0
Review Article

# A Comprehensive Study on Rechargeable Energy Storage TechnologiesOPEN ACCESS

[+] Author and Article Information
Rahul Gopalakrishnan

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: rgopalak@vub.ac.be

Shovon Goutam

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: Shovon.Goutam@vub.ac.be

Luis Miguel Oliveira

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: luis.miguel.oliveira@vub.ac.be

Jean-Marc Timmermans

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: jptimmer@vub.ac.be

Noshin Omar

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: noshomar@vub.ac.be

Maarten Messagie

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: mmessagi@vub.ac.be

Peter Van den Bossche

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: Peter.Van.Den.Bossche@vub.ac.be

Joeri van Mierlo

Mobility, Logistics and Automotive
Technology Research Centre (MOBI),
Department of Electrical Engineering and
Energy Technology (ETEC),
Vrije Universiteit Brussels,
Pleinlaan 2,
Brussels 1050, Belgium
e-mail: jvmierlo@vub.ac.be

Manuscript received September 7, 2016; final manuscript received January 31, 2017; published online April 11, 2017. Assoc. Editor: Partha Mukherjee.

J. Electrochem. En. Conv. Stor. 13(4), 040801 (Apr 11, 2017) (25 pages) Paper No: JEECS-16-1121; doi: 10.1115/1.4036000 History: Received September 07, 2016; Revised January 31, 2017

## Abstract

This paper provides an extended overview of the existing electrode materials and electrolytes for energy storage systems that can be used in environmentally friendly hybrid and electric vehicles from the literature based on lithium-ion and nonlithium technologies. The performed analysis illustrates the current and future evolution in the field of electrode materials development (2015–2040). The investigated characteristics are specific energy, specific power, cycle life, and safety. Furthermore, the proposed study describes the cost and life cycle assessment of the proposed technologies and the availability of these materials.

## Introduction

Since the beginning of the automobile era, the internal combustion engine (ICE) has been used for vehicular propulsion. In addition, motor vehicles powered by the ICE are significant contributors to air pollutants and greenhouse gases linked to global climate [1,2]. As the global economy begins to strain under the pressure of rising petroleum prices and environmental concerns, research has spurred the development of various types of clean energy transportation technologies such as hybrid electric vehicles (HEVs), battery electric vehicles (BEVs), and plug-in hybrid electric vehicles (PHEVs) [3,4]. But the establishment of the energy storage technology, which can support the required output power during acceleration, achieving efficient use of the regenerative energy and considerable life cycle are the critical aspects and no current battery technology can meet these often concurrent objectives [57].

In the last decade, various rechargeable energy storage battery technologies have been developed, such as /lead-acid, nickel-metal hydride, and lithium-based batteries. However, the first two battery technologies seem not convenient for (PH)(B)EVs due to their limited performances in terms of energy density [810]. Therefore, in the last years, a significant research work has been performed in the field of lithium-ion technology. However, the term lithium-ion family encompasses a large number of different chemistries based on the used materials for the anode and cathode as presented in Table 1.

The selection of the appropriate lithium-ion chemistry in HEVs, PHEVs, and BEVs depends of many parameters. In this study, the following requirements for PHEVs and BEVs are considered, based on the goals as specified in United States Advanced Battery Consortium (USABC), Electric Power Research Institute (EPRI), Sloan Automotive Laboratory at the Massachusetts Institute of Technology (MIT), European SuperLIB project and European BATTERIES 2020 project as presented by Table 2.

As we can observe from the requirements for PHEVs and BEVs, the goals are quite high for the existing battery technology. Particularly the high requirements for both energy and power can be considered as a main barrier, since the commercial batteries are often optimized for specific applications (either high energy or high power). In addition, the cost and cycle life goals are high for the available lithium-ion battery (LIB) technologies.

## Lithium-Ion Battery

Lithium ion batteries (LIB) have been in the market for more than two decades since Sony first launched them in 1991. Because of the low atomic weight of the lithium element and the high negative potential of the Li+/Li redox couple (−3.045 V versus standard hydrogen electrode), leading to a much higher battery potential and higher energy densities compared to the previous generations of batteries((e.g., lead-acid, nickel metal-hydride), Li-ion batteries represented a real technological breakthrough. The principle of the Li-ion battery is shown in Fig. 1; it consists of two electrodes, a positive and a negative, which are separated by means of a separator and the electrolyte. The separator is generally a fine sheet of polyethylene or polypropylene, whose main function is to prevent short circuit between the two electrodes. During the charge step, the battery is connected to a current source and electrons move from the positive to the negative electrode. Simultaneously, at the positive electrode, the departure of the electrons occurs together with the oxidation of the transition metal and, due to charge compensation, together with the deintercalation of Li+ ions from the host structure. The Li+ ions then move inside battery through the electrolyte (organic solvents + LiPF6) to the negative electrode. At the negative electrode, the opposite happens: the arrival of electrons leads to a reduction reaction of the active material (AM), and due to charge compensation, to an intercalation of Li+ ions. In this way, electric energy is converted and stored as chemical energy. During the discharge step, all these mechanisms are reversed and the electrons moving in the external circuit allow recovering the electric energy, which was chemically stored during the charge.

The chemical reaction equations involved are the following (example of a LiMO2/graphite cell, M being a metal as for example Co, Mn, or Ni): Display Formula

(1)$Cathode:LiMO2⇄DischargeChargeLi1−xMO2+xLi++xe−$
Display Formula
(2)$Anode:6C+xLi++xe−⇄DischargeChargeLixC6$
Display Formula
(3)$Overall:LiMO2+6C⇄DischargeChargeLi1−xMO2+LixC6$

Generally, the positive and negative electrodes are composed of an active material (AM), a binder (usually a polymer), and a conductive agent (usually carbon black) coated on a current collector (Al for the positive electrode and Cu for the negative electrode) as shown in Fig. 2. The binder brings its mechanical strength to the electrode, allows a good adhesion on the current collector, and also provides internal ionic percolation by allowing necessary uptake of the electrolyte into the electrode [20]. The conductive agent helps to increase the electrode's electric conductivity by ensuring a good connection between the active material particles and the current collector and promotes a good circulation of the electrons to the current collector.

## Battery Materials

The performance of a lithium-ion battery depends on the chemistry of the battery, i.e., the employed positive and negative electrode material, the separator, the electrolyte, and binders. In the application of HEVs, the intercalation and deintercalation of lithium-ions should go rapidly; therefore, the surface area of the active material is significantly higher when compared to BEVs. While in the BEVs, the electrode should be thicker in order to store more energy. Thus, the ratio of the volume and surface is high in the case of BEVs, while for HEVs, this ratio is low. In addition, the selected electrolyte should be electrically and thermally stable and of course highly conductive. Thus, in general, the selection of the materials and the electrolyte will be essential for improvement of a battery technology.

Therefore, in this section the electrochemical properties of various negative and positive electrode materials will be analyzed and compared along with the electrolytes.

###### Graphite and Carbon Based.

As mentioned earlier, the performance of a battery relies on the used anode, cathode, and electrolyte. Nowadays, in the most commercial lithium-ion batteries, graphite and carbon-based materials are used as a negative electrode [21]. However, these materials differ from each other in terms of morphology, structure, microstructure, and texture [22]. From the point of view of the anode materials, graphite seems the most appropriate candidate due to the high specific capacity, 372 mAh/g. Lithium insertion into graphite happens at a potential of less 1 V versus Li+/Li− [21]. At such low potential, the reduction of electrolyte occurs. However, graphite-based anodes have some particular problems such as less beneficial performances at low temperatures (lithium plating) and formation of the passivation solid electrolyte interface (SEI) layer, which consists of a mixture of inorganic and organic phases (2 nm inorganic phase followed by 80 nm organic phase) [23]. In the first cycle, some capacity fade is observed due to the formation of SEI, which leads to irreversible capacity loss. From the experimental results, it has been revealed that the thickness of the SEI layer increases during cycling, which leads to further increase of the internal resistance due to reduction of the active surface area of the electrode and capacity degradation due to Li consumption [2427]. In Refs. [22] and [28], the main properties of graphite, soft carbon, and hard carbon materials are described as a negative electrode in lithium-ion batteries. Graphite (LiC6) is nothing but stacking of individual graphene layers. The space between the planes is about 0.335 nm [28]. It is generally known that during intercalation process, 10% volume expansion of the graphene occurs. Here, it should be noted that the practical specific capacity available is about 310 mAh/g [22]. The potential curve for lithium-ion battery (LIB) anode made of graphite is characterized by a flat and wide range. Graphite became the state-of-the-art anode material because of this trait and fairly simple reaction chemistry with minimal problems like volume expansion, etc. The charge/discharge curve for LIB graphite is characterized by several flat stages, as shown in Fig. 3. Many other variations were developed to achieve lower cost and increased capacity. Among the various types of graphite, modified natural graphite has become the most commonly used. Natural graphite, which is readily reactive with electrolytes, renders itself useless for it is being used as anode material even though it is inexpensive. Thus comes the need to coat them with a carbon layer to enhance its resistance toward the electrolyte. Hard carbons are one of the well-known, highly recommended anode materials used for HEV applications.

