Second Life Batteries Facilitating Sustainable Transition in the Transport and Energy Sectors?

INOM

EXAMENSARBETE MASKINTEKNIK,

AVANCERAD NIVÅ, 30 HP ,

STOCKHOLM SVERIGE 2020

Second Life Batteries Facilitating

Sustainable Transition in the

Transport and Energy Sectors?

- An Exploratory Field Study in Colombia

IRIS VESTERBERG

SOFIA WESTERLUND

KTH

Second Life Batteries Facilitating

Sustainable Transition in the

Transport and Energy Sectors?

- An Exploratory Field Study in Colombia

Iris Vesterberg

Sofia Westerlund

2020-06-17

Master of Science Thesis

KTH School of Industrial Engineering and Management

Energy Technology TRITA-ITM-EX-2020:338

Master of Science Thesis TRITA-ITM-EX-2020:338

Second Life Batteries Facilitating Sustainable Transition in the Transport and Energy Sectors?

- An Exploratory Field Study in Colombia

Iris Vesterberg Sofia Westerlund Approved 2020-06-17 Examiner Per Lundqvist Supervisor Dilip Khatiwada Commissioner

Celsia Colombia S.A. E.S.P.

Contact person

Santiago Lemos Cano

Keywords

This study has been carried out within the framework of the Minor Field Studies Scholarship Program, MFS, which is funded by the Swedish International Development Cooperation Agency, Sida.

The MFS Scholarship Program offers Swedish university students an opportunity to carry out two months' field work, usually the student's final degree project, in a country in Africa, Asia or Latin America. The results of the work are presented in an MFS report which is also the student's Bachelor or Master of Science Thesis. Minor Field Studies are primarily conducted within subject areas of importance from a development perspective and in a country where Swedish international cooperation is ongoing.

The main purpose of the MFS Program is to enhance Swedish university students' knowledge and understanding of these countries and their problems and opportunities. MFS should provide the student with initial experience of conditions in such a country. The overall goals are to widen the Swedish human resources cadre for engagement in international development cooperation as well as to promote scientific exchange between universities, research institutes and similar authorities as well as NGOs in developing countries and in Sweden.

The International Relations Office at KTH the Royal Institute of Technology, Stockholm, Sweden, administers the MFS Program within engineering and applied natural sciences.

Katie Zmijewski Program Officer

MFS Program, KTH International Relations Office

KTH , SE-100 44 Stockholm. Phone: +46 8 790 7659. Fax: +46 8 790 8192. E- mail: katiez@kth.se

Acknowledgment

We would like to thank Sida for granting us the Minor Field Studies (MFS) scholarship which enabled the field study in Medellín, Colombia. Our greatest gratitude goes to all the people who have helped us along the way during this project, both with guidance and information as well as accommodation and transport. First and foremost, we would like to thank Celsia Colombia S.A. E.S.P., and especially our contact person and supervisor Santiago Lemos Cano, for allowing us collaborate with you for this thesis. Special thank you to everyone working at Celsia’s Medellín office for your contribution to this thesis and for making us feel welcome at your office and in Medellín. It was a really great experience, even though it unfortunately came to an abrupt end in advance due to the Coronavirus-pandemic.

Thank you to all interviewees from the pre-study and the field study for your time and valuable knowledge. We would also like thank our supervisor at Kungliga Tekniska Högskolan (KTH), Dilip Khatiwada, for his guidance and valuable advices that helped us form the thesis. We are also grateful to Malin Lönnqvist for being our travel and housing partner, as well as Spanish translator, in Colombia. We would also like to thank Magnus Lindqvist for introducing us to Celsia and Colombia.

Iris Vesterberg and Sofia Westerlund

Abstract

The increasing number of vehicles in Colombian cities have resulted in alarmingly low quality of air, further resulting in increasing health issues. One potential solution to this issue could be a shift from ICEVs (internal combustion engine vehicles) to EVs (electric vehicles). However, EVs in Colombia are still very expensive, an issue that needs to be addressed in order for the EV market to increase enough to be able address the issue of low air qual ity in cities. One way of overcoming these cost barriers could be through implementation of a market for SLB (second life batteries), meaning that a battery retired from usage in EVs would be remanufactured, resold and reused in another application. Through SLB, the owner cost of EVs could potentially be decreased. SLB could also help improve the case for non-dispatchable renewable energy sources by providing low cost BESS (battery energy storage solutions). Thus, SLB has the potential to facilitate sustainable transition within both the transport and the energy sector.

This thesis aims to assess the potential of SLB in Colombia. This is done through a literature review where the current state of SLB is investigated, several interviews with potential stakeholders for a SLB market in Colombia, and a techno-economic assessment of four potential BESS applications in Colombia. The literature review provides with current knowledge and state of SLB in general. The interviews provide important insight to potential stakeholders’ view on SLB for the specific case of Colombia. The techno-economic assessment includes a sensitivity analysis aiming to provide insights in which factors, such as e.g. battery purchasing price or charging cost, that that gives rise to the largest impact on feasibility of SLB.

Findings from the interviews shows a strong collective commitment from the interviewees to working towards cleaner air, resulting in high engagement and collaborative efforts between stakeholders for the SLB case. The main issue highlighted by stakeholders regards techno-economic uncertainties of SLB. Findings from the techno-techno-economic assessment indicates that SLB is viable for larger applications such as BESS at solar farms, but not for smaller applications such as backup power in residential buildings. However, SLB is not deemed to be a game changer for either application, and there are still many uncertainties regarding both technological and economic aspects that needs to be further investigated. The sensitivity analysis shows that the factors resulting in the highest impact on feasibility of SLB is battery SOH (state of health) at the beginning of SLB usage, and battery and repurposing cost. It will be hard to address both of these factors simultaneously due to a higher SOH would render higher battery prices, and vice versa.

Sammanfattning

Det ökande antalet fordon i colombianska städer har resulterat i oroväckande låg luftkvalitet, vilket ytterligare resulterat i ökande hälsoproblem. En potentiell lösning på det problemet kan vara en övergång från ICEVs (förbränningsmotorfordon) till EV (elfordon). EVs i Colombia är fortfarande väldigt dyra, en fråga som måste adresseras för att EV-marknaden ska kunna öka tillräckligt för att kunna ge en inverkan på problemet med låg luftkvalitet i städer. Ett sätt att övervinna dessa kostnadshinder skulle kunna vara genom att implementera en marknad för SLB (second life-batterier), vilket innebär att ett batteri som bedömts inte längre uppfylla kraven för användning i EVs, och därmed byts ut, skulle kunna byggas om, säljas vidare och återanvändas i andra applikationer. Genom SLB kan ägarkostnaderna för EVs potentiellt sänkas. SLB skulle också kunna användas för att tillhandahålla billigare BESS (batterilagringslösningar) hos icke-reglerbara förnyelsebara kraftverk, såsom solkraftverk. Således har SLB potentialen att underlätta för hållbara förändringar inom både transportsektorn och energisektorn.

Den här uppsatsen ämnar att utvärdera SLBs potential i Colombia. Detta görs genom en litteraturöversikt där det nuvarande tillståndet av SLBs undersöks, flera intervjuer med potentiella intressenter för en SLB-marknad i Colombia, och en tekno-ekonomisk bedömning av fyra potentiella BESS-applikationer i Colombia. Litteraturöversikten samlar aktuell kunskap och status inom SLB i allmänhet. Intervjuerna ger viktig insikt om potentiella intressenters syn på SLB för det specifika fallet i Colombia. Den tekno-ekonomiska bedömningen inkluderar en känslighetsanalys som syftar till att ge insikter i vilka faktorer, som t.ex. batteriets inköpspris eller laddningskostnad, som ger upphov till den största effekten på SLBs genomförbarhet.

