

(33)

32

4 C

ONCLUSION

For long term storage, PHES is the strongest contestant yet and should be the main choice due to the maturity of the technology. If expanding PHES further is not possible the options are the unreliable but easily recycled lead-acid batteries or the fossil fuel-reliant CAES which also requires a suitable container. If it was to turn out that cavities in Sweden are suitable for CAES it seems to be the more favourable option, this due to the lower cost, superior life span and high power capacity.

For short term energy storage, Li-Ion batteries or Na-S seems to be the best options. Though Li-Ion batteries outperforms Na-S batteries in terms of energy density, efficiency and lifespan, Na-S batteries are not too far behind. Na-S batteries are also much easier to recycle than Li-Ion, making them the better option from an environmental point of view.

FES are interesting due to their higher power outage and good energy density, but seems to be too expensive to use for the time being.

Outside of UPS purposes, TES seems to be a potentially cost effective and environmentally friendly option for heating and cooling of facilities.

4.1 F

UTURE

W

ORK

Closer studies of several fields in this thesis would provide much needed information which have been left out. Most notably are geographical conditions of PHES and CAES. How much potential for further expansion of PHES is there left and is any of the mines in Sweden suitable for CAES?

As this thesis have mainly been focused on established technologies, further investigation could also be done regarding what is about to come. As mentioned in section 3.1.2, Environmental impact, CAES could use additional systems to make use of otherwise lost heat. Depending of if such a feature could successfully be implemented in five years or in twenty years, the social and environmental impacts could change drastically.

(34)

33

5 A

CKNOWLEDGEMENTS

I am very grateful to have had both of my advisers, Justin Chiu and Saman Nimali Gunasekara, helping me through this report. I thank Justin Chiu for giving me a great start to my work, you helped me understand the goal of my work from the beginning and gave me the starting point to make it possible to finish through. And I thank Saman Nimali Gunasekara for being there for me at all times. You have been a constant source of encouragement, help and constructive feedback, which without I could never have written this report.

I would also like to thank the “KTH School of Industrial Engineering and Management” for providing me with the chance of writing this Bachelor of Science thesis report, and last but not least I would like to thank Catharina Erlich for the trouble you went through helping me find a wonderful subject to write my thesis about.

(35)

34

6 R

EFERENCES

[1] A. Cavallo, “Controllable and affordable utility-scale electricity from intermittent wind resources and compressed air energy storage (CAES),” 2005.

[2] H. Chen, X. Zhang, J. Liu and C. Tan , “Compressed Air Energy Storage,” INTECH, 2013.

[3] I. Hadjipaschalis, A. Poullikkas and V. Efthimiou, “Overview of current and future energy storage technologies for electric power applications,” Electricity Authority of Cyprus, 2008.

[4] A. Nordling, R. Englund, A. Hembjer and A. Mannberg , “Energilagring - Teknik för lagring av el,” Kungl. Ingenjörsvetenskapsakademien (IVA), 2015.

[5] H. Ibrahim, A. Ilinca and J. Perron, “Energy storage systems - Characteristics and comparisons,” Elsevier Ltd, 2007.

[6] H. Zhang, J. Baeyens, G. Cáceres, J. Degrève and Y. Lv, “Thermal energy storage: Recent developments and practical aspects,” Elsevier Ltd., 2015.

[7] P. Breeze, “Chapter 10 - Power System Energy Storage Technologies,” in Power

Generation Technologies (Second Edition), 2nd ed., Elsevier Ltd, 2014, p. 195–221.

[8] R. Hebner, J. Beno and A. Walls, “Flywheel Batteries Comes Around Again,” IEEE Spectrum, 2002.

[9] S. B. Walker, U. Mukherjee, M. Fowler and A. Elkamel, “Benchmarking and selection of Power-to-Gas utilizing electrolytic hydrogen as an energy storage alternative,” University of Waterloo, Waterloo, 2015.

(36)

35

[11] Y. Tian and C. Zhao, “A Review of Solar Colletor and Thermal Energy Storage in Solar Thermal Applications,” 2012.

[12] J.-O. Dalenbäck, J. Dahm, S.-E. Lundin, G. Hellström and B. Nordell, “Solar Heated Residental Area Anneberg,” 2000.

[13] “Snowpower,” Snowpower AB, 2005. [Online]. Available:

http://www.snowpower.se/sundsvalls-kylanlaggning.asp. [Accessed 14 06 2016]. [14] J. Henriksson, “Striden om Vindelälven - Början till slutet på den svenska

vattenkraftsutbyggnaden,” Luleå Tekniska Universitet, 2004.

[15] “DOE Global Energy Storage Database,” Sandia Corporation, 18 06 2012. [Online]. Available: http://www.energystorageexchange.org/projects/113. [Accessed 12 05 2016]. [16] O. Andersson, “www.slideshare.net,” 2 December 2013. [Online]. Available:

http://www.slideshare.net/SGU_Sverige/olof-andersson-akviferlager-arlandatermisk-resurspdf. [Accessed 1 Maj 2016].

[17] “DOE Global Energy Storage Database,” Sandia Corporation, 18 08 2014. [Online]. Available: http://www.energystorageexchange.org/projects/1487. [Accessed 12 05 2016].

[18] “DOE Global Energy Storage Database,” Sandia Corporation, 10 09 2013. [Online]. Available: http://www.energystorageexchange.org/projects/605. [Accessed 19 05 2016]. [19] “DOE Energy Storage Global Database,” Sandia Corporation, 02 03 2016. [Online].

Available: http://www.energystorageexchange.org/projects/2067. [Accessed 19 05 2016].

[20] M. U. Län, “Ansökan om tillstånd enligt 9 kap. miljöbalken (MB) till gruppstation för vindkraft på fastigheterna Forsmark 6:5 och 6:8 i Östhammars kommun,” Länsstyrelsen Uppsala Län, Uppsala, 2012.

[21] Vattenfall, “Biotestsjön - ett forskningsakvarium i jätteformat,” [Online]. Available:

(37)

36

[22] “Oanda,” OANDA Corporation, [Online]. Available: