Long-duration energy storage: A technoeconomic comparative analysis with case studies in Mexico

(1)

Master of Science Thesis

KTH School of Industrial Engineering and Management Energy Technology TRITA-ITM-EX 2020:612

Division of Heat and Power Technology SE-100 44 STOCKHOLM

Long-duration energy storage: a technoeconomic comparative analysis with case studies in Mexico

Nayeli Gallardo

(2)

-2-

Master of Science Thesis TRITA-ITM-EX 2020:612

Long-duration energy storage: a technoeconomic comparative analysis with

case studies in Mexico

Nayeli Gallardo

Approved

Dec. 15, 2020

Examiner

Björn Laumert

Supervisor

Rafael Eduardo Guédez Mata

Abstract

While the interest in energy storage has grown in recent years, attention has been largely focused on short- duration systems with lithium-ion batteries. Long-duration (4-24 h) technologies, their business cases, and their potential to contribute to the energy transition have remained largely unexplored. Drawing from both academic and industry publications, this thesis presents the state of the art of energy storage technologies suitable for long-duration applications and performs a technoeconomic analysis of two technologies (lithium-ion and flow battery) applied to two case studies in Mexico.

This report presents the most relevant energy storage technologies that can provide long duration storage.

It also briefly explores the general use cases for storage and the business models typically employed. Two case studies for PV+storage systems in Mexico are also developed, one for a behind-the-meter industrial user in 2021 and another for an independent power producer in 2025. Two storage technologies, a lithium iron phosphate (LFP) and a vanadium redox flow (VRF) battery, are chosen for both cases, based on the appropriateness and maturity of the technology and the availability of data. The technoeconomic performance of these technologies is evaluated using purpose-built models and varying system size and duration, as well as PV plant size. Additional revenues from Clean Energy Certificates are included.

Sensitivity to key parameters is also assessed. The resulting indicators are compared.

The case studies suggest that PV+storage is attractive in Mexico only when large levels of self-sufficiency and/or clean energy are valued, although the electricity market and rates may change significantly in the coming years. LFP is found to be competitive against VRF for mid-range levels of self-sufficiency, whereas VRF is more competitive in the last 15-30% of self-sufficiency. To meet the last portion of demand and reach complete self-sufficiency, another source (such as a fuel genset) would likely be more economical to avoid drastic oversizing of the PV plant. VRF and LFP systems with fast charging may be significantly more competitive than the base cases, but more research is required to verify the assumptions. Business recommendations for PV+storage projects in Mexico are offered, considering findings from both the technologies and the country’s energy sector.

(3)

-3-

Sammanfattning

Medan intresset för energilagring har ökat de senaste åren har uppmärksamheten i stor utsträckning riktats mot kortvariga system med litiumjonbatterier. Långvarig (4-24 timmar) teknik, deras affärsfall och deras potential att bidra till energiomställningen har förblivit i stort sett outforskad. Denna avhandling bygger på både akademiska publikationer och branschpublikationer och presenterar den senaste tekniken för energilagringsteknologier som är lämpliga för långvariga applikationer och utför en tekniskekonomisk analys av två tekniker (litiumjon- och flödesbatteri) som tillämpas på två fallstudier i Mexiko.

Denna rapport presenterar de mest relevanta teknologierna för energilagring som kan ge långvarig lagring.

Den utforskar också kort de allmänna användningsfall för lagring och de affärsmodeller som vanligtvis används. Två fallstudier för PV + -lagringssystem i Mexiko har också utvecklats, en för en industriell användare bakom mätaren 2021 och en för en oberoende kraftproducent 2025. Två lagringstekniker, ett litiumjärnfosfat (LFP) och en vanadin redox flow (VRF) -batteri väljs för båda fallen, baserat på lämpligheten och mognaden hos tekniken och tillgängligheten av data. Den tekniska ekonomiska prestandan hos dessa tekniker utvärderas med hjälp av specialbyggda modeller och varierande systemstorlek och varaktighet, samt solcelleanläggningens storlek. Ytterligare intäkter från rena energicertifikat ingår. Känsligheten för nyckelparametrar utvärderas också. De resulterande indikatorerna jämförs.

Fallstudierna tyder på att PV + -lagring är attraktiv i Mexiko endast när stora nivåer av självförsörjning och / eller ren energi värderas, även om elmarknaden och priserna kan förändras avsevärt de närmaste åren. LFP har visat sig vara konkurrenskraftigt mot VRF för medelhög nivå av självförsörjning, medan VRF är mer konkurrenskraftigt under de senaste 15-30% av självförsörjningen. För att möta den sista delen av efterfrågan och uppnå fullständig självförsörjning skulle en annan källa (t.ex. ett bränslegenerator) sannolikt vara mer ekonomiskt för att undvika drastisk överdimensionering av solcelleanläggningen. VRF- och LFP-system med snabb laddning kan vara betydligt mer konkurrenskraftiga än basfallet, men mer forskning krävs för att verifiera antagandena. Affärsrekommendationer för PV + -lagringsprojekt i Mexiko erbjuds med beaktande av resultat från både teknologierna och landets energisektor.

(4)

-4-

Acknowledgments

I am profoundly grateful to my academic supervisor Rafael Guédez and my supervisor at Azelio Anna Gerokostopoulou for guiding me throughout this thesis work and patiently teaching me so much about the exciting world of energy storage. You helped me develop valuable skills along the way, and I know that these will serve me well in my professional career.

I am also indebted to many others at Azelio for the support they provided, particularly Nicco Oggioni and Felipe Gallardo, who provided me with helpful information and resources as I built the case studies. I would like to acknowledge the irreplaceable support from Jonas Wallmander and Ralph Wiesenberg as well. I would have been lost without the technical and administrative assistance that Olov Sandzén and Marika Olsson provided me with from Åmål. To everyone I had the pleasure of sharing the Stockholm office with (before the pandemic sent us all home), thank you for your support and insights: Anna, Emil, Felipe, Johan, Martin, Osama, Tine, and Youssef. I wish to also show my gratitude to my friends outside Azelio who answered my questions regarding Mexican policies and other relevant topics: Alicia, Mariana, Natalia, and Carlos.

Moreover, I would not have been able to carry out this work without the opportunity that Azelio granted me by selecting me as an intern. Their sponsorship of my attendance of the Energy Storage Summit also led to invaluable insights for the research phase of the project. Similarly, I extend my gratitude to InnoEnergy for admitting me into the SELECT program and funding my participation in it, which allowed me to begin this thesis project at KTH.