Resultat från intervjuerna visar ett starkt kollektivt engagemang från de intervjuade att arbeta mot renare luft, vilket resulterar i högt engagemang och samarbete mellan intressenterna. Det största problemet med SLB från intressenternas synpunkt berör tekno-ekonomiska osäkerheter. Resultat från den tekno-ekonomiska bedömningen indikerar att SLB är ekonomiskt försvarbart för större applikationer som BESS vid solkraftverk, men inte för mindre applikationer som t.ex. för reservenergi i bostadshus. SLB anses dock inte vara ett genombrott för användning vid någon av applikationerna, och det finns fortfarande många osäkerheter när det gäller både tekniska och ekonomiska aspekter som måste undersökas ytterligare. Känslighetsanalysen visar att de faktorer som resulterar i den högsta påverkan på genomförbarheten av SLB är batteriets SOH (hälsotillstånd) i början av SLB-användning och kostnaden för batteri och ombyggnad av batterier. Det kommer dock att vara svårt att hantera båda dessa faktorer samtidigt på grund av att högre SOH skulle ge högre batteripriser, och vice versa.

Table of content

Acknowledgment ... iv

Abstract ... v

Sammanfattning ... vi

Table of content ... vii

List of figures ... x

List of tables ... xi

Nomenclature ... xii

1 Introduction ... 1

1. 1 Background ... 1

1.1.1 Why electric vehicles ... 1

1.1.2 EV batteries ... 2

1.1.3 Battery value chain ... 5

1.1.4 Second life batteries ... 6

1.1.5 Colombia ... 7 1.1.6 Towards SLB in Colombia ... 9 1.2 Research questions ... 10 1.3 Contribution ... 11 1.4 Delimitations ... 11 1.5 Thesis outline ... 12

2 Second life batteries: An overview on Technical, Economic and Business aspects ... 13

2.1 Types of batteries used in EV ... 13

2.1.1 Lithium Iron Phosphate batteries ... 13

2.1.2 Lithium Nickel Manganese Cobalt Oxide batteries ... 14

2.1.3 Lithium Nickel Cobalt Aluminum Oxide batteries ... 15

2.1.4 Comparison between LFP, NMC and NCA ... 15

2.2 Technical issues and aspects of SLB ... 16

2.2.1 The Battery Life Cycle - from first life to second life ... 17

2.2.2 Battery ageing phenomena and patterns ... 17

2.2.2.1 Battery ageing as a degradation mode... 18

2.2.2.2 Battery ageing as cycle or calendar ageing... 19

2.2.2.3 Battery ageing visualized by “the ageing knee” ... 19

2.2.3 Battery state of health ... 20

2.2.4 Repurposing batteries for a second life ... 22

2.3 The SLB business case ... 23

2.3.1 Economics - Cost structure of SLB ... 24

2.3.2 Applications for SLB - business opportunities ... 25

2.3.3 Commercial and pilot projects ... 27

2.4 Recycling of batteries... 28 3 Method ... 30 3.1 Research design ... 30 3.2 Data collection ... 31 3.2.1 Literature review ... 31 3.2.2 Interviews ... 32 3.2.3 Archival data ... 34 3.2.4 Secondary data ... 34 3.2.5 Observations ... 35

3.3 Theoretical approach to analysis ... 35

3.3.1 Stakeholder theory ... 35 3.3.2 Sustainable transitions ... 36 3.4 Data analysis ... 37 3.4.1 Qualitative analysis ... 37 3.4.1.1 Coding procedures... 38 3.4.1.2 Stakeholder analysis ... 38

3.4.1.3 Transition theory analysis ... 38

3.4.2 Quantitative analysis ... 39

3.4.2.1 Levelized cost of storage ... 39

3.4.2.2 Case descriptions ... 39

3.4.2.3 Data sources and assumptions ... 40

3.5 Research Quality ... 40

4 Findings and Analysis ... 42

4.1 Case settings and preliminary findings ... 42

4.1.1 Entrance to EV market ... 42 4.1.2 Commercial pilot of SLB ... 43 4.2 Stakeholder perspective of SLB ... 43 4.2.1 Socio-technical perspective ... 45 4.2.2 Techno-economic perspective ... 46 4.2.3 Political perspective ... 48 4.2.4 Integrated perspective ... 49 4.3 Techno-economic assessment ... 50 4.3.1 Calculations ... 50 4.3.1.1 Investment cost ... 51 4.3.1.2 N - Lifespan in years ... 51 4.3.1.3 r - discount rate ... 51 4.3.1.4 O&M cost ... 52 4.3.1.5 Charging cost ... 52 4.3.1.6 EOL cost ... 52 4.3.1.7 Electricity Discharged ... 53 4.3.2 Result LCOS ... 53 4.3.3 Sensitivity analysis ... 54 5 Discussion ... 59

5.1 Stakeholder perspective and collaboration towards SLB ... 59

5.1.1 Why SLB is important ... 59

5.2 Techno-economic assessment ... 61

5.2.1 Assessment of SLB... 62

5.2.2 Factors impacting SLB techno-economic viability ... 62

5.2.2.1 High risk techno-economic factors ... 63

5.2.2.2 Mediate risk techno-economic factors ... 64

5.2.2.3 Low risk techno-economic factors ... 65

5.3 SLB facilitating sustainable transition ... 66

5.3.1 How facilitation is achieved ... 66

5.3.2 Sustainability impact ... 67

6 Conclusions and recommended future work ... 69

6.1 Conclusions ... 69

6.2 Implications ... 69

6.2.1 Implications for academia ... 70

6.2.2 Implications for industry ... 70

6.3 Recommended future work ... 70

References ... 72

Appendices ... 79

Appendix I - Summary of the of the SLR ... 79

Business perspective ... 79

Technical perspective ... 81

Appendix II - Interviews... 85

Pre-study interviews ... 85

Field study interviews ... 85

Appendix III - Data sources and assumptions for calculations of LCOS ... 88

Data sources ... 88

List of figures

Figure 1: Global long-term passenger vehicle sales by drivetrain (Bloomberg NEF, 2019a). ... 2

Figure 2: EV battery composition: cell, module, and pack (Farag, 2013). ... 3

Figure 3: Overview of the automotive lithium-ion battery value chain (Lebedeva, Di Persio and Boon-Brett, 2016). ... 6

Figure 4: EV battery life cycle with second use application (NREL, 2015). ... 6

Figure 5: Map of Colombia (Central Intelligence Agency, 2020). ... 8

Figure 6: Illustration of how SLB could affect EV cost and sales. ... 10

Figure 7: Chemical reaction of a LPF cell (Toprakçı et al., 2011). ... 14

Figure 8: Comparison of LIB cells (BIS Research, 2019). ... 16

Figure 9: Potential battery capacity for first and second life applications... 17

Figure 10: Battery degradation scheme (Birkl et al., 2017). ... 18

Figure 11: Illustration of capacity decrease and the ageing knee. ... 20

Figure 12: Resistance growth and capacity fade as indicators affecting second use lifetime (Neubauer et al., 2015). ... 21

Figure 13: Technical parameters of relevance when repurposing batteries for a second life (Klör et al., 2016). ... 22

