1 Introduction

The energy consumption related to human activities always involved a specific energy supply chain, which provided to the final users the exact amount of energy required at a specific time. Since it is not always possible to match the energy supply with the user’s demand, there is a need for storing energy to compensate this mismatch. The storage may be required with a large diversity of durations, ranging from fraction of seconds to months or even years. Different energy carriers involve multiple storage solutions, based on limits and opportunities related to the form of energy that is stored (chemical, potential, kinetic, electro-static, etc.), as well as on technical and economic features of the available storage technologies.

The easiest energy storage usually happens with fuels, especially solid and liquid, which can be generally stored in their normal form without the need of specific solutions. While attention must be paid in avoiding potential self-combustion, chemical degradation, or phase change, solid and liquid fuels are usually stored in simple tanks (eventually cooled or heated in particular climate conditions).

Additional requirements are usually needed for gaseous fuels, mainly natural gas. Due to its low volumetric density, its transportation and storage are usually performed either by compressing it at high pressures or by liquefying it with the need of providing continuous cooling. Natural gas storage is usually performed on a seasonal basis, to match the continuous supply with the fluctuating demand driven by different weather conditions throughout the years. Such storage strategies usually involve large-scale underground formations, either depleted reservoirs or saline formations. The low energy and fuel losses are generally compensated by the significant economic savings that can be obtained with continuous upstream operations for natural gas.

Another energy carrier that is commonly stored is heat, usually in the form of warm or hot water, either in large-scale facilities connected to district heating networks or industrial users or at small-scale heat storage systems for domestic users. Heat storage is mostly used to exploit the better efficiency related to heat generators operating at constant load, especially biomass boilers and heat pumps. However, large seasonal underground systems are being used in some countries to store solar energy in summer and supply district heating in winter. Some systems exploit the ground as storage medium, while others rely on very large water volumes (Bott et al. 2019). For small-scale storage, alternative technologies based on phase-changing materials are the objective of multiple research efforts, although commercial applications are still limited.

Electricity stands out among the most difficult energy vectors to be stored. Electricity storage solutions are usually relying on its conversion to another form of energy. With the exception of superconductivity, other current technological solutions rely on chemical, mechanical, gravitational, or electro-static forms of energy. Nevertheless, electricity storage systems are strongly needed to guarantee the continuous balance of the power grid and provide reliable and effective service to the final users. For this task, a wide range of services is required, which are usually categorized with respect to storage duration: from few seconds or minutes for frequency control; to energy transfers across weeks, days, or day-night (also called arbitrage); and to the need of providing UPS (Uninterrupted Power Supply) for industrial consumers connected to the grid (Crampes and Trochet 2019).

Each available storage technology is usually tailored to a preferred application, based on technical limitations, design choices, and economic features. Today, most electricity storage worldwide is performed by pumped hydro systems, which rely on a mature technology with lower costs in comparison with the alternatives. Although pumped storage may be used also for frequency regulation, the flexibility provided by its potentially long discharge time (up to a few dozens of hundred hours) is usually exploited for arbitrage. Frequency control is provided through flywheels but more often by backup power generators. Batteries are somewhat in between, since they have discharge times that usually reach some hours, but at the same time they are responsive enough to provide frequency regulation services. Compressed-air storage systems have similar applications than pumped hydro, but due to limited available sites few applications exist.

While most storage systems are mature technologies, there is currently an interesting potential in the deployment of electric batteries, especially based on lithium-ion. The two leading drivers are the additional flexibility required by non-dispatchable renewable sources (mainly solar and wind) and the strongly decreasing cost expected by massive upscaling of battery manufacturing for electric vehicles. Although other chemistries may prove to be disruptive in the future, the current choice appears to be firmly oriented toward Li-ion, which is the preferred choice of numerous large-scale factories worldwide (so-called gigafactories). Their modularity also allows a large range of applications, from utility-scale grid storage to beyond-the-meter batteries for final users, usually coupled with distributed PV generation.

For these reasons, this chapter focuses on Li-ion batteries, given their expected central role in the future power systems. Alternative chemistries will be briefly mentioned, with the aim of highlighting the potential advantages they may provide. Section 2 provides a technological perspective to highlight the main aspects that are involved in battery design, deployment, and operation. Section 3 focuses on battery economics, with attention on the manufacturing supply chain and on the sizing and operational logics. Finally, Sect. 4 closes the analysis by recapping the main take-aways, together with some policy implications.