Additionally, I wish to express my dearest gratitude to my now-husband Hugo for the loving support and wise advice he offered me as I worked through this thesis project. Likewise, I am sincerely thankful for the unconditional support from my mother, Rebeca; my father, Ezequiel; and my sister, Berenice. I also thank my friends and my former roommate, Iryna, for their caring words of encouragement.

Thank you.

(5)

-5-

Abbreviations

AA-CAES Advanced adiabatic compressed air energy

storage LMO Lithium manganese oxide

AHI Aqueous hybrid ion NCA Lithium nickel cobalt aluminum

BCS Baja California Sur LTO Lithium titanate oxide

BOS Balance of system LMP Locational marginal price

BESS Battery energy storage system SMES Magnetic energy storage BMS Battery management system MRL Manufacturing readiness level BCSIS BCS Interconnected System SENER Ministry of Energy

BoL Beginning-of-life NREL National Renewable Energy

Laboratory

BTM Behind-the-meter NPV Net present value

BNEF Bloomberg New Energy Finance NMC Nickel manganese cobalt oxide

CAPEX Capital expenses NiCd Nickel-cadmium

CB Carnot battery OPEX Operational expenses

CEL Clean Energy Certificate (in Spanish) PCM Phase-changing material CFE Comisión Federal de Electricidad PCS Power conversion system C&I Commercial and industrial PPA Power purchase agreement CAES Compressed air energy storage PHES Pumped hydroelectric energy

storage

CSP Concentrated solar power RFB Redox flow battery

DoD Depth of discharge RTE Roundtrip efficiency

EoL End-of-life SS Self-sufficiency

EMS Energy management system ZEBRA Sodium metal halide, Zero Emission Batteries Research Activity

E:P Energy-to-power ratio NaNiCl Sodium nickel chloride

EPC Engineering, procurement, and construction NaS Sodium-sulfur

EFC Equivalent full cycle SEI Solid-electrolyte interface

FTM Front-of-the-meter SoC State of charge

GPM Gravity power module SoH State of health

IPP Independent power producer TRL Technology readiness level ISO,

CENACE Independent system operator TES Thermal energy storage

IRR Internal rate of return TMS Thermal management system

IEA International Energy Agency TCES Thermochemical energy storage IRENA International Renewable Energy Agency TOU Time-of-use

KPI Key performance indicator UPHES Underground pumped hydroelectric storage

LCOE Levelized cost of energy UOSS Underwater ocean storage systems

LCOS Levelized cost of storage UF Utilization factor

LAES Liquid air energy storage VRLA Valve-regulated lead-acid

Li Lithium VRF Vanadium redox flow

LCO Lithium cobalt oxide ZBR Zinc-bromide flow

LFP Lithium iron phosphate

(6)

-6-

List of figures

Figure 1. Existing and expected energy storage needs ... 13

Figure 2. Expected additional energy storage capacity, 2017-2030 ... 13

Figure 3. Average system power capacity and duration of energy storage systems installed worldwide excluding pumped hydro, 1958-2017 ... 14

Figure 4. Schematic of the components of a battery energy storage system ... 16

Figure 5. Example of a Ragone plot ... 18

Figure 6. Battery discharge curves for different C-rates ... 18

Figure 7. Voltage differences during charging and discharging ... 20

Figure 8. Efficiency vs. durability for various technologies... 21

Figure 9. Specific power and energy capital costs ... 22

Figure 10. TRL of energy storage technologies ... 23

Figure 11. Schematic of a fuel cell ... 24

Figure 12. Schematic of a Li-ion battery ... 26

Figure 13. Utilization and efficiency for LFP systems and three applications ... 28

Figure 14. Effect of DoD on cycle lifetime for LFP batteries ... 31

Figure 15. Schematic of a Mg-Sb liquid metal battery during charging, storage, and discharging ... 32

Figure 16. Schematics of a battery, a semi-fuel cell, and a fuel cell ... 33

Figure 17. Schematic of a Mg-air fuel cell, its outward appearance, and the Mg anode insert ... 34

Figure 18. Schematic of a classical, separated RFB ... 36

Figure 19. Components of a RFB stack and of a RFB system ... 37

Figure 20. Schematic of a NaS battery ... 40

Figure 21. Schematic of a conventional CAES system ... 42

Figure 22. Schematic of a flywheel ... 43

Figure 23. Schematic of a Carnot battery (the TES.POD) ... 45

Figure 24. Power rating and discharge time for multiple energy storage technologies ... 46

Figure 25. Global energy storage power capacity shares by main-use case and technology as of mid-2017 47 Figure 26. Lowest-cost energy storage technology by application ... 48

Figure 27. Battery ownership models by country (operational and under construction) ... 48

Figure 28. Simplified system layout for a BTM and a FTM system ... 52

Figure 29. Historical and projected market electricity prices used in the BCS case study ... 52

Figure 30. Schematic of the battery systems in the case studies ... 54

Figure 31. Specific initial CAPEX for VRF systems ... 56

Figure 32. Specific initial CAPEX for LFP systems ... 56

Figure 33. Initial and lifetime storage CAPEX for a 20 MW AC system in 2025 [millions of 2020 USD] .. 57

Figure 34. Literature and market review data on DoD vs. cycle lifetime [EFCs] for LFP systems ... 59

Figure 35. DoD vs. cycle lifetime [EFCs] for the PowerBrick LFP battery ... 59

Figure 36. DoD vs. cycle lifetime [partial cycles] for the PowerBrick LFP battery projected for 2025 ... 60

Figure 37. Calendar lifetimes for LFP batteries [y] ... 60

Figure 38. Utility-scale PV system CAPEX in the case study for 2025 [2020 USD/kWpDC] ... 62

Figure 39. PV power potential in Mexico and location of case studies [205] ... 65

Figure 40. Average demand and TOU energy rate for the C&I user ... 66

Figure 41. Monthly demand and total bill for the C&I user ... 66

Figure 42. VRF systems with minimum LCOE in the C&I case ... 67

Figure 43. LFP systems with minimum LCOE in the C&I case ... 67

Figure 44. SS vs. LCOE for VRF systems in the C&I case ... 68

Figure 45. SS vs. LCOE for LFP systems in the C&I case ... 68

Figure 46. Cost of SS comparison in the C&I case ... 69

Figure 47. Cost of SS comparison in the C&I case for 132 kW/13h systems at 50% of the base case PV CAPEX ... 69