Figure 14: Decision making flow diagram for EV batteries and scheme of potential circular economy paths (Canals Casals, Garcia and Cremades, 2017). ... 25

Figure 15: The original stakeholder model (Freeman, 1984). ... 35

Figure 16: Illustration of the three perspectives and their relation (Cherp et al., 2018). ... 37

Figure 17: Cases for LCOS calculations. ... 40

Figure 18: Stakeholder Model. ... 43

Figure 19: SWOT for the case of SLB in Colombia. ... 50

Figure 20: Risk matrix - Impact on LCOS when varying a factor, level of reliability between estimated and actual values, and agency of factors for SLB intermediary. ... 63

List of tables

Table 1: Important battery parameters to consider when matching battery types and applications (Liu et

al., 2019; Osmanbasic, 2019; Koniak and Czerepicki, 2017)... 3

Table 2: Key characteristics of EV battery types (Liu et al., 2019). ... 5

Table 3: Comparison of LIB cells (Koniak and Czerepicki, 2017). ... 16

Table 4: Estimated viable market price ranges for SLB with different approaches. ... 25

Table 5: Estimated cost for SLB repurposing. ... 25

Table 6: Second life projects (commercial and pilots). ... 27

Table 7: First search, literature review approach for SLB. ... 32

Table 8: Second search, literature review approach for SLB. ... 32

Table 9: Pre-study interviews. *Interviews held online. ... 33

Table 10: Field study interviews. *Interviews held online. ... 34

Table 11: Secondary data: non-academic papers. ... 34

Table 12: Case descriptions. ... 39

Table 13: Scenario descriptions. ... 40

Table 14: Identified stakeholders. ... 44

Table 15: Results for each parameter from Equation 4. ... 53

Table 16: LCOS for cases A and B when varying different input parameters. ... 54

Table 17: LCOS for cases C and D when varying different input parameters. ... 56

Table 18: Summary of analyzed publications from the first part of the SLR. ... 80

Table 19: Summary of analyzed publications from the second part of the SLR. ... 84

Table 20: Pre-study interviews. ... 85

Table 21: Field study interviews. ... 87

Table 22: Data sources for values used in LCOS calculations. ... 88

Nomenclature

Symbols used in equations Denomination Unit

Charging cost Yearly charging cost of BESS USD/year

e- Negative electrode -

Electricity Discharged Yearly electricity discharged from BESS kWh/year

EOL cost Cost related to EOL of BESS USD

FePO4 Iron phosphate -

Investment cost Investment cost of BESS USD

LCOS Levelized Cost Of Storage USD/kWh

Li+ Positive lithium electrode -

LiFePO4 Lithium iron phosphate -

LiNiCoAlO2 Lithium nickel cobalt aluminum oxide -

LiNiMnCoO2 Lithium nickel manganese cobalt oxide -

N Final year of BESS Year

n Current year of BESS Year

NiCoAlO2 Nickel cobalt aluminum oxide -

NiMnCoO2 Nickel manganese cobalt oxide -

O&M cost Yearly O&M cost of BESS USD/year

r Discount rate %

Abbreviations Subscript

Al Aluminum

B2U Battery Second Use

BESS Battery Energy Storage Solutions

BEV Battery Electric Vehicle

BMS Battery Management System

CAPEX Capital Expenses

CCET Celsia Commercial Electromobility Team

Co Cobalt

CRM Critical Raw Material

DCFC Direct Current Fast Charge

DOD Depth Of Discharge

DSS Decision Support System

ESS Energy Storage Solutions

EEST Electrical Energy Storage Technologies

EOFL End Of First Life

EOL End Of Life

ESS Energy Storage Solutions

EV Electric Vehicle

Fe Iron

GDP Gross Domestic Product

GHG Green House Gas

ICE Internal Combustion Engine

ICEV Internal Combustion Engine Vehicle

JV Joint Venture

LAMNE Loss of active cathode material

LAMPE Loss of active anode material

LCOE Levelized Cost Of Electricity

LCOS Levelized Cost Of Storage

LFP Lithium Iron Phosphate

Li Lithium

LIB Lithium-ion battery

LLI Loss of Lithium Inventory

KTH Kungliga Tekniska Högskolan (The Royal Institute of Technology)

MaaS Mobility as a Service

MFS Minor Field Studies

Mn Manganese

MRQ Main Research Question

NB New Batteries

NCA Lithium Nickel Cobalt Aluminum Oxide

NiMH Nickel-Metal Hydride

NMC Lithium Nickel Manganese Cobalt Oxide

NPV Net Present Value

NREL National Renewable Energy Laboratory

OECD Organisation for Economic Co-operation and Development

OEM Original Equipment Manufacturer

O&M Operation And Maintenance

PbAc Lead-Acid

PHEV Plug-in Hybrid Electric Vehicle

PV Photo Voltaic

R&D Research And Development

RQ Research Question

RUL Rest of Useful Life

SEI Solid Electrolyte Interface

SLB Second Life Batteries

SLB-EES Second Life Battery - Energy Storage Solution

SLR Systematic Literature Review

SOC State Of Charge

SOH State Of Health

SWOT Strengths, Weaknesses, Opportunities, Threats

TCO Total Cost of Ownership

UPS Uninterruptible Power Supply

1 Introduction

The global threat of climate change has lately driven a shift in the transportation sector, from vehicles operated by fossil fuels to electric vehicles (EV). The electric fleet now comprise of more than five million vehicles globally and is expected to grow rapidly in the coming years (IEA, 2019a). It has actuated a new environmental challenge: batteries, and what to do with them once they no longer meet required specifications for EV usage.

Harper et al. (2019) highlights the battery challenge by providing an example: the EV’s sold in 2017 will result in 250,000 tones of battery packs entering our waste streams. From an environmental perspective, this is a two-fold problem. Firstly, battery manufacturing requires sourcing of rare materials and demands large amounts of energy, thus giving rise to a considerable environmental footprint. Secondly, battery waste is hazardous and should not just be discarded into landfills (New Hampshire Department of Environmental Services, 2017). However, current research has indicated there is potential for retired EV batteries to add value in other sectors. Second life batteries (SLB), batteries retired from their first life application and remanufactured for a second life in another application, can for instance be used in battery energy storage solutions (BESS).

Colombia is leading the shift to electric mobility in the Latin American region, with more than thousand EVs, including both light-duty vehicles and buses, in operation (UN Environment, 2018). With 2.5 million inhabitants, Medellín is the second largest city in Colombia. According to former Mayor Gutierrez of Medellín, Medellín aims “to be the capital of electric mobility in Latin America” (Moloney, 2019). Celsia, a Colombian energy company, has provided Medellín with a network of nine public charging stations and is now interested in turning SLB into a new business opportunity for the energy sector in Colombia (UN Environment, 2018; Interview 1, 2020). However, due to techno-economic uncertainties of SLB, the business opportunity needs to be evaluated given the Colombian context. Moreover, if SLB is to be implemented, the sustainability implications need to be investigated.

1. 1 Background

The background aims to shortly introduce the relevance of the study, as well as the main topics relevant for the thesis. In this case EVs and their batteries, the battery value chain, SLB, and how they relate to Colombia.