2 Battery Technologies

Different technologies exist for electric batteries, based on alternative chemistries for anode, cathode, and electrolyte. Each combination leads to different design and operational parameters, over a wide range of aspects, and the choice is often driven by the most important requirements of each application (e.g. high energy density for electric vehicles, low cost for stationary storage, etc.). The current rise in battery manufacturing capacity worldwide is associated with Li-ion batteries, which are meeting the requirements of the electric vehicles (EVs) industry and offer a viable solution also for stationary storage applications, both for utility-scale batteries and behind-the-meter distributed storage.

The historical trend of global stationary storage capacity (see Fig. 14.1) shows an increase in recent years, from around 0.6 GWh in 2010 up to 3.5 GWh in 2017. While up to 2010 most of the capacity was relying on sodium batteries, in 2017 almost 60% of the total capacity is made up of Li-ion batteries (figures may slightly differ when considering output power, since the energy/power ratio is usually different from one technology to another). This rise is due to different factors, but the most important is surely declining costs driven by manufacturing upscaling of this technology for use in EVs, as is further explained later.

Fig. 14.1
figure 1

Evolution of installed capacity for stationary storage (utility scale), per technology. (Source: Author’s elaboration on (Tsiropoulos et al. 2018))

Figure 14.1 is limited to utility-scale capacity, while there is also a growing, although much more difficult to quantify, amount of behind-the-meter storage.Footnote 1 Estimates for 2016 range from 0.5 to 2.4 GWh, depending on the source, limited to distributed storage operated by residential, industrial, and commercial users. This capacity is made up of a large number of storage systems with small capacity, usually coupled with local generation from RES (mostly solar). While utility-scale batteries are usually managed centrally, an optimized operation of the distributed energy systems requires the operation of smart grids and networks supported by digital platforms (such as virtual aggregatorsFootnote 2).

It is important to highlight that stationary storage may refer to different services for the power network, at both the transmission and the distribution levels, which differ based on the response time of the batteries, the discharge duration, and the size of the system. The applications may include services for the transmission grid (arbitrage, frequency regulation, peak shaving, black start, and rampingFootnote 3) or for the distribution grid and users (voltage support, balance management, uninterruptible power supply (UPS), and support to self-consumption from PV generationFootnote 4).

The following section will describe the main characteristics of the most significant available technologies, not only with a strong emphasis on Li-ion batteries but also with a discussion of the main alternatives: lead-acid (Pb-A) batteries, sodium-sulfur (Na-S) batteries, and vanadium redox (V-R) flow batteries. The main characteristics of different technologies are summarized in Table 14.1.

Table 14.1 Main characteristics of different battery technologies

As already anticipated, each battery shows peculiar parameters that are tailored to specific applications. Particularly, the energy/power (E/P) ratio is crucial for the choice of the application, and while there is some room for adjustment by considering specific design parameters (such as electrodes thickness in Li-ion batteries), each technology usually fits best in a specific application as presented hereafter.

2.1 Li-Ion Batteries

Li-ion batteries are a recent technology, initially developed at Bells labs in the 1960s and first commercialized by Sony in 1990. The Nobel prize in Chemistry in 2019 has been awarded to J. B. Goodenough, M. S. Whittingham, and A. Yoshino for their crucial role in the development of Li-ion batteries at different steps (Nobel Media AB 2019). Their success for portable electronics has been mainly triggered by high cycle life, high energy density, and high efficiency, although at a higher price in comparison with other solutions.

Li-ion batteries were mostly applied to portable electronics (including laptops, phones, etc.), until the rising interest in EVs triggered a significant deployment of batteries, whose price decreases also helped their increased sales for stationary energy storage and other applications (including medical devices, gardening tools, and electric bikes) (Fig. 14.2).

Fig. 14.2
figure 2

Evolution of Li-ion battery sales worldwide. (Source: Author’s elaboration on (Tsiropoulos et al. 2018))

Thanks to their superior performance, they represent the most interesting technology for research and development. In particular, most research is focusing on alternative cathode chemistries to improve energy density and safety or reduce cost through limited use of specific materials (especially cobalt). Other areas of research include anode and electrolyte materials and manufacturing processes.