(7)

-7-

Figure 48. Map of the interconnected electric systems of Mexico ... 70

Figure 49. LMPs in the BCSIS in 2019 ... 71

Figure 50. Average electricity rates in BCS in 2020 ... 71

Figure 51. Range of electricity rates in BCS in Jan & Aug 2020 ... 71

Figure 52. Power demand in the BCSIS in 2019 ... 72

Figure 53. Monthly average electricity market prices in BCS for 2020 ... 73

Figure 54. PV oversize ratio vs. PPA price by duration for VRF and LFP systems in the IPP case ... 74

Figure 55. VRF systems with minimum PPA price in the IPP case ... 75

Figure 56. LFP systems with minimum PPA price in the IPP case ... 75

Figure 57. Cost of SS comparison for 13h systems in the IPP case ... 76

Figure 58. Cost of SS comparison in the IPP case for 13h systems with 93% DoD for LFP ... 76

Figure 59. Output lost to degradation in a 20MW/13h LFP system with 70 MWp of PV ... 77

Figure 60. Sensitivity of cost of SS to PV CAPEX in the IPP case ... 77

Figure 61. Sensitivity of cost of SS to land cost in the IPP case ... 77

Figure 62. Profile for May 2 with 100 MWp of PV and 20MW/13h or 43MW/6h LFP systems ... 78

Figure 63. Cost of SS comparison for 43MW/6 h VRF and LFP systems in the IPP case ... 78

Figure 64. Illustrative electricity storage dispatch including provision of grid services ... 87

Figure 65. General stages of company development ... 88

Figure 66. Li-ion battery companies and their business models ... 89

Figure 67. Flow battery companies and their business models ... 90

Figure 68. Other battery companies and their business models ... 90

Figure 69. A closer look into the business models of four companies ... 90

Figure 70. Potential site for a utility-scale PV+storage system in BCS, Mexico, created in Google Earth .. 96

Figure 71. Results of PVsystems without storage and with net metering for the C&I case ... 97

Figure 72. Results of PV systems without storage and without net metering for the C&I case ... 97

Figure 73. Results of PV systems without storage and for the IPP case ... 98

(8)

-8-

List of tables

Table 1. Description of the main characteristics of energy storage technologies ... 17

Table 2. Li-ion chemistries and their main components ... 27

Table 3. Lifetime expectations for Li-ion batteries in 2020 ... 31

Table 4. Categorization of RFBs by number of phase changes ... 37

Table 5. Total system RTE for LFP and VRF systems in the case studies... 61

Table 6. Opportunities and threats for long-duration PV+storage systems in Mexico ... 80

Table 7. Bulk energy services ... 83

Table 8. Customer/end user energy services ... 84

Table 9. Ancillary energy services ... 85

Table 10. Network energy services ... 86

Table 11. Description of identified business models ... 89

(9)

-9-

1 Introduction 1.1 Context

1.1.1 Definition of long-duration storage

Electrical energy storage can be defined as “a process of converting electrical energy into other forms of energy that can be stored for converting back into electrical energy when needed,” as described in [1].¹ Moreover, storage can be classified by the ratio of energy capacity to power in a storage system. This is called design duration, capacity-to-power ratio, or energy-to-power ratio. Long-duration applications are also termed energy applications, in contrast to power applications that require large amounts of power for short amounts of time. This report is focused on long-duration storage defined as 4 to 24 hours of duration at rated output power. This is also termed daily storage [2] [3].

Although limited attention has been focused on long-duration storage, the concept has been mentioned by significant players. Both the International Renewable Energy Agency (IRENA) and storage technology company ESS, Inc. define long-duration storage as greater than 4 hours [3] [4] [5], while DNV GL considers long-duration to last over 4-6 hours [6]. Industry media refer to a period of 4-5 hours when referring to the threshold for long duration storage [7] [8]. Few mentions of long-duration storage were found in the academic literature, although daily storage is defined by a capacity-to-power ratio of 4-8 h in one paper [2].

Interest in long durations for energy storage are also exemplified in the recent investments into long-duration storage technology developers [9], as emerging technologies tend to focus on providing over four hours of energy output [10].

1.1.2 Storage in the energy transition and sustainable development

While energy storage can provide various specific services, such as bill management and grid services, it is also seen as a key technology in the transition toward a cleaner energy system and in the efforts to advance energy access worldwide. Therefore, energy storage is particularly relevant for Sustainable Development Goals (SDGs) 7 (Affordable and Clean Energy) and 13 (Climate Action), as well as indirectly relevant for several of the other SDGs linked to these.

It is widely accepted that flexibility is required to achieve high penetrations of intermittent renewable energy and energy storage is an important (though not the only) source of flexibility. While many concentrated solar power (CSP) plants today include long-duration thermal storage, short-term storage is in use to address the short-term variability in solar photovoltaic (PV) and wind farms. Long-duration storage may be particularly relevant to address longer-term variability and to back up intermittent renewables. Because the value of grid- connected, PV or wind generation assets decreases as their installed capacity increases due to their non- dispatchable nature, bulk energy storage can help make high penetrations viable [11]. As seen in Germany, spot prices tend to fall when the renewable output is high, reducing the profitability of the renewable assets [12], unless some of that output can be used at a more profitable time.

Storage can contribute to clean, reliable and affordable energy systems in numerous ways. Figure 1 and Figure 2, both by IRENA, depict the electricity storage needs of the energy transition from 2017 to the 2030s and the projected use cases in installed gigawatt-hours (GWh).

1 Supercapacitors are an exception to this definition because they store energy as electrical energy, and therefore do not convert it into another form of energy.

(13)

-13-

Figure 1. Existing and expected energy storage needs [3]

Figure 2. Expected additional energy storage capacity, 2017-2030 [3]

(14)

-14-

1.1.3 Overview of storage installations today

According to the International Energy Agency (IEA), annual energy storage deployments have increased since 2013, except for 2018-2019 [13]. Approximately 3.1 gigawatts (GW) of energy storage projects were deployed in 2019, 42% of which was grid-scale or front-of-the-meter (FTM) and 58% was behind-the-meter (BTM). In 2019, Korea, China, Germany, and the U.S. were the leading countries in storage deployment.

While pumped hydro energy storage represents most of the current installed storage capacity, most systems installed in recent years have been based on Lithium-ion (Li-ion) batteries. As the costs of these batteries have fallen in the past years and continue to fall, largely due to demand in electric vehicles (EVs) & consumer electronics, the power sector has also taken advantage of these cost decreases.