###### Titanate Oxide.

In the last years, new anodic materials have been proposed based on spinel lithium titanate oxides as presented in Fig. 4. According to Ref. [30], lithium titanate oxide is a more ideal insertion material with a specific capacity of 175 mAh/g. It is safer, low cost, nontoxic, and SEI layer formation occurs on the anode surface [21] due to the minimal reducing capacity of the material. The lithium insertion potential of this oxide is around 2 V versus Li. This value lies in the stability window of common organic electrolytes. Lithium titanate oxide-based materials have a spinel structure, whereby the surface area is much bigger (100 m2/g versus 3 m2/g) for carbon-based electrodes [31]. The larger the surface area available, the greater the rate of movement of electrical charges. Thus, lithium titanate oxide does not suffer from applying high current rates. The most used lithium titanate oxide compound is Li4Ti5O12 [31]. Contrary to graphite-based electrode, the volume expansion of the lithium titanate oxide electrode during intercalation of lithium is below 0.2% [28], thus these anodes are also known as the zero strain anodes. In the last few years, several companies started with commercializing this technology, such as Toshiba, EIG Batteries, and Altairnano. The energy density of lithium titanate oxide-based cell is in the range of 70–80 Wh/kg, which is much lower than the NMC and NCA batteries [32,33]. Thus this technology cannot be used in electric vehicles where driving range is of prime importance because of low energy density, but they have high potential in applications where high peak current or fast charging is required, for example in public electric buses. The lower energy density is due to the lower specific capacity and nominal voltage (2.2 V) [11,12]. However, a cycle life of over 20,000 cycles (at 80% depth of discharge) is not a barrier for this technology. Bruce et al. [21] have documented that the specific capacity of Li4Ti5O12 can be improved by using nanotubes/nanowires (NT/NW) composed of TiO2–(B). This technology still has the advantages of Li4Ti5O12. The specific capacity can be improved from 175 mAh/g up to 300 mAh/g. The diameters of the wires are typically 20–40 nm and 0.1–1 mm. According to their experimental results, it is reported that the reversible capacity of nanowires is twice than that of the nanoparticles after 100 cycles. The length of the wire ensures good electron transport between the wires and the small size improves the fast lithium intercalation/deintercalation.

###### Tin.

For tin-based batteries, the specific capacity is almost double than that of the carbon-based electrode (i.e., 372 mAh/g compared to 790 mAh/g) [21,34,35]. The general reaction is presented below: Display Formula

(4)$SnO2+4Li++4e−⇒Sn+SLi2O$
Display Formula
(5)$Sn+xLi++ xe−⇔LixSn (0≤x≤4.4)$

During discharging, Sn reacts with oxygen and creates SnO2, tin metal, and Li2O.

The experimental results reveal that the capacity fade is significant even though the initial capacity was higher than 1100 mAh/g. Similar study was conducted in Ref. [36] where the initial capacity was reported to be around 1400–1500 mAh/g. The discharge specific capacity after five cycles fell down to 650–700 mAh/g and continued to further decrease with increase in the number of cycles. This issue has been related to volume change, which is caused by a high charge voltage and grain size. In order to reduce the capacity retention, the electrode morphology can be changed by nanostructure, nanowires, etc., as presented in Table 3. Bruce et al. [21] proposed to replace the single metal alloy by an AB intermetallic phase, for which the electrochemical process in a lithium cell involves the displacement of one metal (e.g., A) to form the desired lithium alloy, LixA, while the other metal B acts as an electrochemical inactive matrix to buffer the volume variations during the alloying process. Such materials could be Cu6Sn5, InSb, and Cu2Sb, which have strong structural relationship with their lithiated products Li2CuSn and Li3Sb for the Sn and Sb compounds, respectively [39]; however, the capacity fade during cycling remains high. From another point of view, tin transition metal carbon has been proposed to replace the carbon in the negative electrode [40]. The most promising compound is Sn–Co–C, which has been approved based on experimental results during its cycle life. The capacity fade over 100 cycles was significantly less than the above-mentioned compounds. Therefore, Sony Corporation (Tokyo, Japan) started with commercialization of a lithium-ion battery based on tin, cobalt, and carbon as an anode material.

###### Lithium Metal-Based Anodes.

Lithium metal-based batteries suffered from dendrites formation during charge-discharge processes which led to short circuiting of the battery, hence they were replaced by lithium ion-based rechargeable batteries. With the increased interest in recent years for research in advanced lithium technologies like lithium-sulfur and lithium-air batteries, the lithium metal anodes have also seen growing interest. Lithium metal anodes enjoy high theoretical capacity of 3860 mAhg−1. To commercialize this anode, first there is a need to understand the composition and the growth mechanism of the dendrites, followed by ways to suppress their growth. In situ and ex situ techniques using atomic force microscope (AFM), scanning electron microscope, and transmission electron microscope have been used to study the morphology of the dendrite and the composition through X-ray photoelectron spectroscopy (XPS), scanning auger electron microscopy (SAM), and inductively coupled plasma spectroscopy (ICP). In Ref. [41], such techniques have been used to analyze the surface of lithium in dry air atmosphere using in situ AFM and Fourier transform infrared spectroscopy (FTIR); the composition through XPS and SAM to find out that diffusion of lithium through the ridge-line and grain boundary (consisting of Li2Co3 and Li2O) are very important factors for controlling the dendrite formation. Aurbach et al. [42] have extensively studied the lithium surface chemistry in organic electrolytes through FTIR and have found that higher charging rates decrease the size of lithium grains (creating fresh sites for electrolyte reaction) every charge-discharge cycle in the cells. Thus could lead to failure of the battery due to consumption of entire electrolyte and forming lithium dendrites. Two approaches have been provided by research groups regarding controlling dendrite formation, one through mechanical barriers and other through additives in electrolytes which lead to a stable SEI layer formation. Polymer electrolytes and solid-state electrolytes are the two viable solutions as far as mechanical barriers are concerned. In Ref. [43], poly-ethylene oxide (PEO) was used as mechanical barrier as the polymer does not react with fresh lithium; thus, no dendrite formation and the PEO was thermodynamically stable up to 100 °C. Solid inorganic electrolytes also can function as a mechanical barrier due to their strength and does not allow dendrites to penetrate. One such example was shown in Ref. [44] where such an electrolyte was used in Li-air battery for mechanical stability and suppressing dendrite growth. Other approaches do exist [45,46], where the surface morphology of the lithium metal is changed to facilitate uniform deposition or slowing down the dendrite growth but was never successful as the authors could not demonstrate if such a condition can be maintained throughout the cycling life.

###### Silicon.

Silicon seems to be a possible alternative for the graphite or carbon anode since its specific capacity is the highest known. The capacity comparison between various elements is shown in Fig. 5. Silicon is considered as a promising candidate for negative electrode for Li-ion batteries, not only due to its wide abundance but also, as already mentioned, due to its large theoretical specific capacity of 3579 mAh/g, almost ten times higher than graphite with a capacity of 372 mAh/g and its relatively low working potential of 0.05 V (versus Li+/Li). However, the fast capacity fade observed during cycling has prevented the silicon anode from being commercialized until now. The main challenge reported for the implementation of Si-based anodes is its large volume variation (∼275%), which often leads to pulverization of the active alloy particles, isolated particles, and thus loss of conductivity and poor cycling stability [4850]. Silicon particles indeed need to undergo a drastic change in their structure and morphology due to the uptake of 3.75 lithium atoms for 1 silicon atom (275% volume expansion) compared to 10% for graphite, and 33% in generic reactions with MO2 type anodes [47]. A SEI (solid electrolyte interphase) layer is formed during the first lithiation of Si particles when the potential of the anode is below 1 V (versus Li+/Li). The formation is due to the decomposition of the electrolyte, which is not stable at low potentials, on the surface of the silicon particles. This SEI layer prevents further secondary side reactions and is electrically insulating and ionically conducting. Due to huge volume variations during lithiation/delithiation, the SEI layer cracks, exposing fresh Si surface to further decomposition of electrolyte, which leads to endless consumption of Li and thus to capacity fade. The isolation and disconnection of some Si particles from the conductive network and thus from the current collector also lead to a drastic increase of the internal resistance [51]. As a result, the delithiation reaction cannot be fully completed, with some Li remaining in the Si particles, leading to an irreversible capacity loss [52].

Some of the methods employed to try to increase the cyclability of Si-based anodes will be described in the following paragraphs.

Nano sizing: It was generally observed that bulk Si-based electrodes had severe fading and loss of capacity problems [48,53]. Although decreasing the particle size improves the rate of lithiation/delithiation and decreases the volume expansion, it also creates large surface area for electrolyte-decomposition reactions, which results in the continuous formation of solid-electrolyte-interphase (SEI) layer and the associated irreversible loss of capacity [49].

Binders: Binders are essential for electrodes by forming cross-links and providing a good adhesion on the current collector, which to a certain extent helps controlling the volume expansion. PVDF (polyvinylidene fluoride), a widely used commercial polymer binder in the manufacturing of electrodes for batteries, shows very low performances when used in combination with silicon-based electrodes [50]. As a consequence, alternative binders have been investigated for silicon and silicon composite electrodes, such as polyethylene oxide with lithium perchlorate (PEO–LiClO4), polyethylene glycol with lithium perchlorate (PEG–LiClO4), rubber-like ethylene propylene diene monomer (EPDM), alginate (alginic acid + sodium salt from brown algae), poly acrylic acid (PAA), lithium polyacrylic acid (LiPAA), poly imide amide (PAA-Torlon), styrene butadiene rubber (SBR), and sodium carboxy methyl cellulose (CMC) [47,51,52,54,55] where different retention capacity per cycle has been reported. More efforts are still needed with respect to binders in order to shift from the state-of-art binder (PVDF) to others mentioned earlier. One such effort was seen in Ref. [56], where the authors synthesized polymer binders that could be cathodically doped (n-type) and claimed that this polymer showed improved electrical and mechanical characteristics. Here polyfluorene-based polymers (PFFO and PFFOMB) were introduced with carbonyl and methyl benzoic ester group to tailor the electronic states (LUMO) of the polymer. The PFFOB binder when used with the silicon anode material showed 90% capacity retention after 650 cycles (2100 mAhg−1).