2.2 Other Battery Technologies

2.2.1 Lead Batteries

Pb-A batteries are the most mature and diffused battery technology in the world, with their first applications dating back to the 1860s. The extensive research that has been made on many different aspects now guarantees low costs, although with limited life cycles and energy density. Specific additives are available to reach specific objectives, such as reducing the self-discharge or decreasing corrosion issues (Leadbetter and Swan 2012). Lead batteries are seldom used for heavy cycling applications, but they are generally suitable for infrequent cycle applications such as peak shaving or uninterruptible power supplies. Large batteries have been installed as case studies in different countries, up to 20 MW and 40 MWh, demonstrating good performance over several cycles, although requiring appropriate energy management methods. Notwithstanding the technology maturity, research is still active in different domains with the aim of decreasing costs and addressing specific challenges, such as longer lifecycles or more accurate determination of the state of charge (SOC).

2.2.2 Sodium-Sulfur Batteries

Na-S batteries are another relatively new technology, having been developed from the 1960s to the 1990s. While they were initially investigated for electric vehicles without much success, they eventually became among the lowest-cost options for grid storage and renewable applications. The operation of Na-S batteries involves peculiar aspects, including the need of high temperature operation for liquid sodium (300–350°C) and the potential very high reactivity of sodium with air in case of containment losses. While the inefficiency during the operation is generally enough to keep the sodium at the right temperature without the need of an external energy supply, in case of non-operation the battery records up to 20% of daily capacity losses due to heat dissipation. Existing installations have grown rapidly in the last decades, with the largest system for stationary storage reaching to date a capacity of 34 MW and 245 MWh coupled with a 51-MW wind farm to stabilize its power output (Leadbetter and Swan 2012).

2.2.3 Flow Batteries

The most diffused technology for flow batteries is the vanadium redox battery (VRB), whose development began in the early 1980s. Its peculiar features include a very long life cycle, the possibility of independently designing the required power and energy output, very low self-discharge losses, and moderate efficiency and costs. In a flow battery, two electrolytes are stored in two separate tanks, and an electrical current is created through a redox reaction by circulating H+ ions through a membrane. Storage capacity can be raised by increasing the size of the tanks, at constant power output, while increasing the membrane area has the only effect of expanding the power output (i.e. with constant storage capacity). A significant issue is the limited temperature operational range (10–35°C), which usually requires the installation of a temperature control system, although additions to the electrolytes can increase this range. VRB batteries are at a lower technology readiness level in comparison with other solutions, and there are few and small commercial applications to date. An example of application is a 500 kW/1 MWh VRB installed in a wind power research and testing center in Zhangbei, China (IRENA 2015). Its main objective is to support wind generation by storing excess production and delivering it to the grid in hours with higher demand, and the battery can also provide services over a shorter timeframe, such as load following and voltage support. However, experts warn that significant cost reductions would be required to compete with Li-ion or advanced Pb-A technologies, which in turn would require increasing manufacturing and development funding, which may not be the case without increasing revenues (Fisher et al. 2019).

3 Economics of Li-Ion Batteries

Batteries are still an emerging technology in the framework of power systems management and face high upfront costs and regulatory constraints due to lack of technical know-how in governments and public authorities. The investment costs include the battery pack, balance-of-system (BOS) costs and engineering, and procurement and construction (EPC) costs. Battery pack prices are strongly decreasing, driven by economies of scale related to EVs deployment, and the remaining costs are also expected to decrease sharply, thanks to increased standardization of storage modules and increased competition on the market.

The economics of Li-ion batteries can be quantified by defining a levelized cost of storage (LCOS), in analogy to the well-known definition of the levelized cost of electricity (LCOE), with the aim of accounting for all technical and economic parameters affecting the lifetime cost of discharging stored electricity (Schmidt et al. 2019). This metric has been defined to improve the limitations of considering only the investment cost, which is often the only indicator that is analyzed, by including replacement and disposal costs, maintenance and operation costs, as well as performance parameters such as capacity degradation over time. LCOS is thus defined as the total lifetime cost of the investment in an electricity storage technology, divided by its cumulative delivered electricity (Schmidt et al. 2019); the calculation involves a more in-depth analysis on the expected performance of the unit.