Figure 3 shows the average system power capacity and duration of systems installed between 1958 and 2017 (excluding pumped hydro), according to the U.S. Department of Energy’s Energy Storage Database. Molten salt thermal storage, typically installed with CSP plants, and compressed air have the highest durations.

Figure 3. Average system power capacity and duration of energy storage systems installed worldwide excluding pumped hydro according to the U.S.

Department of Energy’s Energy Storage Database, 1958-2017 [14]

1.2 Azelio AB

The work presented in this report has been carried out with support from Azelio AB. Azelio AB is a Swedish energy technology company with 25 years of experience producing Stirling engines [15]. Its electrical thermal energy storage system (the TES.POD®) provides dispatchable power for 13 h by converting heat stored in a phase change material (PCM) into power via a Stirling engine. More information on this type of technology will be presented in the literature review. This work was sponsored as part of Azelio’s market research to better understand the technologies that may offer long-duration storage; therefore, the sizes and durations of the systems analyzed in the case studies are chosen in part to be comparable to Azelio’s TES.POD.

1.3 Objectives and scope

The primary objectives of this thesis research and report are the following:

1. Research, identify and characterize energy storage technologies that can provide long-duration storage, using indicators spanning across technical, economic, environmental, social, and operational dimensions and presenting explanations of these indicators as needed.

(15)

-15-

2. Present an overview of the use cases these energy storage technologies are utilized for and of the business models that are employed for these applications.

3. Explore business cases for long-duration storage coupled with PV power in Mexico; select two cases; and identify the relevant regulation, location-specific inputs, and financial model for each.

4. Identify the most relevant technologies to provide long-duration storage for PV-coupled systems in Mexico, build location-specific cost models for each, and evaluate their technoeconomic performance in the two case studies, using appropriate key performance indicators (KPIs) and varying the system duration in addition to the system rated power and the size of the PV plant.

5. Compare the technoeconomic performance of the two technologies and offer business recommendations for PV+storage projects in Mexico.

The following further define the scope and limitations of this work:

1. This report is focused on long-duration storage as defined in the introduction: 4-24 hours at rated power output.

2. Delivery of electricity from a storage system is the focus of this work. Delivery of heat in addition to electricity is also evaluated when necessary for a fair evaluation.

3. The research is also focused on technologies that are near commercialization or already commercialized. Technologies that have not at least reached the demonstration stage will be only briefly mentioned or entirely excluded.

4. This work focuses on the storage system (not the device) level. For example, for Li-ion, the focus is on the performance of the entire energy storage system, not only of the component cells.

5. Moreover, technologies will be considered individually. In practice, it is common to use multiple storage technologies in a single hybrid system.

6. Lastly, the technology evaluations are done from a user or private perspective. Thus, public or system impacts from energy storage may be overlooked.

1.4 Thesis structure

The rest of this report is structured as follows:

2. Literature review: The next section shows the most relevant findings from a literature review of energy storage technologies, their characteristics and their use cases.

3. Methodology: The methodology section explains the methodological decisions made to build the models and carry out the analyses of the two case studies and includes the data inputs for the models.

4. Results: The results section presents the main findings from the two case studies and provides some discussion of these results.

5. Conclusions: The final chapter contains the principal conclusions drawn from this thesis work and offers some recommendations for both business strategies and future related work.

2 Literature review

This section presents the main findings from a literature review on energy storage technologies, including an overview of the components of energy storage systems, an exploration of the main parameters that describe storage technologies, and a presentation of the most relevant energy storage technologies and their characteristics. Storage technologies are typically classified based on the form of energy used during the storage phase and this work therefore organizes them according to the following five categories outlined in [1] and other papers: chemical, electrical, electrochemical, mechanical, and thermal. In addition, examples of applications or use cases and business models are presented. A more detailed overview of the services that energy storage systems can provide can be found in Appendix 6.1. In addition to this literature review, a market review was performed to identify companies developing or commercializing long-duration energy storage technologies as well as to characterize their business models. Those results can be found in Appendix 6.2.

(16)

-16-

2.1 Energy storage systems

An energy storage system is more than the storage device alone. It requires additional components, as illustrated in the battery energy storage system (BESS) in Figure 4. While the system layout will differ by technology, other modular technologies will likely follow a similar structure.

Figure 4. Schematic of the components of a battery energy storage system [3]

Battery cells are assembled into modules and modules are put into a rack, cabinet, or tray. This forms the battery pack or storage module. The battery pack also includes the battery management system (BMS) and the pack-level thermal management system. The next component group includes the container, the monitors and controls, the overall thermal management system (TMS), and the fire suppression system; these are the balance of system (BOS) components. Another key component is the power conversion system (PCS), sometimes referred to as power electronics. Most storage devices that deliver electricity do so in direct current (DC) and require an inverter to convert this electricity to alternating current (AC) that the power grid and most electricity-consuming devices require [1]. The PCS system includes an inverter and electrical components such as switches and breakers, as well as an energy management system (EMS). A transformer may be necessary in some cases.

In addition to the physical system components described, an energy storage project will incur service-related costs. These include services such as engineering, procurement, and construction (EPC), permitting, commissioning, and project management. A project developer, an integrator, a grid connection, and land costs may also be necessary. When interpreting energy project costs, it is important to understand which costs are included in the reported figure. However, many of the cost figures reported do not declare this.

2.2 Main parameters of energy storage technologies

An understanding of the characteristics or parameters of energy storage technologies is necessary for their proper understanding and analysis. This section presents a summary of the technical, economic, and other characteristics identified in the literature.

2.2.1 Technical

Table 1. Description of the main characteristics of energy storage technologies, based on information from summarizes the most relevant technical characteristics of energy storage technologies. A few of these concepts merit further explanation, which is provided after the table.