Microstructures: Cui et al. investigated core-shell carbon nanotubes (CNT) coated with amorphous silicon particles, where the carbon nanotubes act as a backbone to support the high capacity silicon layer, showing a high capacity of over 2000 mAh/g for more than 50 cycles [57]. Magasinski et al. [58] have also reported a 3D hierarchical structured silicon-coated carbon nanocomposite with approximately 50% silicon content, used as anode for lithium-ion batteries, with an initial discharge capacity of 1950 mAh/g at C/20. Chan et al. [59] have also investigated Si nanowires (NW) anode configuration, as illustrated in Fig. 6. Nanowires are grown directly on the metallic current collector substrate. The capacity measured during the first charge was 4277 mAh/g and both charge and discharge capacities remained nearly constant up to ten cycles. The coulombic efficiency (ratio between discharge and charge capacities) of Si NW was found to be low, which could be due to high internal stresses associated with lithiation and/or accelerated electrolyte decomposition reactions through the formation of SEI layer on the largely exposed surface areas provided by the Si NWs.

It is also claimed that the small NW diameter allows a better adaptation of the large volume variations over lithiation/delithiation cycles, as it involves particle sizes below which particles do not fracture, around 200 nm according to literature [60]. Wu et al. [61] proposed a new concept where the anode material consists of an active silicon nanotube surrounded by an ion-permeable silicon oxide (SiOx) shell. This material can cycle over 6000 times in half cell (against pure lithium), while retaining more than 85% of its initial capacity. As shown in Fig. 7, the volume expansion is prevented by the oxide shell, and the expanding inner surface is not exposed to the electrolyte, resulting in a stable solid–electrolyte interphase. Batteries containing these double-walled silicon nanotube (DWSiNT) anodes exhibit charge capacities approximately eight times larger than conventional carbon anodes and charging rates of up to 20 C. The main drawback of this anode material when compared with pure silicon is the relatively low anode capacity (600 mAh/g) induced by constraints caused by the SiOx layer, but extraordinary cyclability with 100% coulombic efficiency after 6000 cycles at 12 C rate. Liu et al. [62] came up with a so-called Yolk-shell model where Si nanoparticles have been completely “sealed” inside conformal, thin, self-supporting carbon shells, with wisely designed void space in between the particles and the shell. This model is illustrated in Fig. 8. The well-defined void space allows the Si particles to expand freely without breaking the outer carbon shell, therefore stabilizing the SEI on the shell surface. This Yolk-shell-structured Si electrode allows reaching a high capacity (2800 mAh g−1 at C/10), relative long cycle life (1000 cycles with 74% capacity retention), and high coulombic efficiency (99.84%). Nevertheless, it seems that these performances were obtained with very low electrode loadings; this concept still needs to be proven with “acceptable” loadings (4–5 mg/cm2). Furthermore, from an industry point of view, such a process would probably be way too expensive.

Composites: The numerous studies done on composite materials in which active particles are finely dispersed in an active or inactive solid matrix tend to prove that the matrices help buffer the expansion of the active materials. But these studies also reveal that dispersing Si in an inactive host matrix leads to a decrease of the reversible capacity and that dispersing Si in an active metal matrix does not always provide a satisfying capacity retention [63]. Liu et al. proved that conductive additives such as graphite flakes and/or nanoscale carbon black can give a good capacity but only for few cycles [64]. It has also been observed that negative electrodes prepared by depositing 20 wt. % of Si nanoparticles on a graphite surface showed both a high initial charge and high discharge capacities of 1350 and 1000 mAh/g, respectively, and a capacity retention which could reach 900 mAh/g after 100 cycles [65].

Prelithiation: Garcia et al. [66] synthesized LixV2O5 by chemical prelithiation of V2O5 by using n-butyl lithium, which brought new variations in the material and a capacity improved by 15–20% compared to nonprelithiated V2O5. Similarly, the electrochemically prelithiated carbon nanotubes (CNT) by Landi et al. [67] showed an improvement in the cyclability. Li3N was used by Pereira and Klein [68] as well as by Zhang et al. [69]: their respective prelithiated metal fluoride sample and phosphide sample, prepared through high energy ball milling, lead to exceptional results. The phosphide sample for example showed an increase of coulombic efficiency in the first cycle from 62.8 to 72.8%. Zhamu and Jang [60] filed a patent in 2008 stating the techniques for producing prelithiated silicon and claimed that prelithiation of silicon followed by comminuting the particles leads to the production of submicron particles which exhibits high specific capacity, long cycle life and high reversibility. They showed that prelithiated Si anode with carbon coating yielded promising results when compared to nonprelithiated silicon anodes or prelithiated silicon anodes with no carbon coating.

Yakovleva et al. [70] engineered a new type of lithium called SLMP (stabilized lithium metal powder) from FMC lithium which is stable in air, thanks to the coating. They claimed that this SLMP could be used to prelithiate anodes, which could enhance battery performances at a lower cost, since the cathode material is no longer the main provider of lithium for the anode. The SLMP addition reduced the irreversible capacity of the anode from 487 mAh/g to 155 mAh/g. Forney et al. [71] used this SLMP to prelithiate silicon-carbon nanotubes. They used SLMP during battery assembly and prelithiated the Si/C nanotubes to eliminate 20–40% irreversible capacity loss during the first cycle owing to SEI layer formation. The authors made cells with two different variations of prelithiation, one with manual pressing and the other without pressing. The anodes were tested against lithium nickel cobalt aluminum cathodes. They claimed that prelithiation using SLMP by manual pressing showed the most promising results(∼1.1 mAh) when compared to nonprelithiated batteries(∼0.5–0.6 mAh) and SLMP-treated batteries without pressing(∼0.7–0.9 mAh).

Tang et al. [72] successfully lithiated silicon with lithium by ball milling LiBr and KSi in a 1:1 ratio in a tungsten carbide ball mill. They claimed that if the x in LixSi is greater than 1, the capacity drastically decreases after some cycles, but if x is restricted to a value lower than 1, they observe a stable capacity of about 870 mAh/g (0.9 Li+ exchanged) for at least ten cycles, with an electrochemical signature that is different from the classical Si lithiation/delithiation.

###### Lithium Cobalt Oxide.

Since the beginning of commercialization of lithium-ion batteries in mobile applications, lithium cobalt oxide LiCoO2 has been proposed as an interesting solution due to its high specific capacity 120–140 mAh/g (2–3 times higher than nickel cadmium-based batteries) and high nominal voltage 3.7 V (three times higher than alkaline batteries (1.2 V). It is reported that the specific capacity of LiCoO2 can be improved up to 170 mAh by coating a metal oxide on the surface of the LiCoO2 particles and when the battery will be cycled between 2.75 and 4.3 V [11]. The main reason for this enhancement has been related to minimizing the reactivity of Co4+ on charge with the acidic HP in the electrolyte [73]. Furthermore, LiCoO2-based batteries show high reversible intercalation and de-intercalation of lithium-ion, which results in a long cycle life (up to 800 cycles) [40]. In 1991, Sony Corporation started with commercialization of LiCoO2-based batteries in mobile applications. The positive electrode was composed of LiCoO2 and the negative electrode was carbon based in combination with an organic liquid electrolyte [74]. However, it should be noted that LiCoO2 is the most reactive material and has less thermal stability compared to other cathode materials. During over-discharging and particularly over-charging, LiCoO2 undergoes several degradation mechanisms, which could make the system instable, which in the worst case could even result in explosion. For this reason, the exploitation of LiCoO2 has never been considered as a good solution in electric vehicles. Moreover, cobalt is an expensive material (we need to define here an indication of the cost), which increases the manufacturing cost of this type of battery. Finally, the resources of cobalt in the world are limited, which will be discussed in more detail in the “Resource Constraints” section of this paper.

###### Lithium Nickel Oxide.

From the electrochemical point of view, LiNiO2 is also an interesting candidate for positive electrode material. The specific capacity is much higher (170 mAh/g) than that of lithium cobalt oxide. Also, nickel is more available and less toxic compared to cobalt. It is generally known that LiNiO2 and LiCoO2 have an α-NaFeO2 layered structure with the oxygen in a cubic close-packed arrangement [75]. But the structure of the LiNiO2 is less stable than the LiCoO2 [7678]. This instability has been related to many issues [73], the main problems being cation mixing and off stiochiometry. Hereby, the lithium diffusion can be limited and the power capability reduces. A second mechanism that occurs is the fact that low lithium content makes the system unstable due to the high effective equilibrium oxygen pressure. In other words, this makes the battery unstable when it comes in contact with organic liquids. Therefore, the cycle life of LiNiO2 is too short compared to cobalt oxide batteries. These obstacles are partially removed by substituting nickel with aluminum, magnesium, titanium, and gallium [40]. Some researchers reduced the capacity fade of LiNiO2 by using the compound LiNi0.3Mn0.33Co0.33Al0.01O2 [40,79]. It is generally known that cobalt increases the conductivity and the degree of ordering, while manganese and aluminum improves the safety [20,80]. Another attempt has been made by making use of the compound LiNi1−xTix/2Mgx/2O2 [81]. From experimental results, it has been concluded that the specific capacity of the cathode material was enhanced to 180 mAh/g. However, the cyclability of this configuration is still unknown. Sawai et al. [82] proposed a LiNi1/2Co1/2O2 composed of the latter compound in the positive electrode and Li [Li1/3Ti5/3]O4 in the negative electrode. Accelerated cycle life tests at 10% depth of discharge have been conducted and they observed that the capacity degradation after 50,000 was 10%. Moreover, they concluded that the power fade was significantly higher than the capacity fade. Huang et al. [83] proposed NiO hollow spheres with a diameter of about 2 μm, which are composed of NiO particles, of about 200 nm in size for enhancing the charge and discharge capacity (560 mAh/g and 490 mAh/g, respectively) as well as the cyclability. The main improvement in this field is related to the large specific surface area and the good electric contact among the particles due to the NiO hollow spheres.