A general formulation of the LCOS is represented in Eq. (14.1), defining the discounted cost per unit of electricity delivered by the batteries, in line with the most recent publications on the subject (Jülch 2016; Lazard 2018; Schmidt et al. 2019). The main aspects included in this formulation are the investment cost, the operation and maintenance cost, the charging cost, and the end-of-life cost, all divided by the sum of the electricity discharged by the storage system over the entire economic lifetime (N), discounted by the discount rate (i).

$$ LCOS\ \left[\frac{\$}{MWh}\right]=\frac{INV_{cost}+{\sum}_{n=1}^N\frac{O\&{M}_{cost}}{{\left(1+i\right)}^n}+{\sum}_{n=1}^N\frac{Charging_{cost}}{{\left(1+i\right)}^n}+\frac{EOL_{cost}}{{\left(1+i\right)}^{N+1}}}{\sum_{n=1}^N\frac{E_{discharged}}{{\left(1+i\right)}^n}} $$

The LCOS is generally defined with respect to the energy discharged, but for specific applications that focus on services related to active power, a more suitable definition would consider the available output power rather than the energy delivered. Some literature works evaluate also an LCOS based on power, by considering the net power capacity that can be provided each year (Schmidt et al. 2019).

The investment cost is usually parameterized on both power output and energy capacity of the battery, and some components need to be replaced in the lifetime of the battery. The replacement costs may be included in the investment cost, properly discounted based on the estimated year of replacement, or they may be considered part of the maintenance costs, without any difference on the final calculation of the LCOS.

Annual costs include O&M costs and charging costs, both affected by the annual number of cycles of the battery. Charging costs are also related to the specific price of electricity, which can show large variations, and the round-trip efficiency, for which a degradation over time should be considered. End-of-life costs are usually calculated as a fraction of investment costs, but the evolution of recycling procedures (and dedicated regulations) may have a significant impact.

The following sections will focus on the main economic aspects involved with investment, operational, and maintenance costs, as well as on the performance parameters that affect the LCOS both on the annual charging cost and on the electricity discharged.

3.1 Investment Cost

The investment cost of Li-ion batteries significantly declined in recent years, and the trend is expected to continue in the future. As already discussed, the most important trend is currently the strong demand of batteries for the EV sector, which is leading to factory capacity expansion in different regions of the world. While this trend is pushing toward a decrease of battery packs cost, Li-ion batteries for stationary storage also include additional components, such as balance of system, power conversion system, energy management system,Footnote 5 engineering, procurement, and construction. Some of these additional components may face similar cost decreases in the future thanks to potential synergies with other industries (e.g. inverter costs decrease thanks to their application in the PV deployment).

Detailed information on the investment cost breakdowns is usually not available, due to confidentiality restrictions. Moreover, due to the high variability of both technologies and battery configurations related to specific applications, it is difficult to draw conclusions related to the weight of each component of investment costs. Material-related costs analyzed in different literature studies range from one-third to almost two-thirds of the total system costs, depending on the source, as illustrated in Fig. 14.3 (IRENA 2017).

Fig. 14.3
figure 3

Investment cost breakdowns from five different sources. (Source: Author’s elaboration on (IRENA 2017))

However, when considering the breakdown of material costs, the figures show less variability: electrode materials (anode, cathode, and electrolyte) constitute roughly half of the cost, with the main contribution related to cathode (between 31% and 39% of the total cost of materials). Notwithstanding the variable impact of materials in the total investment cost of batteries, the increase of the energy density driven by technology innovation will eventually lead to cost savings, thanks to the lower material input required for the same output capacity.

Many authors calculate learning curves based on the historical trend, assuming that the cost decrease has no significant limitations related to external constraints (Berckmans et al. 2017; Kittner et al. 2017; Schmidt et al. 2017). However, other works highlight the fact that the cost of active materials, especially under rising global demand, may act as a strong constraint to further reduce battery costs and may slow down the learning curves (Hsieh et al. 2019).