1.1.1 Why electric vehicles

The electric vehicle was invented 100 years ago, but was out-competed by the internal combustion engine vehicle (ICEV) (U.S. Department of Energy, 2015). However, the innovation of the lithium-ion battery, allowing for higher energy density and longer cycle life, coupled with the climate urgency of developing fossil-free alternatives, has accelerated the latest development and interest of EVs. Raised awareness of carbon emissions, increased oil prices and government incentives promoting EV usage have resulted in EV usage increasing daily. However, EVs do not only hold environmental value for customers, there are also techno-economic benefits. For instance, EV motors are more efficient than motors in ICEVs (U.S. Department of Energy, 2019). Moreover, EVs are more cost-efficient during usage due to lower fuel and maintenance costs (Sivak and Schoettle, 2018). Therefore, projections show EV will reach break-even with ICEV of total cost of ownership (TCO) by year 2025 (Baik et al., 2019). Indeed, in a near future EV will become superior to ICEV.

The sales of EVs have rapidly increased in the last years. According to the Bloomberg NEF 2019 Electric Vehicle Outlook report, the growth of the EV market will continue. They estimate that the annual passenger EV car sales will rise to 10 million in 2025, 28 million in 2030, and 56 million by 2040. Estimations of future vehicle sales for internal combustion engine (ICE) vehicles, plug-in hybrid electric vehicles (PHEV) and battery electric vehicles (BEV) are illustrated in Figure 1 below. Accordingly, there are expectations that EVs will account for 57 percent of all passenger vehicle sales and 30 percent of the global vehicle fleet by the year 2040.

Figure 1: Global long-term passenger vehicle sales by drivetrain (Bloomberg NEF, 2019a).

1.1.2 EV batteries

Figure 2: EV battery composition: cell, module, and pack (Farag, 2013).

There are several different types of EV batteries currently available at the market. Different batteries are more or less suitable for different vehicles, with no ideal universal battery currently existing. In order to optimally match battery types and vehicles, a few battery parameters need to be considered. These include, but are not always limited to, the parameters presented in Table 1 below.

Battery parameter Notes

Weight and volume Some applications have higher tolerance for larger weights and volumes than others. The energy density per weight or volume differs significantly between different types of EV batteries. Applications with high loads may need batteries with a high specific power, i.e. batteries with a high power capacity per kg. Applications in need of storing large amounts on energy may instead need batteries with a high specific energy, i.e. batteries that can store a large amount of energy per unit mass or volume.

Temperature This parameter is affected by climate conditions and heating and cooling abilities. The performance of many batteries depends on temperature, with some battery types operating better in extreme temperatures than others.

Maximum charge and discharge rate

Some applications are in higher need of fast charge and discharge rates, meaning that they need higher battery working currents. The working current is specified as a multiplier of C number, where e.g. batteries with 1C are able to fully charge or discharge the battery in one hour and batteries with 5C are able to fully charge or discharge the battery in one fifth of an hour, i.e. 12 minutes.

Life span How many work cycles, years or how long distance the battery is estimated to last. This also depends highly on several usage factors, such as operating conditions, depth of discharge (DOD), and charging and discharging currents.

Cost EV battery costs are still high compared to ICE vehicles, and may influence the battery choice.

Safety The battery needs to be well monitored and have a good battery management system (BMS). The safety issue may be higher for some applications than other.

Environmental impact

The manufacturing process and recycling possibilities of some battery types have higher environmental impact than others, which may be more or less important to different consumers.

The main EV battery types include lead-acid (PbAc), nickel-metal hydride (NiMH), and lithium-ion.

The most mature and reliable battery technology is lead-acid, or PbAc, batteries. It is the earliest type of rechargeable battery and was invented in 1860 by the French physicist Gaston Planté (Kurzweil, 2010). This battery type is the cheapest, and was until recently the most commonly used battery type in EVs due to its mature technology, high availability and low cost. PbAc batteries are heavy, and usually end up accounting for as much as 25-50 percent of the vehicle’s total mass when used for EV applications (Osmanbasic, 2019). EVs using PbAc batteries tend to have a higher total mass than ICEVs. The high weight leads to a very low energy-to weight and energy-to-volume ratios. However, the batteries still have a relatively large power-to-weight ratio due to their ability to supply high surge currents (Osmanbasic, 2019). PbAc batteries have, like all battery types, an environmental impact during their construction, use, and disposal or recycling process. This battery type is deemed to pose a threat to human health and the environment if not disposed properly. This is due to the fact that the main components of the batteries are sulfuric acid and lead, which both can contaminate solid and ground water (University of North Carolina, 2009). Sulfuric acid is also corrosive, and lead is linked to negative health effects. PbAc batteries have relatively low purchase and installation costs compared to newer technologies, but their life-time value is lower (Osmanbasic, 2019). The PbAc battery technology for EVs is today widely considered to be obsolete and is commonly not used in newer EV designs.

Nickel-metal hydride batteries, or NiMH batteries, are considered to be a relatively mature technology. They are less efficient than PbAc batteries when it comes to charging and discharging, but they have a much higher specific energy allowing for lighter and smaller batteries. The high heat generation rate during fast charging and discharging however means that the batteries require a cooling system that increases the battery’s weight and cost (Osmanbasic, 2019). Other disadvantages to NiMH batteries includes difficulties with performance in colder temperatures, low efficiency, finicky charge cycles, and high self-discharge (Cobalt Institute, 2017; Deevi and Zhang, 2001; Osmanbasic, 2019). An upside of the NiMH battery is the proven long lives when used correctly. In hybrid cars they have been proven to operate well after as long as ten years and 160,000 km (Osmanbasic, 2019). Another upside is the low levels of toxic material in the batteries and the fact that they are recyclable, making them a better environmental option than the PbAc batteries (Deevi and Zhang, 2001). NiMH batteries have been used in most first-generation hybrids and was until around 2010 the main battery technology used in hybrid vehicles (Delucchi and Lipman, 2010). Several legal disputes regarding patents have slowed down the development of NiMH batteries, allowing the focus to shift to the Lithium-ion battery technology (Paine, 2006). NiMH batteries are still used in some hybrid vehicles, but have recently been surpassed by lithium batteries when it comes to all-electric and plug-in hybrid vehicles (Delucchi and Lipman, 2010; Osmanbasic, 2019).

Motors, 2008). The Roadster was also the first car that could travel over 320 km per charge (Shahan, 2015). The cathode of most LIBs is made of lithium cobalt oxide while the anode consists of graphite in combination with varying metals. This allows for the LIBs to achieve many benefits compared to other battery types, such as lower weight, a longer life span, lower self-discharge rate, and higher density, cell voltage and specific energy rate (Osmanbasic, 2019). A LIB is one-third of the weight of a PbAc battery, while also being three times more powerful with a three times as long lifespan (Osmanbasic, 2019). The largest disadvantage of LIBs is their high price. Their production cost have however decreased largely the last years (McKinsey & Company, 2017). Another issue with the LIB technology is safety. An advanced BMS is needed in order to avoid fires or explosions caused by overheating within the battery (Osmanbasic, 2019). Despite of these issues, LIBs are still the superior option for EVs today and are currently the most commonly used batteries for EVs.

Table 2 below illustrates some key characteristics of the different battery types. This clearly show that LIB is the superior available battery type, explaining why it is the most commonly used battery type in EVs.

Battery type Life span [cycles] Nominal voltage [V] Specific energy [Wh/kg] Specific power [W/kg] Charging efficiency [%] Self- discharge rate [%/month] Charge temp. [°C] Discharge temp. [°C] PbAc 200-300 2.0 30-50 180 50-95 5 -20 to 50 -20 to 50 NiMH 300-600 1.2 60-120 250-1,000 65 30 0 to 35 -20 to 65 LIB 600-3,000 3.2-3.7 100-270 250-680 80-90 3-10 0 to 45 -20 to 60

Table 2: Key characteristics of EV battery types (Liu et al., 2019).