###### Lithium Nickel Cobalt Aluminum Oxide.

In the last few years, lithium nickel cobalt aluminum oxide (LiNiCoAlO2) received considerable high attention due to its high specific capacity (180 mAh/g) [84]. This technology has been commercialized by many manufacturers like Saft, Nissan, Gaia, Panasonic, and Leclanché. The most used compound structure is LiNi0.85Co0.1Al0.05O2. Aluminum doping for this compound is beneficial in order to suppress the impedance increase by stabilization of the charge transfer impedance on the cathode side and by improvement of the electrolyte stability. Therefore, the capacity decrease can be reduced. However, the cycle life tests performed at the Vrije Universiteit Brussel on various NCA cells demonstrated a rather short cycle life [10]. The energy density of these cells can be in the range of 140–180 Wh/kg [8]. However, the high cost, short cycle life, and thermal instability make them unpopular for electric vehicles, with the exception of Tesla vehicles. Finally, this battery technology suffers significantly at high temperatures owing to SEI layer formation [85].

###### Lithium Nickel Manganese Cobalt Oxide.

From the shortcomings of LiCoO2 and LiNiO2, lithium nickel manganese cobalt oxide (LiNiMnCoO2 or NMC) has been proposed as a possible solution for electric vehicles. This compound combines high performances, lower cost, and higher safety properties. The structure of NMC is shown in Fig. 9. The role of these elements is illustrated below in Fig. 10: Material triangle for LiNiMnCoO2 electrode material Fig. 10. The well-known composition of LiNi1/3Mn1/3Co1/3O2 (NMC 111) has already been commercialized. In the last years, several companies such as Panasonic, Sanyo, Hitachi, GS Yuasa, Samsung, EnerDel, EIG, Kokam, Lechanché, Evonik/Litarion, Enax, and Imara started with commercialization of LiNiMnCoO2-based batteries. Figure 11 shows the open circuit potential of NMC 111, which was done at VUB in collaboration with Umicore. Figure 12 shows the Open Circuit Voltage (OCV) curve for NMC 442 from SBO project BATTLE (IWT130019). The NMC electrodes can be used against Lithium-Titanate Oxide (LTO) anodes as well as graphite anodes. The cycle life of the NMC cells is around 3500–4000 cycles at 80% depth of discharge. Amatucci and Tarascon [89] have tried to increase the nickel content and have analyzed their main electrochemical parameters. The authors investigated synthesized Li[NixCoyMnz]O2 where (x = 1/3, 0.5, 0.6, 0.7, 0.8, 0.85). It was found that indeed the capacities of the materials increased with increasing Nickel content but the capacity retention and thermal stability decreased with increasing nickel content. From the study, it is evident that, in order to increase the capacity retention and thermal stability, addition of metals seems to be the direction in which the research must proceed. Another lead could also be surface treatments as well as playing with the architecture of the materials. The conductivity and lithium ion diffusion values are higher with increase in nickel content.

###### Lithium Manganese Oxide.

In order to have a cheap, safe, and environmental friendly material as the positive electrode, lithium manganese oxide (LMO)-based batteries have been introduced, which has been commercialized some years ago [90]. This battery technology has a high nominal voltage (3.7 V) and a relatively high specific capacity of 100 mAh/g as presented in Table 1. It is generally known that lithium manganese oxide (LMO) batteries can be found in two different structures: orthorhombic LiMnO2 and spinel LiMn2O4 (Fig. 13). However, at elevated temperatures, some stability concerns occur in the battery system [91]. At such conditions, lithium manganese oxide can be corroded in organic electrolyte, whereby the Mn2+ ions can be dissolved in the electrolyte. Thus, in the working stage, the active material reduces, the resistance increases, and the capacity decreases. In order to increase the performance of LiMn2O4 batteries, the cathode materials can be treated by coating or doping aluminum LiAl0.1Mn1.9O4 or by cationic substitution with Cr, Ti, Cu, Ni, Mg, and Fe [9294].

###### Lithium Iron Phosphate.

Lithium iron phosphate (LiFePO4 or short LFP) with olivine structure (see Fig. 14) has been proposed as a promising candidate to overcome the weakness (e.g., thermal stability, cost) of the earlier cathode materials. The phosphate material was first proposed as a candidate in lithium-ion batteries in 1997 [95]. The nominal voltage for this material is about 3.3 V and the specific capacity varies between 150 and 160 mAh/g as presented in Ref. [86]. The LiFePO4-based batteries have high thermal stability and are much less expensive than the earlier mentioned battery chemistries [96].

Moreover, they are not toxic. In contrast to LiCoO2, LiCo0.2Ni0.8O2, LiNi0.8Co0.15Al0.05O2, and LiNi1/3Mn1/3Co1/3O2, LiFePO4 (LFP) batteries do not release oxygen at elevated temperatures, which can react with the electrolyte and result in a thermal runaway [96]. When comparing LiFePO4 to other materials, others are unstable at elevated temperatures, whereby the cathode material starts to decompose and produce oxygen, which reacts further with the organic electrolyte. Such situation leads to uncontrolled reactions and possible thermal runaway as a consequence. LiFePO4, however, has low electric conductivity (10−9 S/cm) [ref]. The ionic conductivity has been related to lithium-ion transport and hopping of small polaron, two phenomena which are highly coupled [73]. Thus, the improvement of the conductivity was necessary before commercialization of this electrode technology. Improvement of the electrochemical performances and of the conductivity in particular was achieved by using metallic phosphide on the surface of LiFePO4 [97]. In Ref. [98], it is reported that the electrochemical properties of LiFePO4 can be enhanced by using LiFePO4 particles with carbon. In Ref. [99], calendar life aging of 7 Ah LFP cells was investigated at different temperatures and different state of charge levels as shown in Fig. 15. From this study, one can observe that high storage temperature and high state of charge levels lead to accelerated capacity degradation. The impact of the storage temperature is more evident than the state of charge at which the battery is stored. Further, in Ref. [100], it is indicated that the electrical conductivity can also be improved by doping LiFePO4 with materials such as Cu phosphate LiFe1−xCuxPO4. In the literature, one can find several phosphate-based cathode materials. In this context, lithium manganese phosphate oxide (LiMnPO4) has been proposed as a potential candidate for it has higher nominal voltage (4.2 V versus Li) [101]. However, the low conductivity and structural distortions upon oxidation to MnPO4 are the main hinders of the ionic transport, which makes the suitability of this compound less interesting for hybrid applications [73]. In order to improve the ionic transport, significant research work was done by Drezen and coworkers whereby the particle size has been reduced and carbon has been added to improve the surface conductivity [102]. The experimental results showed over 200 cycles with improved cyclability. However, the main obstacle for commercialization of this cathode technology is related to the fact that carbo-thermal reduction of LiMnPO4 does not occur at temperatures below 1000 °C, which is making the production of surface phosphides more difficult. In the category of phosphate-based electrode materials, one can find fluorophosphates cathode electrode materials, which exhibit a higher operating voltage compared to LiFePO4. The first successful fluorophosphates electrode material is LiVPO4F (vanadium fluorophosphate), which has been proposed by Barker and coworkers [103]. The nominal voltage of this fluorophosphate-based cathode material is 4.2 V versus Li and the specific capacity is close to 154 mAh/g. Furthermore, an interesting new development is sodium vanadium fluorophosphate Na3V2(PO4)2F3 [103]. The specific capacity is around 120 mAh/g and nominal voltage is about 4.1 V versus Li. In 2007, sodium iron fluorophosphate (Na2FePO4F) was introduced as a new cathode material with a nominal voltage of 3.5 V versus Li and a specific capacity of 135 mAh/g [104]. The experimental results showed that the capacity fade over the first 50 cycles is very limited. However, these materials have not been commercialized yet. In hybrid applications, the battery should be able to provide peak power during short durations. This can be realized if LiFePO4 were heat-treated to increase the hole conductivity by adding appropriate additives or by employing nanostructured electrode morphologies such as nanoparticles (2 nm diameter) and nanofibers, which allow to reduce the diffusion length of the lithium-ion in and out of the electrodes based on the following relationship [21,105,106]

Display Formula

(6)$t=L2/D$

where t is the diffusion time constant (in seconds), L is the diffusion length (in meters), and D is the diffusion constant (in m2/s).

Moreover, nanoparticles increase the contact area with the electrolyte and hence allow a high flux of lithium-ions across the solid-electrolyte interface. In the field of LiFePO4 batteries commercialization, the most pronouncing companies are A123 Systems, EIG, Saft, Panasonic, Enerdel, Altairnano, LG, Valence, ASEC, K2, and GAIA.