The rise of battery demand will translate to fast-increasing raw materials requirements, as estimated in the chart of Fig. 14.4 with reference to the expected increase of Li-ion battery production capacity worldwide. In particular, cobalt demand could roughly triple in the period 2018–2028, lithium and graphite demand would grow by 5.5 times, and nickel demand may increase ninefold. Although there has been much debate on the possible lack of materials to support such an expansion, the most critical bottlenecks are expected in the short term, due to the need of adequate planning to upscale the mining industry and the downstream supply chain. Particular issues are related to cobalt, both for the spatial concentration of the resources (more than two-thirds of global cobalt in 2018 have been mined in the Democratic Republic of Congo) and for the fact that is usually obtained as a by-product of nickel and copper mining, making a production upscale more difficult. Moreover, the market concentration of raw materials processing is even more critical, with China representing the largest part of products manufacturing for lithium (51%), refined cobalt (62%), and spherical graphite (100%) worldwide in 2018 (Colbourn 2019). Industry concentration also limits market opportunities, since the lack of diverse perspectives may result in conservative supply expansion plans from existing players (IRENA 2017).

Fig. 14.4
figure 4

Expected increase in raw material demand for global Li-ion batteries manufacturing. (Source: Author’s elaboration on (Benchmark Mineral Intelligence 2018))

The increase of materials demand may be partly compensated by a development of recycling procedures for the depleted batteries, which will need to demonstrate their effectiveness in the coming years, when the first Li-ion batteries used in EVs will start to approach their technical lifetime.

3.2 O&M and Charging Costs

Operational costs of stationary storage are mainly related to electricity cost for charging and maintenance procedures, and the latter may include the replacement costs for components with durability lower than the lifetime of the battery. O&M costs vary depending on the application, but their share on the total LCOS shows limited variation, in the range of 16%–24% for current installations (Lazard 2018). Higher costs are related to wholesale and transmission and distribution (T&D) applications (22%–24%), while utility-scale or behind-the-meter applications coupled with PV usually have a lower impact of O&M costs (16%–19%).

The electricity cost for charging is an important aspect, and its variability is related not only to the application, with large differences in electricity prices for T&D and behind-the-meter systems, but also to the location of these systems, as electricity price has very large variations from country to country. Moreover, all the applications coupled to variable RES generation are usually considering a null cost for charging, although a part of the investment cost of generation plants should somehow be factored in.

Charging costs in LCOS studies generally consider a fixed average price (and in some cases some increment over the lifetime of the system), usually around 50$/MWh for T&D applications and 100$/MWh for behind-the-meter applications (Schmidt et al. 2019). However, while it is difficult to forecast more accurate values on such a long interval, it is important to remind that this cost often shows significant variability over time, and for some applications (e.g. energy arbitrage), it is the main driver of the frequency and duration of battery charge/discharge cycles. Behind-the-meter applications usually face rather constant electricity prices, albeit higher.

3.3 End of Life: Decommissioning and Recycling

The end of life of Li-ion batteries may include different pathways: reusing or repurposing for other applications, recycling of materials, and disposal. While current research is strongly focused on potential recycling procedures, it is estimated that the large majority of batteries is currently disposed, and recycling of Li-ion batteries has not yet emerged as a competitive solution on the market (Pellow et al. 2019). However, other more mature battery technologies, especially lead-acid, have already established recycling pathways, but establishing clear policy targets is a key component in the development of adequate technological solutions.

As discussed for battery manufacturing, also in recycling the larger share of EV batteries will probably drive the market for recycling processes. However, end-of-life conditions of these two applications may broadly differ. Research studies suggest the possibility of reusing EV batteries as stationary storage for residential and industrial applications (Mirzaei Omrani and Jannesari 2019). While a certain level of performance degradation of battery packs may not be acceptable for transport requirements, they could be repurposed for stationary applications thanks to their very low cost. If this option gains interest, the direct material recycling of EV batteries may remain limited, thanks to their extended lifetime through this potential second life.

Few studies currently estimate the potential recycling cost of Li-ion batteries, and the very different assumptions across research works lead to very low comparability of the results. Recovery rates of specific materials are very highly variable, and it is difficult to compare academic studies with the few real applications. There is still a lack of consensus on the sustainability of the end-of-life of Li-ion batteries, both concerning specific energy consumption and environmental impacts (Pellow et al. 2019).