That LIBs currently are the most commonly use battery type in EVs is a well-known fact. LIBs are accounted for as much as 90 percent of the EV market year 2016 (Pillot, 2015).

1.1.3 Battery value chain

Figure 3: Overview of the automotive lithium-ion battery value chain (Lebedeva, Di Persio and Boon-Brett, 2016).

Drabik and Rizos (2018) have identified a potential seventh segment to the automotive LIB value chain: second-life applications. This segment, if implemented, would occur after the EV manufacturing and before the recycling. Drabik and Rizos (2018) stresses the importance of considering second-life applications when interpreting the value chain of EV LIBs. LIBs contains materials that are considered by the European Commission to be critical raw materials (CRM) (European Commission, 2017). The European Commission defines CRMs as raw materials that are both of high economic importance for the EU and vulnerable to supply disruptions. One strategy suggested by the European Commission for securing the supply of CRMs includes boosting resource efficiency. Increased resource efficiency for LIBs from the EV sector can thus be achieved by using them in second-life applications, decreasing the demand for new batteries.

1.1.4 Second life batteries

According to the National Renewable Energy Laboratory (NREL) (2015) second application strategies, SLB, for batteries from the EV sector could help overcome the lithium-ion cost barriers that both EVs and BESS currently are facing by providing additional value to EVs and decreasing the cost of BESS. SLB are batteries from the EV sector which once deemed to not be appropriate for their first use in EVs are redeployed for a second applications, such as BESS. A simplified life cycle of LIBs used first in EVs and then for a second use have been illustrated in Figure 4 below.

SLB strategies in order to reduce the battery cost impact of an EV was proposed already in 1990 during the development of California’s first zero emissions vehicle (ZEV) (Neubauer et al., 2015). This inspired several studies regarding SLB, the first being Argonne National Laboratory’s “Electric Vehicle Battery 2nd Use study,” conducted for the United States Advanced Battery Consortium in 1998 (Neubauer et al., 2015). The study by Argonne National Laboratory focused on NiMH batteries and the results showed that the used EV batteries potentially could be used for other energy storage applications than vehicles, motivating further research within the area.

According to Neubauer et al. (2015) at NREL, the interest in SLB faded in 2003 after changed ZEV requirements in California, shifting the focus to hybrid EVs. When California again changed their ZEV requirements in 2008 focus shifted back to fully electric vehicles and the interest in SLB increased again, resulting in many new studies within the field. The battery technology focus had by now shifted from NiMH batteries to LIBs.

The SLB research area is thus novel and incused with many uncertainties, however, has seen a boosted interest from researchers during recent years as battery development and EV growth have been progressing. The major issues with SLB are of techno-economic nature, where previous research not yet has been able to conclude fundamental aspects (Hossain et al., 2019; Martinez-Laserna et al., 2018a). The following aspects have been found while reviewing the techno-economic performance of SLB (see Sections 2.2 and 2.3 for details):

1. Lack of knowledge of ageing pattern of batteries. Why and how do batteries degrade? 2. Lack of standardization. There are several types of cell standards and battery pack

configurations, leading to difficulties in developing SLB manufacturing processes. 3. Uncertain profitability. Given the rapid battery development, can SLB become

cost-competitive?

Despite the techno-economic uncertainties and issues, researchers are intrigued by the prospects of SLB enabling for sustainable development. It has been suggested that SLB can contribute in several ways in a future sustainable society, by increasing resource-efficiency and possibly enabling for both renewable energy and EV (Olsson et al., 2018). To be able to realize the value of SLB, there is need for further research with business aspects of SLB (see Section 2.3), which include:

1. SLB applications. Which showcase the most potential?

2. Business models for SLB. Indications show there is a need for developing collaborative business models, connecting diverse sectors throughout the battery value chain. 3. Governmental support and regulation might be needed if SLB business models and

markets are ever to succeed.

1.1.5 Colombia

America and the Caribbean, and the third by population. The economy of Colombia is as of 2019 the fourth largest in Latin American and the 38th largest in the world as measured by gross domestic product (GDP) (International Monetary Fund, 2019).

Figure 5: Map of Colombia (Central Intelligence Agency, 2020).

Colombia has made major economic and social developments in the recent years (OECD, 2015). In the first quarter of 2014, Colombia was the second fastest growing economy in the world, after China, with an economic growth of 6.4 percent (Curacao Chronicle, 2014). The country is considered to be an emerging market economy by most major analyst groups, meaning that the country’s market has some characteristics of a developed country without fully meeting its standards (MSCI, 2020; MSCI, 2014). Since 2018 Colombia is a member of the Organisation for Economic Co-operation and Development (OECD). This makes Colombia the organisations 37th member country and the third member country from the Latin America and Caribbean region, joining Chile and Mexico (OECD, 2018). The economic growth in combination with improved access to education and social transfers has contributed to increased living standards, fallen poverty levels and increased social conditions in Colombia (OECD, 2019).

Colombia have a diverse and strong economy, traditionally based on agriculture but nowadays in combination with a well-developed industry as well (ui landguiden, 2019). Colombia is currently highly dependent on energy and mining, as well as oil and coal exports (Central Intelligence Agency, 2020; ui landguiden, 2019). However, the country’s development have historically been hampered by a lot of violence due to over 50 years of civil wars, drug trafficking and smuggling. Later years peace processes have led to a more stable and growing economy (ui landguiden, 2019). The then sitting president, Juan Manuel Santos, received the Nobel Peace Prize in 2016 after negotiating a peace with the FARC guerrilla, ending an over 50 years long civil war (The Nobel Prize organization, 2016).

development in the world have also led to increased amount of sold cars. Monthly private car sales in Colombia have increased with almost 200 percent since year 2004 (Trading Economics, 2020).

The increased electricity demand as well as car sales increases the incentives to develop the country’s renewable energy sources, e.g. by using BESS, and to promote EVs in order to reduce Colombia’s GHG emissions. Both renewables and electric transportation has been incentivized by Colombia’s government as a way for Colombia to meet the Paris Climate Agreement (Interview 1, 2020). Colombia has optimal conditions for several non-dispatchable energy sources, such as wind and solar power, and are, as mentioned, heavily dependent on their energy export (Central Intelligence Agency, 2020; Currie, 2016). By utilizing BESS, Colombia could better make use of this opportunity by making these energy sources dispatchable. The increased amount of cars in Colombia, specifically in the cities, have led to issues with high levels of traffic and air pollutants in the cities. The air pollutant issue could be decreased by increasing the use of EVs in Colombia, and thereby decreasing the usage of ICEVs. By July 2019 Colombia had 22,873 registered EVs, including all kinds of electrified vehicles and not only cars, but only 203 of these were fully electrified vehicles (Bland, 2019). The Colombian government have implemented some benefits for EV owners in an attempt to increase the country’s EV fleet. These include tax benefits, preferential public parking, and exemption from the Pico y Placa circulation restriction, a restriction policy aiming to mitigate traffic congestion by restricting usage of cars based on their license plates during certain hours (Viscidi, 2016).