###### Lithium Silicates.

Due to the success of LiFePO4 batteries, another family of electrode materials has come into the limelight and has been studied: silicates in the form of Li2MSiO4 (where M = Fe, Mn or Co), where M2+ is a transition metal. In this architecture, the phosphate group from LFP materials has been replaced by an element of the sulfate group, which is more electron drawing. The specific capacity of Li2FeSiO4 is about 140 mAh/g and the nominal voltage is around 3.1 V versus Li [107]. Muraliganth et al. [108] have shown that Li2FeSiO4 and Li2MnSiO4 could be synthesized by microwave synthesis and can provide capacities of 148 mAh/g and 200 mAh/g, respectively. Li2CoSiO4 was synthesized by Lyness et al. [109] and by Gong et al. [110] through ball milling and solution-hydrothermal route with which charge capacities of 170 mAh/g and 234 mAh/g where achieved. The structure of Li2FeSiO4 is shown in Fig. 16. Despite being an attractive choice with regard to safety, availability of materials and high theoretical capacities, lithium silicates research generally suffers from lack of understanding the capacity fade mechanisms, optimal conditions for synthesis and the nest route to mitigate low conductivity in this class of materials. Options like composites, elemental substitution, and tuning architectures [110,112,113] have been proposed but the challenges like reproducibility and scalability still hinder this chemistry to break through, although the results are promising. Figure 17 shows a comparison between four different large format commercial cells with LFP, NMC, and LTO chemistry, respectively. They have been cycled at 25 °C, with 80% depth of discharge and at the nominal C-rate, as specified by the manufacturer. Every 100 cycles, the cell has undergone a discharge capacity check to determine its State of Health (SoH). One can clearly see that the NMC is the least performing cell which reaches 80% SoH after 1800 cycles followed by the LFP 14 Ah. The LTO 15 Ah cells seem to be the better performing in terms of capacity retention while the LTO 16 Ah is following more or less the trend of LFP. The difference in the two LTO cells could be because of the cathode used in them, the LTO 16 Ah has NCO (lithium nickel cobalt oxide) as cathode whereas the LTO 15 Ah has NCA (lithium nickel cobalt aluminum oxide) cathode.

###### Electrolytes.

The enhancement of the battery concept is not only limited to different electrode materials, separators, and current collectors. The used electrolyte has a significant impact on the overall performance of the battery. According to Taige et al. [114] and Janek et al. [115], the requirements of the electrolyte can be summarized as follows:

• large electrochemical stability

• high thermal stability

• wide operating voltage range

• low vapor pressure

• high conductivity

• high capacity

• high cycling rates

In the literature, one can find several lithium salts such as lithium hexafluorophosphate (LiPF6), lithium tetrafluoroborate (LiBF4), lithium triflate (LiSO3CF3), and lithium tris(trifluoromethanesulfonyl)methide (LiC(SO2CF3)3) [116]. In Fig. 18, these different lithium salts are plotted with their respective electric conductivity (in mS/cm). Xu et al. [118] concluded that LiPF6 is the most common lithium salt in lithium–ion batteries and it has a high conductivity compared to many other electrolytes. However, during charging, the organic solvent forms an SEI layer with the graphite-based anode (<1.2 V), which decreases the conductivity of the electrochemical system. Moreover, lithium-ions cannot be used with aqueous electrolyte because lithium ions react with water and form lithium hydroxide (LiOH). While at a higher charging voltage (>4.4 to 4.6 V), the electrolyte decomposes and forms a new passivation layer on the active surface area. The electric conductivity of LiPF6 electrolyte is 10–1000 times higher than aqueous electrolytes as used in nickel–metal hydride and lead–acid batteries. During the last few years, significant attention has been paid to ionic electrolytes due to their higher thermal stability (nonflammable), wider voltage operating window, and higher conductivity by using optimal mixture of ionic liquids [119]. In this context, the electrolytes N-methyl-N-propylpiperidinium-bis (trifluoromethylsulfonyl) imide, n-hexyltrimethylammonium-bis (trifluoromethylsulfonyl) imide, and 1-ethyl-3-methylimidazolium-bis (trifluoromethylsulfonyl) imide have been developed with small capacity degradation and good cyclability [120122]. LiTFSI (bis(trifluoromethanesulfonyl)imide) is also another salt which could be used as an alternative; it enjoys the best thermal stability, solubility, ion pair dissociation, and SEI formation ability among many salts [123]. However, there are few drawbacks with this salt; one is the possibility of aluminum corrosion which could damage the cell since the positive electrode current collector is aluminum and the other is poor ion mobility. LiTFSI salt is commonly used in Li-sulfur cells; it inhibits the polysulfide solubility and has high electrolyte conductivity when mixed with different carbonate type solvents [124].

## Other Promising Technologies

###### Lithium-Sulfur.

Among the recent solutions of high density rechargeable energy storage for automotive applications, lithium-sulfur system is very promising. Especially on the account of the recent progress, lithium-sulfur can be seen as a most promising commercially viable product in the near future [125132]. Theoretically a lithium-sulfur system can provide an energy density as high as 2600 Wh/kg−1 (considering lithium anode and sulfur cathode) and in a practical system it can be as high as 600 Wh/kg [2], which will probably improve as the research progresses. Moreover, sulfur abundance in the nature makes the system economically beneficial. Lithium-sulfur is based on conversion chemistry and a typical system consists of a Li anode and S cathode. During discharge, at the anode Li metal is oxidized to Li+. At the cathode, solid sulfur (S8) is stepwise reduced to lower oxidation state and forms sulfide precipitates Li2Sx (where x values are 8, 6, 4, 2 and, 1 at different discharge state). Figure 19 is a schematic, explaining the general configuration and working principal of a Li-S cell.

Currently, commercially applicable lithium-sulfur batteries are not available. This technology is still under development for a long time now. However, Oxis energy is producing lithium-sulfur pouch cells since 2013 and is expected to produce commercially with an expected specific energy of 165 Wh/kg and cycle life of 1500 cycles. In addition, Saft Batteries have developed the first prototypes of cylindrical lithium-sulfur cells with an energy density of over 500 Wh/kg and lifetime over 500 cycles. Sion Power and BASF are some further commercial actors who are working on lithium sulfur battery technologies for a long time.

There are a number of technical challenges to be overcome to make it commercial. First, at the cathode, the low ionic and electric conductivity of sulfur and the polysulfides make the electric efficiency of the system very low (<70%). Second, the complex electrochemistry of the Li-S system. The polysulfides create a number of difficulties during operation. They (mainly Li2S4–Li2S8) dissolve in the conventional organic electrolytes, which can result in serious capacity fade [1,2,4]. Moreover, postdissolving, these polysulfides can migrate to the Li anode and upon reaction, can form Li+ movement resisting insulation layer of Li2S2, Li2S. This complex phenomenon is referred to as polysulfide shuttle [129,131,133136]. Third, the overwhelming volume change in the system, which can be as high as 79% [137,138]. Due to this volume change phenomenon, contact between cathode and current collector can be very loose and can worsen with cycling. As a result, the battery performance deteriorates and increases safety concerns. All the more, the presence of Li metal anode has its own technical challenges. These include the strong SEI formation and dendrite formation. This has been discussed earlier in the section related to other battery technologies. Finding a suitable electrolyte, which has similar properties of Li-ion and other Li metal systems, which can additionally act against the dissolution of polysulfide, is also an impending challenge.

In general, to overcome the shortcomings of low conductivity, sulfur composites are used, mostly sulfur–carbon. These composites are commonly synthesized in two forms, coated sulfur and sulfur-loaded porous substrates. Thanks to nanotechnology, for further advancements were made to overcome many technical challenges. According to recent findings, 2D graphene [139141], conductive polymer such as polypyrrole [142,143], and metal oxide such as TiO2 [144,145] coatings have not only showed improvement in terms of conductivity but also their flexibility, which has minimized the detrimental effect caused by volume change. A unique S-TiO2 yolk-shell nano-architectured cathode allowed Li+ transport but restricted polysulfide dissolution. With this configuration, an impressive cycle life of 1000 cycles was achieved [144]. Comparatively more focus has been put on porous substrates, and in particular on carbon-based sulfur composites. Macro to meso to micro (i.e., reducing pore size) has proven to be extremely beneficial in terms of performance, rate capability, and life cycle [125,126,128130,133]. For instance, in Ref. [146], reported that, confining sulfur allotropes in small pores within a matrix can diminish other undesirable allotrope formation thereby controlling polysufide shuttle phenomenon. Another direction of the cathode development is using metal sulfides such as Li2S, FeS2, and MoS2; beneficial improvement was reported in Refs. [125,126,128,130,131,133]. Development in cell design can also be beneficial. For instance, Su and Manthiram introduced a carbon interlayer between the cathode and the separator, which can preferentially block polysulfide shuttle, thereby improving the capacity and life cycle [147]. Figure 20 shows the discharge characteristics of lithium-sulfur cell where one can clearly see the loss of capacity from first and second discharge (C/30 rate) probably due to the polysulfide shuttle phenomenon. In a very recent report, a novel nitrogen and sulfur co-sulfur doped and 3D structured mesoporous carbon-based cathode has been reported [148]. The reported cell showed high rate capability, as high as 4 C with capacity fade of only 0.085% per cycle for over 250 cycles. Another recent report showed a cell with a specific capacity of ∼750 mAh/g for over 500 cycles and the capacity fade was limited to less than 25% [149]. Song et al. [150] managed to demonstrate a long cycle life Li/S cell by employing a cetyltrimethyl ammonium bromide (CTAB). CTAB is a surfactant which was coated on top of the sulfur to form a sulfur-graphene oxide (S-GO) nanocomposite cathode. It is reported that such an electrode can be discharged and charged at rates as high as 6 C and 3 C, respectively, with a long cycle life exceeding 1500 and only minimum capacity decay every cycle.