(17)

-17-

Table 1. Description of the main characteristics of energy storage technologies, based on information from [16] [17] [18] [1] [19] [20]

Characteristic Description Unit Comments

Calendar life Amount of time a device is expected to be operational

Years -

Capacity fade Loss in original capacity as a function of another variable; determined by assessing the state of health (SoH)

% A function of cycles undergone at a specific DoD and/or a function of temperature

Cycle life Number of cycles a device can withstand using the manufacturer’s cyclic charging recommendations and maintain the minimum capacity required for an application

- Numerous rates and DoDs can be used for cyclic discharge testing, depending on the intended application

Energy capacity Rated capacity multiplied by the voltage Wh -

Energy density Energy stored per unit of weight (gravimetric/specific energy) or volume (volumetric density)

Wh/kg or Wh/l

-

Energy-to-power ratio (E:P)

Ratio of rated energy to rated power h Equal to duration

Maximum Depth of Discharge (DoD)

Maximum proportion of rated capacity that should be used, as recommended by the manufacturer

% Exceeding this can lead to shorter lifetime and efficiency losses; inverse of minimum SoC

Minimum State of Charge (SoC)

Minimum ratio of the remaining capacity to the rated capacity at a particular moment, as recommended by the manufacturer

%, Ah, coulomb s or Wh

Operating below this can lead to shorter cycle life and/or efficiency losses; inverse of maximum DoD

Power Current multiplied by the voltage W Different manufacturers may report different power concepts²

Power density Power that can be delivered per unit of weight (gravimetric/specific power) or volume (volumetric power)

W/kg or W/l

-

Rated capacity Amount of charge expected to be delivered by a new, fully formed cell under standard conditions

Ah Standard conditions used depend on the manufacturer and battery type

Response time³ Time to transition from a non- discharging state to full discharge

ms, s, or min

Can range from a few ms to several min

Self-discharge rate Percentage capacity loss during a specified period due to lack of use (or open-circuit condition)

Wh/d, or Wh/y

Sometimes expressed as a % per period

2 Peak power, rated power, or other measurements of power may be reported. Some sources note that it is measured at a specific SOC (usually 20%).

3 Response time is limited by the PCS [37].

(18)

-18-

Ragone plots compare energy density and power density, providing insight into the duration of an energy storage device. The Ragone plot was first developed by David V Ragone and an example from [16] is shown in Figure 5.

Figure 5. Example of a Ragone plot [16]

2.2.1.1 Capacity, C-rate, and discharge curve of batteries

The capacity of a battery will vary depending on the conditions under which it is measured. As noted in Table 1, these conditions vary and may not be fully stated in technical specifications. The actual capacity at a given moment depends on temperature, cell age, state of charge, and rate of discharge. The charge delivered to a load will also vary depending on the current drawn by the load [21].

Because the rate at which current is drawn from a battery affects the amount of energy that the battery will deliver, the C-rate is important. It is the ratio of the current to the battery capacity at that current [22].

Essentially, it describes how much the battery can be discharged every hour. It is also described in the literature as the ratio between the power rating and the energy rating [4], which is equivalent to the inverse of the design duration. Thus, if a battery is discharged over 10 hours, the 10-h discharge rate current is used, and the C-rate is C/10 or 0.1C. Similarly, a C-rate of 1 indicates a 1-h discharge [23]. A 2C rate indicates 30 minutes are required for complete discharge [4].

Cells of different sizes will charge or discharge in a similar way if the C-rate is the same. A lower C-rate will yield a greater capacity than a higher C-rate for the same battery [21]. That is, capacity is lost as C-rate is increased. This is evident in the battery discharge curve, which shows the relationship between discharge voltage and capacity. Examples of this are shown in Figure 6. The midpoint between 0% and 100% capacity is the nominal cell voltage [21]. The C-rate can be varied to maximize profits, as in the analysis in [23]. While the discharge curve of batteries is exponential, those of other technologies exhibit different shapes. For example, a supercapacitor has a linear curve (as its power decreases steadily during discharge) and the curve for compressed air is the inverse of the battery curve [24].

Figure 6. Battery discharge curves for different C-rates [21]

(19)

-19- 2.2.1.2 Lifetime & state of health

The lifetime of a storage device is the expected useful life and is sometimes referred to as durability. This parameter is particularly important for electrochemical devices, as they can degrade considerably over time.

The end-of-life (EoL) of a battery is typically defined as the point when its capacity has decreased to 80%

of its initial capacity, also known as beginning-of-life (BoL) capacity, considering nameplate conditions [21].

However, 70% is sometimes used as well [25]. Of course, a battery may be used past this point if it is appropriate for the specific project. The SoH of a battery is an instantaneous indicator determined by the BMS and is useful for estimating the remaining useful life of a battery pack [21]. Lifetime and degradation in this report refers to capacity, not power or other parameters.

Product specification sheets may contain lifetime data, but the type of information presented varies greatly and some key details are often omitted, which complicates comparisons. Battery manufacturers typically report lifetime in simplified ways such as the number of cycles (ideally specifying a DoD) or calendar days.

A warranty is sometimes included, which serves as an indication of minimum calendar lifetime.

A cycle is typically defined as a complete charge and discharge process [26]. However, reported cycle lifetimes may refer to equivalent full cycles (EFCs), which are measured by throughput and do not reflect the physical realities of partial cycles [27] or to cycles at a specified DoD, which may or may not be 100%. It is well understood that two half-cycles (at 50% DoD) do not have the same impact on battery degradation as one full cycle (at 100% DoD) [22]. Various modeling approaches convert the irregular discharge history to constant amplitude events for a more accurate estimation of the cycle life, and these typically use a rainflow cycle-counting algorithm [20] [28] [27].

2.2.1.3 Efficiency

Throughout this report, efficiency refers to energy efficiency. While several scopes are applied to the efficiency of storage devices and systems, these scopes are often not explicitly stated in the sources. Often, the terms “efficiency,” “peak efficiency,” and “roundtrip efficiency” are used in the academic literature and in technical specifications with few or no details as to what is considered. This renders the interpretation and comparison of efficiency data difficult. Moreover, several operational or environmental factors, such as utilization rate and temperature, can affect the efficiency of an energy storage device or system.

The use of a storage system involves three primary stages that must be considered to assess efficiency:

charging, storage, and discharging. During storage, some energy may be lost through time (indicated by the self-discharge rate) or some energy input may be required to maintain the charge of the storage device [18].

Additionally, energy storage systems include energy conversion processes for charging and discharging that exhibit losses (with the exception of technologies that store energy directly as electricity) [18].

The roundtrip efficiency (RTE) includes all three of these stages and is therefore the most often referred to figure for efficiency, although different scopes for it exist. It considers a full cycle—charging and discharging—and is affected by voltage variations in the charging and discharging processes. For batteries, the voltage increases during charging, and more power is required. While discharging, the voltage decreases, and the power delivered decreases [28]. Figure 7 shows this graphically. Separate efficiencies for charging and for discharging are seldom provided by manufacturers.