1.1.6 Towards SLB in Colombia

Celsia Colombia S.A. E.S.P. is an electricity company, part of the Colombian conglomerate Grupo Argos. They have operations in Colombia, Costa Rica, Honduras and Panama. As an electricity company, their operations consist of power generation and distribution. Five years ago, Celsia supplied almost 600 thousand customers with electricity. Through an aggressive acquisition strategy they expanded their service to new areas, and as of 2020 they provide electricity to almost 1.15 million customers (Interview 1, 2020). Their operative headquarters are in Yumbo, Valle del Cauca, Colombia.

As part of its strategy, Celsia has been developing innovative products and services focused on improving the quality of life of its clients and guaranteeing the sustainability of cities in the long term, and therefore, it has identified the electrification of transport as a good opportunity to advance towards this end (Interview 1, 2020). While there are many barriers to EV, the main barrier is the price of EVs, which is still too expensive compared to ICEV. EV business models thus need to address the price issue in order to be successful. One way of overcoming the price barriers, is to increase the profitability of the battery by expanding its lifetime. Hence, if SLB is implemented the price of EV can be reduced. The purpose of SLB can therefore be considered a potential enabler for EV, and by extension increased electricity sales. At the same time, the very existence of SLB is dependent on the success of EV.

Figure 6: Illustration of how SLB could affect EV cost and sales.

1.2 Research questions

Celsia is taking initiative to commence one of the first commercial pilots of SLB in the world which will take place in Medellín, Colombia. While SLB has been uplifted in the context of enabling sustainability, it has not been clarified how. It has although been implied that SLB can enable sustainability within several sectors and aspects of society, thus indicating broader transformation, i.e. sustainable transition. More importantly, SLB as enabling sustainability has been merely theoretically suggested, and not substantiated with empirical findings. As such, the initiative taken by Celsia towards SLB, constitutes an interesting research opportunity to further validate if and how SLB can facilitate a sustainable transition, through empirical evidence.

Thus, to explore this, the main research question (MRQ) of this thesis is:

MRQ: How can a SLB initiative facilitate sustainable transition within the transport and energy

sectors?

To be able to answer this question, attention must first be brought to SLB. Due to the novelty of SLB, current research and empirical findings are scarce, which creates many uncertainties which needs to be investigated, as described in Section 1.1.4. Moreover, most research has been conducted in Europe, China and the U.S. and it is ambiguous if conclusions drawn from these studies is applicable in the Latin American region. The lack of studies within the Latin American region coupled with the inherent uncertainties of SLB, thus raises the need for contextual evaluation as the commercial pilot of SLB is progressing.

RQ1: Why do stakeholders perceive SLB as important, and which issues and

properties/dimensions are emerging from the SLB initiative?

RQ2. From a techno-economic perspective, what is the assessment of SLB and which factors

give rise to the largest impact of feasibility?

The research questions are addressed by conducting a field study, thus realizing the case identified in Medellín, Colombia. To ensure in-depth knowledge, mixed methods will be used, described further in Chapter 3.

1.3 Contribution

The overall aim of this thesis, to explore how SLB can facilitate sustainable transitions by studying an empirical case emerging in Medellín, can contribute both theoretically and practically. Theoretically to understand and substantiate SLB’s role within sustainability as it has been merely theoretically suggested by researchers previously, but not yet empirically validated. Moreover, the exploration of the connection with SLB and sustainability can be helpful in identifying further research gaps within the same theme. Practically, it can give insight to industry and governmental institutions when developing strategies for sustainable development.

The aim of conducting this study in Colombia, exploring contextual conditions as well as techno-economic viability of SLB, can be regarded as a first step in researching the opportunity of SLB in the Latin American region. Thus, it diversifies the current SLB research geographically, which can contribute in extending current research as well as finding new perspectives to SLB. Moreover, it gives a further understanding if and how SLB is viable, which has theoretical and practical value. Theoretical as it deepens SLB research by exploring not only techno-economic viability, but also which factors that impact the viability the most. Thus practical value is supplied as it gives insight to any industry actor interested in pursuing SLB. Simultaneously, contributing with practical value for the sustainable development of Colombia, and addressing environmental challenges in terms of the battery value chain.

1.4 Delimitations

The scope of this thesis is delimited to studying the SLB case identified in Medellín, Colombia. Thus addressing the battery value chain, described in Section 1.1.3, from first life in EV and beyond, but not considering the sourcing nor the manufacturing of batteries. Furthermore, as EV batteries are converging towards LIB, explained in Section 1.1.2, the scope of this thesis is further limited to solely address LIB. Other EV battery types will not be studied.

The study is conducted as an exploratory field study in Medellín, see Section 3.1.

1.5 Thesis outline

This section briefly introduces each of the thesis’s six chapters.

Chapter 1, Introduction

This chapter aims to introduce the reader to the thesis subject and why it is of relevance. This is done by providing background information on the development of EVs, what kind of batteries they use and the battery value chain, as well as what SLB is and its purpose. Background information on Colombia and why the case of SLB is of interest for the country is also provided to introduce the settings of the case study. Thereafter, the thesis purpose and research questions are stated. Lastly, the thesis intended contribution and delimitations are defined.

Chapter 2, Second life batteries (SLB)

This chapter presents a literature review of the existing literature on the subject of second life batteries. The literature review is divided into four parts: types of batteries used in EVs; technical issues and aspects of SLB; the SLB business case; and recycling of batteries. The aim is to investigate the current state of SLB.

Chapter 3, Method

This chapter explains the methods used to carry out the thesis. First, the research design and a research outline is presented. Thereafter the data collection process and sources are introduced. This is followed by an introduction to the theoretical approaches used for the qualitative analysis, and descriptions of the qualitative and quantitative analysis. Lastly, the research quality, e.g. the research validity and credibility, is discussed. The aim of this chapter is to explain to the reader how the research was conducted as well as why certain choices were made.

Chapter 4, Findings and analysis

This chapter first describes the case setting and introduces preliminary findings in order to understand the specific local context better. Thereafter, the findings and analysis of the empirical case study, using the theory described in Section 4.2, is presented. Lastly, the results from a techno-economic assessment of potential real life cases presented in Section 3.4.2.2 as well as a sensitivity analysis and its important findings are presented.

Chapter 5, Discussion

This chapter examines and discusses the findings from Chapters 2 and 4. The discussion is divided into three sections, one for each sub-research question, and one for the main research question. The aim of this chapter is to answer the research questions presented in Section 1.2 and discuss them.

Chapter 6, Conclusion

2 Second life batteries: An overview on Technical,

Economic and Business aspects

This chapter presents a literature review of the existing literature on the subject of second life batteries. The literature review is divided into four parts. Section 2.1 presents the different battery types currently most commonly used in EVs, as well as a comparison between them. Section 2.2 presents the technical issues and aspects of SLB, including battery life cycle, ageing, state of health (SOH) and repurposing. Section 2.3 present the SLB business case and Section 2.4 presents issues related to recycling of the batteries. The aim is to educate the reader on the current state of SLB in order for them to be able to fully follow the thesis analysis, discussion and conclusion.

2.1 Types of batteries used in EV

As previously mentioned in Section 1.1.2, this report will focus on LIBs since they are the most commonly used batteries in EVs today, accounting for 90 percent of the EV market year 2016 (Pillot, 2015). There are several types of LIBs currently available in the market. All modern LIB types use the same materials for the anode and electrolyte, while the cathode material usually varies between different metal oxides (Interview 3, 2020; Sepasi, 2014). Therefore, the battery types are usually classified by cathode material alone. The most common types of LIBs used in EVs are Lithium Iron Phosphate (LFP), Lithium Nickel Manganese Cobalt Oxide (NMC) and Lithium Nickel Cobalt Oxide (NCA) (Interview 3, 2020; Sepasi, 2014). These battery types and the differences between them will be further described below.