###### Lithium-Magnesium Batteries.

Magnesium-based batteries can be considered as one of the future successors, replacing lithium-based batteries. The factors therefore are its abundance in earth's crust and its bivalency. Low cost, higher energy density, higher theoretical capacity (2205 mAh/g), and high safety also make it a contender in powering of PHEV and BEV. Commercialization of these types of batteries has been plagued by the electrolytes and the nature of magnesium metal itself. Similar to an SEI layer in lithium ion batteries, a layer is formed on the magnesium metal anode upon the decomposition of electrolyte [151]. However, unlike the SEI layer in the lithium system, this layer does not allow the Mg2+ to pass through, thus blocking the ion pathway. Hence, pure magnesium metal anodes had to be overlooked as was the case with lithium metal anodes (dendrite formation). Two approaches can be employed to accomplish the goal of producing high energy rechargeable magnesium batteries, either insertion type cathodes with high voltage and low capacities or conversion type cathodes like Mg/S systems. Problems related to the mentioned routes are sulfur and polysulfide dissolution in the case of Mg/S and electrolyte stability in the case of intercalation cathodes. Toyota research institute [152] demonstrated the use of Bismuth and Tin-based insertion anodes for magnesium ion batteries. They achieved 215 mAh/g (initial—298 mAh/g) after 100 cycles for Bi0.88Sb0.12 in Mg (TFSI)2, Bis(trifluoromethanesulfonyl) imide in acetonitrile. They also proved that by using a different electrolyte for Sn and Bi, they could get higher capacity values and lower insertion/extraction voltage. Even though the capacity value achieved could be increased, the retention was low due to the volume expansion of the anodes. Toyota, Sony, LG, Hitachi, and Pellion Technologies have made significant efforts in order to provide breakthrough in this field. There is now a drive and it has been suggested that these systems can be tuned for rechargeable energy storage systems through refueling [153]. Such systems are still used as primary batteries and need more research to be done in order to convert this into a rechargeable system where anodes should be reversible, the cathode having redox catalysts, and an electrolyte sustaining such an operation. Future development should be focused toward finding an electrolyte, which can handle the high voltage cathodes and facilitate high reversibility of magnesium; and in case of cathode development, finding a stable insertion cathode, which can accommodate Mg2+ insertion and exertion.

###### Solid-State Batteries.

Solid-state batteries are probably the closest successor to conventional lithium-ion batteries in the market. In the past few years, big companies like Samsung and Toyota have shown lot of interest in this technology. According to some recent technical reports, solid-state batteries are being tested for wearable devices. For instance, vacuum cleaner producer Dyson recently invested \$15 M on solid-state battery manufacturer Sakti3 for their future cordless vacuum machines (Source: Technology Review). SEEO also is involved in producing DryLyteTM batteries which is claimed to be highly reliable with extended cycle life compared to liquid electrolyte-based batteries, which could be used to power EVs [154]. Solid-state batteries basically differ in the type of electrolyte compared to other battery technologies. Conventional electrolytes are generally in liquid forms while in the case of solid-state batteries, the electrolytes are in solid form (e.g., glass, ceramic, new organic structures of thermotropic-ionic liquid crystals). There are numerous benefits of solid-state batteries which is why impactful interests have grown on this technology. With conventional liquid electrolyte system, leakage can cause safety problems, which can be overcome by solid-state electrolytes. If lithium metal is used as an anode, the energy density can be highly increased (Fig. 21). However, due to the high risk of reaction with conventional electrolyte, lithium has to be protected by an additional layer which results in poor performance. Moreover, the dendrite formation of lithium can cause short circuit. All these problems can be eliminated by using solid-state electrolytes. Additionally, conventional electrolytes cause self-discharge, but as solid electrolytes only allow ion (Li+ for example) to transfer, minimizing self-discharge. Moreover, cell can be designed in desirable shapes with flexible solid-state electrolyte. Another benefit is the higher voltage operating window of solid-state electrolyte compared to the conventional electrolytes. Although solid-state battery possesses great potential, there are challenges faced in developing commercial grade batteries. Most important is the ionic conductivity of the electrolyte at room temperature, which is poor compared to liquid electrolytes. For instance, optimized liquid electrolyte has a conductivity of 5–10 mS/cm where highest achieved conductivity (see Figs. 22 and 23) of solid-state electrolyte is 2 mS/cm [158]. Other problem is the higher manufacturing cost of such solid electrolytes. Due to rigidity of these electrolytes, maintaining good interfacial contact becomes a challenge. In Ref. [159], the authors have managed to quantify the polarization losses occurring at the organic liquid electrolyte/single-ion conductor interface using a customized diffusion cell. The polarization losses were found to be very significant, which could lead to decrease in cell voltage. Ongoing research activities suggest that solution will be found in a relatively short term (5–10 years). According to the thickness, solid-state electrolyte can be broadly divided into two types—bulk solid-state electrolyte (several hundred micrometers) and thin-film solid-state electrolyte. Most commonly used solid-state electrolytes are the solid polymeric electrolytes (SPEs). In order to increase the conductivity, SPEs are complexes with inorganic salt. One of the important type is the NASICON type electrolyte (General formula LiM2(PO4)3, where M is usually Ti, Ge, or Hf [160]. Solid-state electrolyte with garnet structure (general formula A2B2C3O12) has also been frequently used [160]. Vanadium-based electrolytes were studied as well [161]. Apart from these inorganic solid-state electrolytes, organic solid-state electrolytes have also been studied and used. Kim et al. have developed polymer solid electrolytes, more particularly, polyethylene glycol (PEG) coupled with organic hybrids and plasticizers [162], which showed better conductivity (∼1 mS/cm) at room temperature. Among the thin-film solid-state electrolyte, LiPON (lithium-phosphorus-oxy-nitride) is the most commonly used [162] due to its conductivity and stability. With some modification, thin-film solid-state electrolyte showed optimal conductivities compared with bulk solid-state electrolytes. Small size (few Ah) thin film Li battery is being used in portable electronics but high power application commercial cells are not available yet. Solid electrolyte has been implemented with Li and non-Li-based systems. Lithium-based systems are more interesting for the near future as conventional Li-based systems are already used in many applications. Table 4 represents an overview of Li-based solid systems with some main output characteristics. Nonlithium-based systems have also been tested with solid electrolytes. Among them, ceramic-based electrolytes are most common. As some of these systems, for example, sodium-based batteries, work at a high temperature (about 270 °C), these solid-state electrolytes' conductivity increases to optimum level (∼2 mS/cm) and shows superior performance [168]. In order to overcome the limitation of solid-state battery systems, possible direction of development is to produce very thin electrolyte along with decreasing the thickness of the other components. This will allow to further decrease the internal resistance and in this way the power performances can be enhanced. Modeling of optimal property of a solid-state electrolyte will also play a vital role in this further development.

###### Metal-Air Batteries.

Metal-air batteries use metal to store energy. The electron flow is achieved through conversion reactions. In this category, lithium, zinc, iron, and aluminum are metals that can be used as anode. And, air cathode or oxygen from the environment acts as cathode. Among all the metal-air batteries, lithium-air and zinc-air batteries show most promising performances. Following this, Fig. 23 illustrates an overview of the limiting energy densities of various rechargeable energy systems where one can see the potential of metal-air batteries over other conventional systems.

Lithium-Air: Very high theoretical specific energy is the most attracting property of the lithium-air battery technology. In fact, theoretical specific energy of a lithium-air system is more than 11,000 Wh/kg if the mass of the oxygen is ignored, which is close to gasoline and ∼10 times of the state-of-the-art lithium-ion technology [169172]. Additionally, low cost of active materials makes it commercially interesting. Rather than intercalation of Li+ ions in conventional lithium-ion batteries, a lithium air system consists of a lithium metal anode and a porous air cathode where reduction of oxygen takes place during discharge. Two most basic and primitive configurations differ in the type of the electrolytes used and therefore also the discharge products. Li-air system with aqueous electrolytes produces soluble LiOH through oxygen reduction reactions (ORRs) upon discharge, whereas solid Li2O2 is produced in the nonaqueous or aprotic system, and deposited on the pores of air cathode as shown in Fig. 24. Respective discharge products are oxidized upon charging through oxygen evolution reactions (OERs). The aqueous system possesses inferior energy density and high risk of contamination of lithium anode, thus favoring the aprotic system in terms of both performance and safety [171,174]. In the case of aprotic lithium-air, the main challenges are that, at one hand the anode suffers from water contamination and on the other hand, the process at the cathode level is affected by clogging with Li2O2 due to insolubility of Li2O2. Clogging of the cathode pores by Li2O2 gradually restricts the access to active reaction sites and thus limits the discharge capacity. Lack of suitable electrolytes, dendrite formation at the anode, formation of irreversible intermediate discharge product (e.g., Li2O), lack of understanding of the mechanism of ORR and OER are the other crucial challenging factors because of which the present Li-air cells show low efficiency, low cycle life, and poor power capability [169176]. Significant progress has been achieved to overcome these challenges over the last few years [177]. Conventional carbonate, ester, ether, amides, sulfoxide, sulfones, ionic liquids-based electrolytes have been used and it was found that they are either unstable against the superoxide species in short or long term, or have high risk of contaminating anode, have low solubility, have high volatility, or have poor ionic conductivity [176]. Polymer and ceramic-based solid-state electrolytes showed better stability. For instance, lithium conducting glass electrolytes, NASICON, LISICON, etc., not only showed better chemical stability but also reduced the risk of anode contamination [178,179]. Another very fruitful solution is to use a two-compartment hybrid lithium-air system (see Fig. 25). In this system, the cathode compartment consists of an aqueous electrolyte where solubility of discharge product is an issue and the anode compartment consists of a nonaqueous electrolyte where anode contamination is the main issue [176,180]. Additionally, hybrid electrolyte such as ionic liquid-organic electrolytes blend, conventional electrolyte with additives showed better performance in terms of chemical stability and solubility of Li2O2 [176]. Nanostructure morphology tailoring of carbon and graphene such as increasing porosity and pore widening showed better electric efficiency and power capability by offering more space for Li2O2 precipitates [169,181184].