(20)

-20-

Figure 7. Voltage differences during charging and discharging [22]

Several sources, such as [29], refer to DC-DC and AC-AC RTE. The first is limited to the storage device, while the second one widens the scope to include ancillary loads up to the connection point. In either case, there may be variations in how manufacturers measure and report efficiencies.

Moreover, two concepts of RTE for batteries are defined in [30], although they are applicable to other storage technologies as well. The first is the conversion RTE, which includes losses during conversion in charging and discharging. This efficiency is limited to the losses in the battery as well as in the power electronics and is therefore calculated in AC.

𝜂_{𝑐𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛} = 𝐸𝐷𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒,𝐴𝐶

𝐸_{𝐶ℎ𝑎𝑟𝑔𝑒,𝐴𝐶}

The second is the total RTE, which in addition to the above includes energy consumed by the overall system for thermal management, control, and monitoring.

𝜂_{𝑡𝑜𝑡𝑎𝑙}= 𝐸𝐷𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒,𝐴𝐶

𝐸_{𝐶ℎ𝑎𝑟𝑔𝑒,𝐴𝐶}+ 𝐸𝑆𝑦𝑠𝑡𝑒𝑚𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛

For storage systems that involve heat engines, the efficiency is limited by the Carnot limitation. This maximum efficiency limit applies only to storage technologies that make use of energy conversion processes with temperature changes and not to isothermal processes such as those in fuel cells and batteries [31]. The maximum efficiency is expressed as

𝜂 = 𝑇_𝐻− 𝑇_𝐿

𝑇_𝐻 = 1 − 𝑇_𝐿 𝑇_𝐻

Lastly, methodological aspects can also impact reported efficiencies. For example, some models for batteries scale cell-level results to the rack level, possibly ignoring differences in the dynamics at each of these two scales. Modeling approaches may be categorized as either cell-focused (where a system is mostly a scaled up single cell plus power electronics efficiencies from the literature), system operation-focused (where a specific case is analyzed utilizing overall system efficiency values from the literature) or holistic (which evaluate component interdependencies) [30].

DNV GL reported that while the RTE for flow batteries tends to be 5-10% lower than for than the DC RTE for the best Li-ion batteries, the difference is negligible against the AC RTE because system losses are considered [6]. For a clear understanding of what a reported efficiency entails (and an accurate comparison of different technologies’ efficiencies), the following should be identified:

• Component scope: Reported efficiencies may be at the level of a single unit (cell for batteries), a module, or a system of varying scope. The existence or lack of auxiliary components, such as a thermal management system, could significantly impact the efficiency.

(21)

-21-

• Process scope: The efficiency may refer to only the charging or discharging process or both (roundtrip). Self-discharging during storage may or may not be included as well.

• Current type: While efficiencies reported in DC typically indicate that they only include the storage device, this would not be the case if the system is DC-connected (as opposed to the dominant AC- connected).

• Relevant operational and environmental parameters: Certain parameters related to the operation of the system for a specific application impact the efficiency, as do environmental factors. These include but are not limited to DoD (or SoC), ambient temperature, and C-rate.

However, some of this information may not be provided and assumptions may be necessary.

2.2.1.4 Other technical considerations

Often, the technical parameters of technologies present trade-offs. For example, Figure 8 from a 2013 publication contrasts efficiency and durability (or lifetime) for several energy storage technologies; while some of this data is outdated, the graph illustrates an important trade-off.

Figure 8. Efficiency vs. durability for various technologies [18]

Moreover, for electrochemical technologies, the voltage affects the quality of the stored energy, with higher voltage energy being more useful than low voltage energy [31]. Some studies comment on the overcharge ability of batteries [17] [32]. This refers to the behavior of the battery when charging continues after reaching full capacity and is relevant because, after all the active material has been converted to the charged state via chemical reactions, the energy received is consumed in the production of gases or for other non-desirable reactions [21]. Conventional cells tend to experience temperature rise when this occurs [21]. Moreover, memory effect is the dramatic shortening of the lifetime and capacity of a battery after it is repeatedly recharged after being partially discharged [1].

2.2.2 Economic

2.2.2.1 Capital cost

One important economic characteristic of a storage device or system is its capital cost or capital expenses (CAPEX), the initial investment required. Specific cost is a useful metric to make the costs of different power ratings or capacities comparable. This can be defined as the cost per unit of power [USD/W] or of energy [USD/Wh] [23] because some of the components in an energy storage system are sized based on power and others based on energy. The components that are included in reported costs can vary greatly, particularly for the soft costs, such as project development and margins. It is estimated that for Li-ion systems, the batteries

(22)

-22-

generally represent only about 60% of the total system cost [33]; this value has been found to be 55% for a 4-hour system but only 23% for a 30-minute system [14].

Figure 9 summarizes two reviews of specific capital costs performed in 2016 [23] and 2012 [34]. While the data captured in these graphs is largely outdated, they show that the specific cost per unit power and per unit energy varies widely across different technologies. Technologies with lower energy-related capital costs tend to be more appropriate for long-duration applications than those with higher specific energy costs.

Figure 9. Specific power and energy capital costs from [23] (left) and [34] (right)

2.2.2.2 Operational cost

The operational costs or expenses (OPEX) express the costs associated with using the system. These are typically expressed as a fixed annual amount based on system power rating [USD/kW], as a variable amount per unit of energy discharged [USD/kWh], or both. OPEX may include insurance.

2.2.3 Other

Other important parameters for storage systems include the environmental impacts (such as sourcing of raw materials or degree of recyclability), and safety (such as toxicity and flammability) [17]. The presence of materials sourced from conflict areas (such as cobalt from the Democratic Republic of Congo) is a serious industry concern, particularly for Li-ion battery chemistries that contain cobalt. The lead time, or the required time to get a system to be operational, may also be relevant for some users or applications.

The authors of [35] have noted that there is a research gap concerning the environmental impact of batteries, and that batteries that have a low environmental impact and a high degree of operational safety are encouraged. The toxicity of materials is an important factor in the environmental impact and operational safety characteristics. Life cycle assessments (LCAs) are an insightful approach to evaluate environmental impacts, but relatively few have been done on energy storage technologies. LCAs of the following four technologies are documented in [36]: aqueous hybrid ion, lithium-iron phosphate with a graphite anode (LFP-C), lithium-iron phosphate with a lithium-titanate anode (LFP-LTO) and sodium-ion with a layered oxide cathode and hard carbon anode. The analysis considered not only the production of the batteries, but also two applications: residential PV and island microgrid. The aqueous hybrid ion battery had the lowest impact in four of the six categories, although its performance is significantly worse than that of the rest for ozone depletion and greenhouse gas emissions. Aspects such as the energy density and the lifetime affect a battery’s impact per unit of storage capacity.