2.1.1 Lithium Iron Phosphate batteries

Lithium Iron Phosphate batteries, more commonly called LFP-batteries, uses the cathode material LiFePO4. LFP-batteries are used in most electric busses and larger trucks (Interview 3, 2020). It has also been popular for applications such as commercial use, high power applications, and military applications (Sepasi, 2014). A big advantage of LFP-batteries is the high availability of the materials in their cathodes, iron and phosphate, making them less expensive than other cathode types (Sepasi, 2014). Iron and phosphate are less toxic than both cobalt, nickel and magnesium which is used in the other LIB types, making LFP a better option from an environmental perspective (Yamada et al., 2001). All LIBs contain some materials that are or may become susceptible to supply risk. For LFP-batteries these materials are lithium (Li) and iron (Fe), where lithium is very likely to be susceptible to a high supply risk and iron have a medium supply risk (Thomas, 2018).

The chemical reaction in LFP cells is shown in Equation (1) and illustrated in Figure 7. Lithium acts as a positive electrode material, being ionized during charge and moving from layer to layer in the negative electrode (Sepasi, 2014).

𝐿𝑖𝐹𝑒𝑃𝑂4⇔ 𝐹𝑒𝑃𝑂4+ 𝐿𝑖++ 𝑒− (1)

Figure 7: Chemical reaction of a LPF cell (Toprakçı et al., 2011).

The battery efficiency, defined as the ratio of the energy retrieved from the battery to the energy provided to the battery, of LFP-batteries is assumed to be one fixed value of 95 percent between SOH 100-80 percent and another fixed value of 94 percent between SOH 80-60 percent (Redondo-Iglesias, 2019; Venet and Pelissier, 2019).

2.1.2 Lithium Nickel Manganese Cobalt Oxide batteries

Lithium Nickel Manganese Cobalt Oxide batteries, more commonly called NMC-batteries, uses the cathode material LiNiMnCoO2. NMC-batteries are used in most electric cars, with the exception of Tesla who uses Lithium Nickel Cobalt Aluminum Oxide (NCA) batteries (Interview 3, 2020). Toyota, Mitsubishi, Honda, BMW and Volvo all use NMC-batteries for their EVs (Pillot, 2017; Sakti et al., 2015). Tesla do not use NMC in their vehicles, but they do use it for their Tesla Powerwall and Tesla Powerwall 2 AC (Lithium Ion Battery Test Centre, 2018). Tesla Powerwall is battery pack and BMS manufactured by Tesla and Panasonic. Other applications in which NMC-batteries are sued includes smartphones, laptops, industrial storage and grid storage (Garche and Brandt, 2019). The last few years NMC-batteries have also been used in storage systems. One example of this are two storage systems for frequency regulation in Korea which were installed in 2016 (Kokam, 2016).

et al., 2001). Since cobalt is used in both NMC and NCA-batteries, the battery types which are frequently used in electric cars, the demand for cobalt for battery manufacturing is expected to increase while the available supply of cobalt will decrease (Thomas, 2018).

The chemical reaction in NMC cells is shown in Equation 2. As in LFP cells, lithium acts as a positive electrode material, being ionized during charge and moving from layer to layer in the negative electrode. See Figure 7 for an illustration of this process.

𝐿𝑖𝑁𝑖𝑀𝑛𝐶𝑜𝑂2⇔ 𝑁𝑖𝑀𝑛𝐶𝑜𝑂2+ 𝐿𝑖++ 𝑒− (2)

2.1.3 Lithium Nickel Cobalt Aluminum Oxide batteries

Lithium Nickel Cobalt Aluminum Oxide batteries, more commonly called NCA-batteries, uses

the cathode material LiNiCoAlO2. NCA-batteries are, as mentioned, used in electric cars by

Tesla, while most other car manufacturers use NMC-batteries (Interview 3, 2020; Zackrisson, 2017). NCA cells are also used by several electronic manufacturers, such as Panasonic, Sony and Samsung, in batteries for electronic devices and power tools (Pillot, 2017).

As previously mentioned, NCA and NMC-batteries have several similarities. They both have advantages such as good capacity and working potential, and disadvantages such as their performance degradation, behaving as a decline in capacity and working voltage as well as battery swelling and impedance growth, and a safety hazard over the battery life, arising from abuse conditions such as overcharging, overheating and electric shorting (Zhang, 2020). However, there are some differences between the two battery types, including NCA having higher energy and power densities but shorter life spans (Koniak and Czerepicki, 2017). A study by Popp et al. (2014) showed that NCA cells have the best performance value out of all the LIB types. As mentioned previously, all LIBs contains materials that are susceptible to supply risk. For NCA-batteries these materials are lithium (Li), nickel (Ni), cobalt (Co) and aluminum (Al), where lithium and cobalt are very likely to be susceptible to a high supply risk, nickel have a medium supply risk, and aluminum having a relatively small supply risk (Thomas, 2018). Nickel and cobalt are as previously mentioned both more toxic than e.g. iron and phosphate, making NCA batteries less environmentally friendly (Yamada et al., 2001). Since cobalt is used in both NMC and NCA-batteries, the battery types which are frequently used in electric cars, the demand for cobalt for battery manufacturing is expected to increase while the available supply of cobalt will decrease (Thomas, 2018).

The chemical reaction in NCA cells is shown in Equation 3. As in LFP and NMC cells, lithium acts as a positive electrode material, being ionized during charge and moving from layer to layer in the negative electrode. See Figure 7 for an illustration of this process.

LiNiCoAlO2 <=> NiCoAlO2 + Li+ + 𝑒− (3)

2.1.4 Comparison between LFP, NMC and NCA

or by volume is defined as the amount of energy that can be stored in one kilogram or one liter. The operating current defines the allowed current during charging and discharging, and is measured as a multiplier of a C number, where e.g. batteries with 1C are able to fully charge or discharge the battery in one hour and batteries with 5C are able to fully charge or discharge the battery in one fifth of an hour, i.e. twelve minutes. Charging and discharging temperature is defined as the temperature ranges for battery operation at charging or discharging.

Battery type Life span [cycles] Energy density / weight [Wh/kg] Energy density / volume [Wh/l] Operating current charging [C] Operating current discharging [C] Charging temp. [°C] Discharging temp. [°C] LFP LiFePO4 3,600 130 247 1 3 -20 to 55 -30 to 55 NMC LiNiMnCoO2 3,000 150 300 1 2 to 3 0 to 55 -20 to 55 NCA LiNiCoAlO2 500 240 670 0.5 2 0 to 45 -20 to 60

Table 3: Comparison of LIB cells (Koniak and Czerepicki, 2017).

Figure 8: Comparison of LIB cells (BIS Research, 2019).

The comparison shows that LFP-batteries have the advantage of longer life and safer operating conditions, while the NCA-batteries have better performance values.