Although the lithium-air system possesses very promising specific energy characteristics, due to the practical challenges, commercial implementation will most likely take another decade if not more. Nevertheless, the contemporary significant research effort might achieve a breakthrough for this technology. The most known research program in this field was BATTERY500 of IBM, which has been stopped due to the technical challenges that could not be overcome.

Zinc Air: Zinc-air batteries are similar to lithium-air batteries in working principle, as shown in Fig. 26. However, Zn-air systems possess certain advantages over lithium-air systems.

Zinc is stable against moisture; thus, it is not needed to protect the Zn anode while it is needed for Li anode. Thus, manufacturing environment can be more easily designed. For the same reason, the cell design can be much simpler compared to Li-air [186]. Zn is cheaper than lithium; therefore, the system can be comparatively cheaper. In the principle reaction, Zn can donate two electrons, which is very beneficial in terms of energy density. Compared to Li-air system, Zn has lower nominal voltage (∼1.65 V) which means lower energy density. However, the biggest problems with Zn-Air systems are very poor reversibility. Due to high overpotential of ORR and OER, the electrical efficiency of the system is inferior compared to Li-air (less than 60%). Zinc corrodes highly in the acidic media. Alloying and coating can improve the electrochemical performance in this sense. Moreover, Zn2+ migration from anode to cathode can occur and can decrease the capacity of the cell. Separator with proper pore size can help in this regard. Overall, due to these technical challenges, Zn-air system is still used as primary battery in hearing aid devices, and no commercial product of the secondary type is available [187]. EOS introduced a prototype zinc-air battery, which provides over 10,000 cycles. The cost estimated is \$160/kWh. However, the real performances of this prototype are still unknown. Electric Fuel Ltd. has developed a mechanically rechargeable zinc/air battery that power electric vehicles for more than 200 km on a single discharge. This technology has been tested on-road in the service of German postal delivery.

For metal-air system, success lies in the electrolytes. Better understanding of reaction mechanism and mathematical modeling can help to identify suitable electrolytes. Solid-state electrolytes can play a vital role here. Modification of the anode at micro and nano level can be very beneficial. Proper designing of the cell is another potential direction for development of this technology.

###### Sodium-Based Batteries.

There are mainly two types of sodium-based batteries. One type is the high temperature large-scale sodium battery. This type works at a high temperature (>300 °C) and is mainly used in large-scale stationary energy storage applications. Other type of sodium battery is Na-ion battery, which follows the similar principle insertion chemistry as Li-ion at room temperature and which could potentially replace Li-ion battery.

High Temperature Sodium Battery: High temperature sodium batteries (HT–NaB) offer a cost-competitive solution for large-scale stationary storage. The general battery system consists of negative sodium electrode and beta alumina as both electrolyte and separator. Depending on the positive electrode, there are two main types of sodium battery systems. Sodium-sulfur consists of liquid sulfur and sodium halide consists of liquid halide (chloride) as liquid positive electrolyte. Both systems work above 300 °C. These battery technologies exist since early 1970s. Due to practical issues, the development discontinued recently. As sodium is abundant in the nature, system based on Na would be much cheaper than lithium technology. Moreover, sodium is less reactive and has a longer cycle life (∼4500 cycles) compared to some lithium-based systems [188]. Because of the fact that the anode and cathode are in a liquid state, careful cell design should be implemented in order to avoid leakage. And as the operating temperature is above 300 °C, this system is not suitable for portable application. Additionally, the practical energy density is around 200 Wh/kg, which is unattractive for portable system. However, for stationary and grid applications, these systems are very attractive, in terms of cost of components and manufacturing [188190]. Sodium-sulfur system are in application, since 2010 and meanwhile 316 MW storage capacity has been built and deployed by NGK according to Electric Power Research Institute (EPRI) research. Sodium-chloride systems are more expensive than sodium-sulfur as they incorporate expensive cobalt. This technology is novel compared to sulfur system and further improvement is needed. But sodium-chloride system manufacturing is less complex, which can compensate for extra cost of the cobalt. General Electronics is the biggest producer of sodium-chloride batteries.

Sodium Ion battery: Recently, interests and effort have increased significantly for the development of Na-ion battery technology. Na ion possesses the similar potential of insertion chemistry and is much safer and less expensive than Li chemistry, making it the center of the trending interest. With Na+, limitation of Li metal chemistry can be suppressed. Compared to Li, Na does not have the tendency to alloy with Al; thus, current collector can be made of Al. For the same reason, the Na-ion battery can be discharged to zero volts. As Li is used in different configurations such as Li-ion, Li-Air, and Li-S, sodium has been implemented in similar configurations. According to recent reports, practical energy densities of Na-ion, Na-Sulfur (S8), and Na-air are, respectively, ∼150, 300, and 400 Wh/kg [190,191]. Until now, at laboratory scale, a working Na-ion battery has been achieved with operating voltage of ∼3 V. Although this technology has numerous advantages, commercialization will take time. One of the biggest concerns is the size of Na+ that makes it difficult for insertion mechanism. The energy density is much lower than similar Li system with a lower operating voltage window. Power density is not comparable with that of Li system. Significant number of work is currently ongoing to overcome these challenges. For high temperature Na batteries, the future direction is to minimize the operation temperature. In this regard, several works have been undertaken. One solution is to find intermediate temperature working material for Na batteries. Such a battery has been reported by Pacific Northwest Laboratory for which the operating temperature is ∼150 °C [188]. Another direction of improvement is the electrolyte. NASICON-Polymer electrolyte can be used for improved performance [185]. For sodium-ion battery, the development is still in the early stage. Possible direction of improvement is metal oxide Na cathode along with solid-state electrolyte.

###### Redox Flow Batteries.

Redox flow batteries differ from conventional batteries in that their active material is in the form of redox couple solutions. They are generally stored in external tanks and pumped through a stack of electrochemical cells where the solutions are separated by an ion exchange membrane. Due to active material in the liquid state, the systems offer long life. Two principal types of flow batteries exist today, zinc and vanadium based. Zn-based flow batteries are one type of flow battery where Zn is deposited on the negative electrode during charging and the capacity of the system is determined by the amount of deposited metal. Zn/Cl2 and Zn/Br2 are the prominent Zn-based flow batteries. Other type of flow battery contains two liquid redox couple solutions and the capacity of the system is determined by the size of the external reservoir of the redox couple solution. This type of battery is called redox flow battery (RFB). Zn-based flow batteries, especially the Zn/Br2 technology is available since long time and there are several companies producing commercially applicable Zn/Br2 batteries. Australia-based “Redflow” is producing commercial grade 48 V DC Zn/Br2 battery packs and these have been tested extensively. These Zn-based batteries producing companies, currently engaged in the commercialization of Zn/Br2 batteries, include Premium Power in the U.S. and the Australian company ZBB Energy that has recently moved its activities to the U.S. [192]. Most recent work from Pacific Northwest National Laboratory (PNNL) shows a Zn-based flow battery with an energy density of 160 Wh/l [193]. A lack of funding on this technology is restricting this technology to come into the market.

Redox flow batteries have several key advantages over other battery technology:

• no solid-phase changes occur,

• can be fully charged and discharged,

• almost zero self-discharge,

• electrolyte flowing through the stack can act as a coolant to prevent thermal runaway,

• low cost system for large storage systems, and

However, the energy density is too low for an EV application. Fe/Cr, sulfur-bromine (S/Br), and vanadium redox flow battery (VRBs) are the promising candidates of this technology [192]. Commercial Fe/Cr batteries have been in the market by an U.S.-based company Deeya in early 2000s; but later, they switched to vanadium redox battery under the name Imergy energy [192]. Most promising RFB is vanadium-based batteries whose schematic is shown in Fig. 27. According to the development phase of the components, there are three generations of vanadium RFBs. Several companies in U.S., Japan, and Europe have invested during last two decades in this battery technology. G1 batteries are commercially suitable for large-scale stationary energy storage. From the cost perspective, the cost is near the cheapest battery lead-acid batteries. It also showed unprecedented life cycle of over 200,000 cycles. However, it could not provide the demanded energy needed for EVs. Thus, direction moved for second- and third-generation vanadium RFBs. There is a lot of development being done, but still the energy density is around 25–40 Wh/kg. It is Important to mention here that the energy efficiency of vanadium-based batteries is around 65–75% and for Zn/Br2 batteries, the energy efficiency is around 60%. This technology is a success in stationary applications and can be transferred or adapted to EV application through further engineering. Some prototypes for cars (e.g., nanoflowcell) and busses with redox flow technology have been proposed.