Two other relevant characteristics for any technology are the technology readiness level (TRL) ranging from 1 to 9 and the manufacturing readiness level (MRL) ranging from 1 to 10, both of which are evaluated in [37]. Other similar parameters include the Level of Dissemination (LoD) and the System Readiness Level (SRL). Figure 10 shows the results of a 2017 evaluation of the TRL of energy storage technologies. This study employed the European Commission’s TRL scale [38], while the researchers in [37] used a similar one from the U.S. Department of Energy.

(23)

-23-

Figure 10. TRL of energy storage technologies [39]

2.3 Chemical energy storage technologies

Chemical energy storage includes power-to-gas (such as hydrogen or methane) and power-to-liquids (such as ammonia or methanol). Power-to-gas with hydrogen fuel cells is perhaps the most promising and likely the most utilized of the chemical energy storage technologies.

Hydrogen can be produced via the electrolysis of water utilizing electrical energy, and the electrical energy can be recovered by either burning the hydrogen in a thermal power plant or using it in fuel cells [40]. Water is the only by-product in either case. Electrolysis for hydrogen production has been performed in industrial settings for many years, with large facilities reaching 100 MW capacities. Alkaline electrolyzers are most common, with proton exchange membrane (PEM) and high-temperature ceramic electrolyzers under development [40]. PEMs, however, require a platinum catalyst. Hydrogen storage can involve high-pressure steel or composite tanks, liquefaction, or storing it as a solid within alloys [40].

While some researchers consider fuel cells to be energy converters and not energy storage devices [17], others classify them as energy storage devices [1]. Regardless, a fuel cell converts the chemical energy of a fuel into electrical energy [1] [16]. Even if considered energy converters, fuel cells are included in the academic literature on storage devices because they can be used as a storage device. The production of hydrogen via electrolysis is akin to charging because it consumes electricity. This hydrogen can be stored in a vessel. Later, it can be used to generate electricity during a “discharging” phase [18], also releasing heat in the process [1].

The conversion to electrical energy occurs when the fuel is fed to the anode (the positive electrode) and air or oxygen (oxidants) is fed to the cathode (the negative electrode), resulting in a potential difference between the two electrodes [1]. The reverse process, where electricity is converted to chemical energy, occurs when a current is applied to an electrolyzer, separating water into hydrogen and oxygen. Regenerative or reversible fuel cells can reverse the process because they integrate an electrolyzer and a fuel cell [1].

The anode receives electrons from the fuel, and these electrons travel to the cathode due to the potential difference between the electrodes. In contrast to a battery, a fuel and an oxidant must be supplied, and the flow of electrons is through an external circuit [16]. As an oxidant is supplied to the cathode side of the fuel cell, an exhaust is produced. Figure 11 depicts the general operation of a fuel cell.

(24)

-24-

Figure 11. Schematic of a fuel cell [16]

Fuel cells are compatible with many fuels [1], but hydrogen is often used. A key challenge for this technology is therefore the storage of hydrogen. Four main hydrogen storage technologies are outlined in [1]. The two most mature technologies are pressurized hydrogen and metal hydrides that absorb the hydrogen; the latter requires a thermal management system because heat is released and consumed during the absorption and desorption reactions. The less developed technologies are liquid hydrogen and carbon nanofibers.

Fuel cells have a TRL of 8 [39]. They may be suitable for long-duration storage if an effective storage method is used, but real-world storage projects with hydrogen fuel cells have thus far been very rare. Self-discharge rates range from 0.5-2%/day generally and 3%/day for liquid hydrogen range [39] [1]. However, they have a relatively slow response rate in the minutes range [39] [40]. Efficiencies of 20-70% can be expected [39]

[1] [18] [40]. Lifetimes of 1,000 – 20,000 cycles and 5-15 years have been published [39] [1] [18].

2.4 Electrical energy storage technologies

Electrical energy storage is unique in that it does not require energy conversions because the energy is already stored as electrical energy. It includes capacitors and superconductive magnetic energy storage (SMES).

Capacitance is the ability to store energy in an electric field and capacitors store energy in electrostatic charge [40]. Capacitors form an electric field between its two electrodes and store energy in this field [18]. The two metal plates (the electrodes) are separated by a small air gap that does not allow DC current to pass through.

These plates become electrostatically charged when a voltage is applied; a short circuit between the plates can release this charge [40].

Supercapacitors were introduced in the 1970s to store larger amounts of energy than capacitors [40]. They are also known as ultracapacitors or as electric double layer capacitors (EDLC) [1]. In an electrostatic capacitor, the voltage will eventually increase enough to break down the air and allow electricity to pass through. Placing a dielectric material between the plates increases the amount of charge that can be held.

The dielectric material can be a liquid electrolyte, such as sulfuric acid or potassium hydroxide [40].

SMES, in contrast, creates a magnetic field and stores energy in it. It lets direct current flow in a superconducting coil to create this magnetic field [18]. To increase conductivity and eliminate resistance, the coil is cryogenically cooled, which makes it superconducting [16]. These systems are fast-responding, are used in power ratings ranging from the hundreds of kW to a few MW and have been applied for adjustable speed drives, power quality improvements, and back-up power [16].

2.5 Electrochemical energy storage technologies (batteries)

Electrochemical storage technologies are commonly referred to as batteries. Batteries use a chemical reaction to produce electricity by separating the processes of the reactions with a selective filter [40]. Two reduction- oxidation (redox) half-reactions occur in a battery, one in each half-cell. A negative electrode (the cathode)

(25)

-25-

and a positive electrode (the anode) are associated with each of these half-reactions. The selective filter that separates the two electrodes is called the electrolyte. In these half-reactions, charged ions are created and electrons are either captured (when charging) or released (when discharging). The electrolyte allows the charged ions to travel between electrodes, but the electrons are forced to go through an external circuit to get from one electrode to the other. This produces the electrical current.

Batteries that can only be used once are called primary batteries, while batteries that can reverse the reaction and therefore be recharged are known as secondary batteries [40]. When the battery is charged, the application of an external voltage across the electrodes reverses the reaction [18]. While the chemistries of batteries vary, they share several characteristics. For example, their lifetime depends heavily on DoD [18].