3.4 Performance Parameters

While it is important to focus on the total costs over the life cycle of a battery, its performance is another relevant aspect for the comparison of different solutions, since it directly affects the available electricity that can be supplied by the battery for a given electricity input. Batteries are usually compared based on their energy capacity, although their nominal charge/discharge rate, the maximum depth of discharge (DoD), and their cycle efficiencyFootnote 6 are just as important. An additional aspect is the potential degradation rate of these parameters over time, which can lead to total life-cycle performances lower than the nominal conditions for a new battery. Some aspects are related to the specific technological solution, while others can be adjusted by an accurate choice of design parameters, often based on the specific application that is of interest.

An additional aspect that has an impact on Li-ion batteries performance is the operation temperature, which can affect efficiency, safety, and lifetime. High temperatures accelerate the rate of unwanted chemical reactions that degrade the battery cells, reducing the total lifetime up to 50% for each 10°C of difference with respect to design temperature (IRENA 2017). The longest lifetime is usually achieved in the range 20°C–30°C, resulting in the need of cooling systems in hot climates. On the other hand, operation at extremely low temperatures leads to significant power loss, resulting in significant limitations for electric transport systems in some locations.

Therefore, attention should be paid on the discrepancies for actual operational performance in comparison with expected ratings from manufacturers or testing results, especially considering the different cycling hypotheses and their effect on battery degradation. Multiple circumstances occurring during the operation may lead to degradation of the batteries, including overcharging/discharging, high currents, and mechanical stresses, such as electrode material expansionFootnote 7 (Li et al. 2019).

3.5 Comparison of Different LCOS Studies

As discussed in the previous sections, the hypotheses required to calculate the LCOS are abundant, resulting in a low comparability of different studies. Nonetheless, some information can be retrieved from the most recent literature available on the subject, to represent the range of variability of LCOS results related to Li-ion applications. Figure 14.5 reports the average values of LCOS for Li-ion batteries calculated in different studies (Comello and Reichelstein 2019; Jülch 2016; Lazard 2018; Schmidt et al. 2019).

Fig. 14.5
figure 5

Comparison of average values of LCOS for Li-ion batteries from selected studies. (Source: Author’s elaboration)

While many analyses exist for current LCOS, few studies extend the analysis to the future evolution of LCOS values for storage. Since Schmidt et al. (2019) is the only study providing detailed projections of future trends, as well as a differentiation of results for large batteries and behind-the-meter batteries, it has been given more relevance in the chart. It is important to underline that these numbers are strongly affected by the high uncertainty associated both to investment costs of the technology and to the market conditions for the electricity supply. For this reason, the values reported in Fig. 14.5 should only be considered as a potential future indication based on the most recent available literature, but since the commercial deployment of Li-ion batteries for stationary storage is only beginning, these numbers may be subject to significant revisions in the years to come. Additionally, just like any other comparison of literature results, it is important to highlight the caveat that the calculation of the LCOS requires multiple assumptions, which may differ across multiple research works.

3.6 External Context and Revenue Opportunities

While much attention is generally paid to energy storage costs, since this aspect is often the more limiting factor, a brief analysis of the potential revenue opportunities can provide additional insights on the economics of Li-ion batteries. The opportunities for any storage technology are related to the variable value that a commodity can have over time, and electricity storage is thus most required when there is a larger mismatch between the electricity demand and supply.

Such mismatch was generally tackled through bids and offers in capacity markets at the transmission level, whose participation was usually limited to dispatchable power plants such as thermoelectric and pumped hydro storage. However, such mechanisms may not be enough in the current transition toward higher shares of RES and distributed generation. In this transition phase, virtual power plants are being deployed: a virtual aggregation of several small units of different nature (i.e. electricity producers, storage units, demand-side management) thanks to the use of digital technologies. In the context of this transition, energy storage can fit at different levels thanks to the possible scalability of system size and the flexibility of operation.

The involvement of different stakeholders may be tightly related to the specific policies and regulations that will be implemented, but the flexibility requirements of a low-carbon energy system will necessarily include storage among other different solutions. While transmission and distribution systems operators are evaluating batteries for a wide range of network services, they are also being considered by large-scale variable RES producers to increase their capacity of dispatching electricity and the associated value and market opportunities (IRENA 2019a). A similar driver exists at residential level, where households equipped with a PV system try to maximize their self-consumption if they face high electricity prices (IRENA 2019b). Final users are generally more affected by stringent regulations, and the profitability of behind-the-meter storage may exhibit strong differences across countries.