###### Hybrid Batteries.

The above described technologies are all based on faradaic reactions, meaning that the charge of the ions within the electrolyte is transferred to the electrode according to the intercalation or redox process. Electrical double-layer capacitors (EDLCs) on the other hand store their charge electrostatically at the electrode–electrolyte interface. The ions in the electrolyte are adsorbed at or desorbed from the interface of the activated carbon (AC) electrode, meaning that no charge transfer reaction occurs. This process is thus nonfaradaic, as illustrated in Fig. 28. This suggests that EDLCs can obtain a very high power density (10 kW/kg) and lifetime (over one million cycles), but as a drawback have only a limited energy density of around 5 Wh/kg [196200]. In order to increase their energy density, several concepts are currently under research: new materials such as MXenes [201], pseudo capacitors that use redox-active materials like RuO2 and MnO2 [202,203], and hybrid batteries. Hybrid batteries combine a (battery-like) insertion electrode with an (EDLC-like) AC electrode, as illustrated in Fig. 28. This concept results in a rechargeable energy storage systems (RESS) that combines the high power capability of EDLCs with the high energy density of LIBs [32,204,205]. This can clearly be seen in the Ragone plot of Fig. 29, where the hybrid capacitor technology has its own specific place on the Ragone plot in between the LIBs and EDLCs. The hybrid capacitor technology is thus very interesting for those applications that need a RESS with a high power capability combined with the need for a considerable amount of energy content, such as heavy transport applications like hybrid busses, trams, trains and metros, and uninterruptible power supply (UPS) systems. The concept of hybrid batteries was applied for the first time in the academia by Amatucci, who presented in 2001 the combination of a lithium titanate Li4Ti5O12 negative electrode with an AC positive electrode, obtaining a cell with an energy density of more than 20 Wh/kg [206]. Today, several companies are already commercially offering specific types of hybrid batteries. JM Energy, as subsidiary of JSR Corporations, started commercialization of lithium-ion capacitors (LICs) in 2007. As shown in Fig. 28, JM Energy uses a prelithiated graphite negative electrode in combination with an AC positive electrode. The prelithiation of the graphite electrode allows decreasing its reference potential meaning that the cell voltage can be increased up to 3.8 V, compared to a maximum voltage of 2.7 V for typical EDLCs. LICs, however, also have a minimum voltage of 2.2 V, while EDLCs can be discharged up to 0 V. The Ukrainian company Yunasko offers, aside from several EDLC cells, also a hybrid capacitor with a capacitance of 5200 F. It is a pouch-type device with a very low mass of only 85 g and as a result a very high energy density of 30 Wh/kg according to Burke et al. [207] and 37 Wh/kg based on the provided datasheet [208]. The high energy content of the Yunasko cell, however, has a drawback in terms of limited cycle life of around 10,000 cycles, while the JM Energy cells can obtain a cycle life of over 1 × 106 cycles. A comparison of the highlighted cells has been provided in Table 5, where the characteristics of the Maxwell 3400 F cell are presented as well. Future development will encompass the usage of new materials in the hybrid capacitor technology and the further optimization of the prelithiation process. From the point of view of new materials, several examples are already observed in the academic literature, such as the nanohybrid super capacitor, which can obtain 30 Wh/l, presented by Naoi et al. [209]. The hybrid capacitor technology is still a “young” technology. Their performance has already been demonstrated in several cases, such as in the hybrid bus technology and UPS installation, but further market acceptance must be gained in order to be implemented on larger scale.

## Battery Material Resources

###### Resource Constraints.

The main material concern when it comes to traction batteries relies on lithium ore availability. In recent years, lithium-based batteries have been gaining relevant importance in the mobility sector, where they equip most of the electrified vehicles. Concurrently, in the past years, awareness on the available lithium reserves has risen. As effective vehicle range increases, the demand for battery grade materials follows up the trend. This trend is summarized in Table 6 and Ref. [211], where the energy content and the needed lithium-ion per specific electric driving range are displayed. Figure 30 shows the global EV forecast in 2015 by Bloomberg new energy finance; electric vehicles would be 35% of the new car sales by 2040. According to Internal Energy Agency (IEA), the global PHEVs and BEVs in 2015 will increase from 6.9 million vehicles to 106.4 million vehicles. Based on these data, the needed lithium has been calculated over time. As it is observed, the total required lithium carbonate demand will increase significantly based on the prediction of IEA. In 2050, the production could be over 0.57 million tons as shown in Fig. 31, which is around 3% and 4.1% of the estimated lithium resources, respectively, according to Tahil et al. and USGS. However, resources and reserves are a different concept according to some authors [213]. Resources are considered “geologically assured quantities that are available for exploitation” while reserves are “the quantity that are exploitable in the current technical and socioeconomic conditions.” This consequently means that 3–4.1% of estimated resources will become a much higher amount of reserves. The geological spread of lithium ore can be seen in Fig. 32 according to Ref. [214].

The situation in the future can be critical due to geopolitical reasons as the lithium resources are mostly allocated in the countries like Argentina, Australia, Austria, Afghanistan, Brazil, Bolivia, Canada, Chile, China, Finland, Russia, U.S., Congo, and Zimbabwe. Therefore, the price of lithium carbonate could increase significantly as a consequence of demand increase. Other materials that are crucial in lithium-ion battery technology include manganese, cobalt, and nickel. Manganese is one of the main materials in use in NMC and LMO batteries for battery electric vehicles. 75% of the manganese resource is identified in South Africa and it holds currently 150 million tons of reserve followed by Ukraine (10% share in world resource). Manganese is a material that never has been considered as a critical material for energy storage technologies. Compared to manganese, cobalt is a key component in the NMC, LCO, and NCA batteries. Congo is the main supplier of this material—56,000 Mt in 2014 followed by China (7200 Mt). The worldwide reserves of cobalt are estimated to be about 7,200,000 Mt. This material is not only of interest for mobile and stationary applications but also for portable applications. Therefore, this material is a critical one when we start with mass production of large batteries for stationary and automotive applications. Nevertheless, the recycling of cobalt is very efficient and this process will continue in the future to maintain the use of this material for portable applications. Nickel prices increased by 12% in 2014 despite weak economic conditions in some developing countries and parts of the European Union. In May 2014, the London Metal Exchange (LME) cash mean for 99.8% pure nickel peaked briefly at \$19,434 per metric ton. The major nickel mine production countries are Philippines, Russia, Indonesia, Australia, and Canada. The total worldwide reserves of nickel are estimated by 81,000,000 Mt. EU does not consider nickel as a critical material since it can be obtained from different sources in the world. However, the cost of nickel can increase as the cobalt one, which makes this material in the long term for energy storage technologies less interesting. From this point of view, in the long/near future, it would be wise to focus on new technologies like zinc, sulfur, and tin-based materials and commercializing them because of their abundant availability around the world. For these materials, there are negligible resource constraints compared to lithium specificities.

###### Environmental Burden of Lithium Battery Materials.

The environmental burden of lithium battery production is mostly spread throughout all the impact categories that comprehend every impact assessment methods. As Oliveira et al. [214] did during an assessment on LMO and LFP environmental impacts using ReCiPe [215], it was found that impacts per energy unit (kWh) can be significant from one chemistry to the other by up to three fold. Impacts during the usage of these technologies is normally tied to electricity production and charge-discharge efficiency, whereas manufacturing and recycling efforts are different for both technologies. For example, LFP has a three times higher impact for recycling in the climate change category than LMO and a negative impact (beneficial) opposed to LMO in the human toxicity category, Figs. 33 and 34. Also, Oliveira et al. [214] performed a detailed manufacturing analysis on LFP batteries by trying to identify environmental hotspots. This lead to the identification that roughly 38% of the burden is originated from the positive electrode manufacturing processes. In addition, the energetic demands of the manufacturing process account for 21% of the impacts followed by the cell container manufacturing burden. Analyzing electrode (positive) materials, LFP involves high costs due to its specific product chain. This high cost makes the recycling process even more relevant in an effort to close the material loop and thus reduce costs for cell manufacturers. Similar results have been found by Majeau-Bettez et al. [216] and Notter et al. [217]. Nevertheless, a transversal issue to all the lithium-based chemistries is that most of the minerals used are not primarily extracted and therefore, depend on the three main extraction industries. They are subproducts of the processing of iron, copper, and aluminum. These industries do have the monopoly on the prices of these subproducts and therefore, recycling has never been so relevant.

The geographical dispersion of these ores is of relevance as poorer countries with abundance of resources have worse development indexes than those well endowed. Most of the times, the mineral extraction activities are related to corruption and armed conflicts—as a consequence, the supply of these ore materials is constantly at risk. The economic importance and supply risk of these materials are displayed in Fig. 35. Resource extraction contributes heavily towards total environmental impact which can be linked to the extraction companies. As an example, other chemistries such as NMC contain fundamental materials that are originated from countries where accountability is not a priority. This includes sub-Saharan countries such as Congo and Angola where the extraction industries of copper and cobalt account for a significant share of the economy. Most of the companies acting in these scenarios do neither document their processes nor are they accountable for environmental stresses. They make profit of the socio-economic situation of the country as well as relaxed regulations. This issue creates difficulty for the analysis of batteries that contain these materials. Sometimes, by default, assumptions are taken in order to overcome this issue. The environment can sometimes be affected, not by excess but by shortage. Alongside this problem, vehicle manufacturers (where most modern day lithium-based batteries end up) struggle to maintain the companies that contribute to their product chain away from practices that can harm their business.

## Battery Cost Evolution

It is a well-known fact that the cost of the battery is one of the most crucial factors affecting the commercial development of BEVs and PHEVs. In literature and in general practice, several approaches could be found regarding the cost calculation of batteries. One of these approaches is presented below [19]. Display Formula

(7)$Wh=(/gmAh/gmanode)+(/gmAh/gmcathode)$