Battery technologies have largely been considered too expensive for energy-intensive applications such as load leveling and arbitrage; they are therefore most used for niche power-oriented applications like frequency regulation and peaker replacement [33]. However, their continued development and decreasing costs, particularly of Li-ion batteries, may change their viability for long-duration applications.

The following sections provide more detailed information on the principal types of batteries suitable for stationary energy storage today. In addition to these, several novel battery technologies are at early stages of research and development, including advanced sodium-metal chloride batteries, high-performance copper- chloride batteries, and lithium batteries with nanostructured materials [35].

2.5.1 Lead-acid

Lead-acid batteries are the oldest battery technology in use today [35], represent approximately half of all secondary cell sales globally and are the most widely available battery worldwide [40]. Lead-acid battery systems for stationary energy storage have a TRL of 8 and an MRL of 9 and will reach the maximum values of 9 and 10, respectively, by 2025 [37]. This battery contains metallic lead (Pb), lead dioxide (PbO2) and sulfuric acid as the cathode, anode, and electrolyte (respectively), all housed in a thermoplastic container [17].

Both flooded and sealed lead-acid batteries are commonly used. Flooded batteries require “topping up” with distilled water, while sealed batteries are nearly maintenance-free [17]. The valve-regulated lead-acid (VRLA) battery and the gelled/absorbed electrolyte-based lead-acid battery are examples of sealed batteries [17] [35].

In general, flooded batteries are less costly, slightly more efficient, have longer operational lives, require more maintenance, and pose higher safety risks, compared to sealed varieties.

While lead-acid batteries are widely used for stationary back-up power and have a response time as fast as under 5 ms [1], their limitations include relatively short lifespans of 200-3,100 cycles depending on the DoD [1] [35] [41] and a maximum (recommended) DoD of 50% [19] [42]. At 80% DoD, they are expected to provide only 900 cycles and last 2.6 years [37]. Calendar lifetime with various cycling patterns can range from 2 to 30 years [17] [40], although lifetimes of approximately 10 years are most cited [41].

The reference AC-AC (system) efficiency for a lead-acid battery according to [37] is 72% and it is also subject to a steep efficiency degradation of 5.40%/y. The efficiency of the battery itself is 85-95% according to [19].

The reported efficiencies ranging from 65-90% in [1], [18], [17] and [35] likely refer to system-level AC-AC efficiency. IRENA reports a reference efficiency of 81% for VRLA and 83% for flooded batteries (and a range of approximately 76-93% for both). Self-discharge is relatively low at 0.1-0.4%/day [41] [39]. Lead- acid batteries may require a thermal management system due to poor performance in low or high temperatures [1] [43] outside its 18-45°C operating range [17].

Several advanced lead-acid batteries, also known as advanced lead-carbon batteries, are under development.

Advanced versions use a combined Pb and modified carbon electrode as the cathode (instead of only Pb);

the battery then acquires qualities of a supercapacitor [44]. The resulting advantages over conventional lead- acid technology include greater charge acceptance (faster charging), faster response times, and longer lifetimes. One advanced lead-acid battery was found to reach up to 17,000 cycles [1]. Another lead-carbon battery achieved a DC-DC efficiency of 92% and an AC-AC efficiency of 85% and it can operate at SoC levels between 30% and 80% [29], although the IFC places the average RTE of advanced lead-acid systems

(26)

-26-

a bit lower at 80% [45]. Moreover, they can operate in freezing temperatures and do not need active cooling, although they exhibit a steep voltage drop when discharging [44].

2.5.2 Lithium-ion

2.5.2.1 Overview

Lithium-ion (Li-ion) batteries are the dominant technology for energy storage today, excluding pumped hydro. In 2017, they represented nearly 70% of the installed capacity according to [46] and 99% of the capacity added in the previous 10 years. Li-ion batteries receive their name from the Li-ion-conducting electrolyte they employ. The cathode typically consists of metal oxides with layered or tunneled structures on an aluminum current collector, while the anode is typically a graphitic carbon on a copper current collector [17]. A schematic of a Li-ion battery is shown in Figure 12.

Figure 12. Schematic of a Li-ion battery [1]

The commercialization of Li-ion batteries began in 1991 when Sony introduced them in consumer electronics [1] and they are now considered to have a TRL of 8 and an MRL of 9 [37]. The advantages of Li-ion batteries include a fast response time—as short as 1 ms [29]—a relatively high efficiency, and rapidly decreasing costs. However, their degradation is highly influenced by DoD and can lead to relatively short lifetimes.

Li-ion batteries have generally been limited to durations of under 6 h [29] and durations under 2 h seem to be common for systems focused on ancillary services and/or customer demand charge management. Li-ion systems with durations of 6 h and 8 h have been installed for a demonstration microgrid in California and for transmission infrastructure deferral on a U.S. island, respectively [46].⁴

2.5.2.2 Categories

Li-ion batteries can be categorized into liquid, semi-solid or solid-state batteries, based on the phase of the electrolyte. Liquid electrolytes are usually lithium salts dissolved in organic carbonates. Semi-solid electrolytes are often a polymer made of lithium salts and high-molecular-weight polymer matrices and may include solvents. Solid-state electrolytes consist of mobile ions and metal and non-metal ions in a crystal structure [17] and they make batteries safer and more energy-dense.

Moreover, Li-ion batteries can also be categorized based on the shape their cells are manufactured into, which includes pouch, cylindrical, and prismatic for those with liquid electrolytes [47]. A comparison of these three shapes and an illustrative example of a solid-state Li-ion battery by [47] showed pouch and cylindrical batteries as the lowest-cost options and pouch and solid-state batteries as the most durable (longest lifetimes). However, it also noted that pouch batteries have poorer safety features than the others.

4 Although IRENA assigns a minimum C-rate of 0.25 for Li-ion technologies in [4], these examples show that longer durations (and therefore lower C-rates) are possible.

Long-duration energy storage: A technoeconomic comparative analysis with case studies in Mexico

Long-duration energy storage: a technoeconomic comparative analysis with case studies in Mexico

Nayeli Gallardo

Abstract

Sammanfattning

Acknowledgments

Abbreviations

List of figures

List of tables

Table of contents

1 Introduction 1.1 Context

1.2 Azelio AB

1.3 Objectives and scope

1.4 Thesis structure

2 Literature review

2.1 Energy storage systems

2.2 Main parameters of energy storage technologies

2.3 Chemical energy storage technologies

2.4 Electrical energy storage technologies

2.5 Electrochemical energy storage technologies (batteries)