Current opportunities are emerging with an uneven distribution at global scale, since countries with favorable regulations are already seeing deployment of battery storage systems at different levels, including Germany, Australia, South Korea, and the United States (IRENA 2019a).

3.7 Future Deployment of Stationary Li-Ion Batteries

In parallel to the economic analysis that has been presented before, it is important to discuss the expected scenarios for stationary battery deployments. While these numbers are continuously being updated based on the evolution of the energy systems and energy markets, the comparison of the current scenarios of different international organizations underlines the strong momentum and the high potential of stationary storage.

Figure 14.6 reports a comparison of the future trends expected by some of the most influential energy organizations, that is, the International Renewable Energy Agency (IRENA 2017), the International Energy Agency (IEA 2019), and Bloomberg New Energy Finance (BNEF 2019). These scenarios differ for the final capacity deployed, and it is not always clear which kind of applications are included in the forecast, in particular as far as behind-the-meter applications are concerned. Nevertheless, in all the cases the expected battery storage capacity reaches a considerable total volume, although stationary storage will likely remain a minor market in comparison with Li-ion batteries used in electric vehicles.

Fig. 14.6
figure 6

Comparison of future scenarios for stationary batteries deployment. (Source: Author’s elaboration)

To give some context to these volumes, the current energy storage capacity of pumped hydro storage, as of 2017, sum up to 4.5 TWh worldwide (IRENA 2017). However, it is important to highlight that batteries and pumped storage are not in competition but rather provide complementary services, since electric batteries will mostly be used for short-term services (up to some hours or in some cases few days), while pumped hydro storage is characterized by longer charging and discharging times (in the order of weeks or months).

4 Conclusions

This chapter described the main aspects of the economics of battery storage systems and provided a qualitative discussion of battery technology and potential. Due to the high momentum of Li-ion batteries, especially in connection to the expected strong manufacturing capacity increase for electric vehicles applications, updated figures may exhibit strong variations from a year to another. On the other hand, underlying trends related to the main cost drivers and revenue opportunities will likely show lower variations and maintain their importance.

Li-ion batteries for stationary storage have recorded massive upfront cost decrease in the last years, and this trend is expected to continue in the coming decade. The reason is the expected increase of batteries supply capacity at a global level, driven by rising demand of electric vehicles, which is benefitting from economies of scale as well as technological improvements related to both battery performance and manufacturing efficiency. A secondary effect of large deployment of EVs may be the availability of cheap second-life batteries, whose remaining performance level not suitable for transport could be acceptable for stationary storage requirements.

While much emphasis is usually put on Li-ion batteries investment costs, there are other factors that affect the real total cost of batteries operation, which can be evaluated through the levelized cost of storage (LCOS). Such factors include not only the O&M costs, the electricity costs for charging, and the end-of life costs, but also a number of technical parameters that affect the performance of the battery and thus the electricity output that can be achieved. They include the energy/power ratio (usually resulting from design choices), the round-trip efficiency, the calendar and cycling lifetimes, the degradation over time, as well as the operation logic of the battery in terms of number of cycles per year and the average discharge duration. An understanding of these parameters is essential to have a complete picture on the economics of Li-ion batteries operation for electricity storage, since the results of available research studies are strongly dependent on underlying hypotheses.

Current applications of Li-ion batteries for stationary storage, both as utility-scale and behind-the-meter systems, demonstrate the crucial importance of policies and regulations in fostering the adoption of such technologies and improving their maturity. While upfront costs remain the main barrier to widespread adoption, existing regulations are often limiting the development and deployment of batteries for different applications and network services. Moreover, key stakeholders are not always aware of the potential of this technology and the results from existing case studies.

If the current trend of declining costs will continue in the future, without being hindered by issues related to the lack of raw materials or bottlenecks in the supply chain, Li-ion batteries are expected to play a crucial role in providing the required flexibility for low-carbon electricity systems. A crucial aspect will be the competition with the EVs market, since its expansion may lead to either positive or negative impacts on stationary storage applications.