2.2 Technical issues and aspects of SLB

2.2.1 The Battery Life Cycle - from first life to second life

Neubauer et al. (2015) state current research on battery life cycle is scarce, and limited to simplified estimations in determining years of operation in first and second life. However, it is also recognized that there is a lack of battery degradation life models (Neubauer et al., 2015). Furthermore, different manufacturing processes of batteries and their first life usage complicates the development of accurate models (Du et al., 2018) Most researchers agree that the EOFL is when SOH is at 80 percent (Hossain et al., 2019; Martines-Lazerna et al. 2018; Jiang et al., 2018; Ioakimidis et al., 2019; Du et al., 2018). After that, a useful second life is deemed to be when the SOH decreases to 60 or 50 percent. This is illustrated in Figure 9 below. However, the actual SOH for a second life is still ambiguous, since there are few empirical cases to study. Interview 3 (2020) states:

“The ageing that occurs from SOH 100 to SOH 80 percent is known to the research community. That is, we understand the chemical reactions and can make accurate assumptions of how the battery will degrade. What happens from 80 to 60 percent is still unknown, and in need of further research. From 60 percent and lower, there are even larger knowledge gaps.”

The statement from Interview 3 highlights the main challenge for SLB. Since the research area is still novel, there are many uncertainties in need of more data, so that accurate models can be constructed and used.

Figure 9: Potential battery capacity for first and second life applications.

Whether EV batteries are used in SLB applications or not, eventually they will need to be recycled. From an environmental point of view battery recycling is crucial. This makes it of high importance to further research and develop the recycling process and regulations for LIBs.

2.2.2 Battery ageing phenomena and patterns

discharging the battery. However, it is possible that cycle ageing and calendar ageing interacts, making it difficult to distinguish the two types of battery ageing processes completely (Redondo-Iglesias et al., 2020).

Another way of conceptualizing battery ageing is to divide it into a typology of different degradation processes, i.e. which chemical result will follow by the degradation. For instance, the loss of lithium inventory within the battery, can be defined as one specific ageing process. That ageing process can then be studied, where causes and effects of it, becomes determined.

Finally, battery ageing can be understood by the concept of the ageing knee. The ageing knee is a point during the battery’s lifetime where the speed of ageing rapidly increases, understood by research as a drastic change in the ageing mechanism. In that regard, ageing models and related conclusions will be different before and after the ageing knee has occurred, making it possible to distinguish ageing into two different modes.

The different ways of conceptualizing ageing patterns and processes are not mutually exclusive. Current research typically uses a combination of the above stated ageing concepts to conduct tests, describe ageing phenomena and draw conclusions valuable for second life battery operations.

2.2.2.1 Battery ageing as a degradation mode

When regarding ageing as a chemical or physical process, the interest lies within determining the changes to the composition of materials within the battery. Battery degradation can then mainly be categorized into three different modes: loss of lithium inventory (LLI), loss of active anode material (LAMPE) and loss of active cathode material (LAMNE). An overview of battery degradations processes is provided in Figure 10.

The LLI occurs due to parasitic reactions, most commonly the increase of the solid electrolyte interface (SEI). The SEI-layer forms due to a chemical reaction between Lithium Ions and electrolyte, consequently decreasing the amount of Lithium-ions available for cycling. Thus, the capacity of the battery decreases directly (Birkl et al., 2017). The SEI-layer growth increases with higher temperatures and higher SOC (Redondo-Iglesias et al., 2020). The loss of active materials, both on the cathode and anode, occurs due to particle cracking, lithium plating and loss of electrical contact. Thus, the capacity decreases and can also lead to a power fade (Birkl et al., 2017). During second life operation of LFP-battery, tests have shown that batteries age in a degradation mode of LLI in combination with LAMNE (Jiang et al., 2018).

2.2.2.2 Battery ageing as cycle or calendar ageing

By differentiating cycle from calendar ageing, a greater understanding of how to optimally store and operate the battery can be achieved. One of the key findings is related to temperature, where it has been concluded that temperature has an effect on both calendar and cycle ageing. Higher temperature increases calendar ageing, while lower temperature increases cycle ageing (Li et al., 2016; Hossain et al., 2019). Therefore, geographic regions experiencing stable ambient temperatures have more suitable conditions for lithium-ion batteries.

The ageing mechanism mainly attributed to calendar ageing is the SEI-layer growth, while cycle aging is mainly caused by lithium plating of the negative electrode (Redondo-Iglesias et al., 2020). SEI-layer growth is further induced by high SOC range and high temperature, while lithium plating of negative electrode is increased by high charge rate and low temperatures (Redondo-Iglesias et al., 2020; Wikner, 2017). That a high SOC range leads to a larger capacity fade is further confirmed by Ecker et al. (2014) and Dulout et al. (2017) in life cycle studies. Furthermore, in a calendar ageing study conducted by Wu and Lee (2017) it was concluded that a high SOC when storing batteries increases the capacity fade. Therefore, batteries should not be stored at high SOC. In a study conducted by Jiang et al. (2018), it is concluded that a second life battery should be operated with a maximum current rate at 1C and a maximum SOC range of 90 percent. Their test also shows that the factor impacting SOH negatively the most is a high current rate. A high current gives rise to a large capacity fade and resistance growth (Jiang et al., 2018).

Expected lifespan in cycles of LFP-batteries from tests conducted by Ioakimidis et al. (2019) indicates that the capacity fade during second life is about 1.5 times higher than during first life.

2.2.2.3 Battery ageing visualized by “the ageing knee”

regarded as factors increasing risk of reaching the ageing knee. High and low temperatures, large current rate, fast charging, high mean SOC, large SOC operation range, high power applications, is all factors contributing to battery degradation. Consequently, the same factors increases the risk of reaching the ageing knee.

Figure 11: Illustration of capacity decrease and the ageing knee.

Figure 11 above illustrates an example of a batteries capacity fade, decreasing at a significantly larger rate once hitting the ageing knee. As mentioned above, several factors impacts when this occurs and it varies for each unique case. When the battery reaches its ageing knee, it can be considered to have reached EOL, which in the figure occurs after x years or cycles and at a SOH of y percent.

2.2.3 Battery state of health

One of the largest issues, discussed and highlighted in several of the publications, when it comes to the use of SLB regards the difficulties in estimating their remaining ability for a second application (Hossain et al., 2019; Martinez-Laserna et al., 2018a; Klör et al., 2018; Martinez-Laserna et al., 2018b; Quinard et al., 2019; Abdel-Monem et al., 2017; Jiang et al., 2018; Martinez-Laserna et al., 2016; Nguyen et al., 2018). Also Klör et al. (2018) discusses the difficulties when it comes to deciding if a battery cell should be reused or recycled. Their study shows that human errors are common during this decision process, and suggests the implementation of a decision support system (DSS) in order to maximize business and environmental value.

A straightforward definition of SOH is actual capacity divided by its nominal capacity (Hossain et al., 2019). This definition of SOH thus refers solely to capacity fade. This is the most widely accepted criteria, which should be of at least 80 percent of the initial capacity in order to be deemed suitable for a second life (Martinez-Laserna et al., 2018b). Martinez-Laserna et al. (2018b) although reflects on that criteria have been widely questioned and that other criteria should be taken into consideration as well when deciding whether a battery cell is of good enough health for a second life application. The study by Nguyen et al. (2018) on the other hand claims that the frequently used end of life (EOL) criteria of 80 percent state of health (SOH) is very useful. Neubauer et al. (2015) also argue that capacity fade is the most important variable when assessing SOH: however, differentiate capacity fade due to cycle ageing, with capacity fade due to calendar ageing. In their study, capacity fade due to calendar ageing is concluded to be of higher importance, since it will have a higher effect on second use lifetime (Neubauer et al., 2015). The test results achieved by Neubauer et al., (2015) are depicted in Figure 12.

Figure 12: Resistance growth and capacity fade as indicators affecting second use lifetime (Neubauer et al., 2015).