Eco-Efficiency Assessment of the Application of Large-Scale Rechargeable Batteries in a Coal-Fired Power Plant

Abstract

This article presents the results of an eco-efficiency assessment of the application of large-scale rechargeable battery technology in electricity generation from coal. The eco-efficiency of electricity production in a 350 MW coal-fired power plant was calculated. Two production variants were compared: with the use of a lithium-ion battery of a 400 MWh capacity to optimize the operation of power blocks and without using the battery. Hard coal is one of the main fossil fuels used to generate electricity in Poland. Despite the growing share of electricity from renewable sources, this situation will persist for many more years. The main reasons for this are the high costs and long-lasting process of moving away from fossil fuels in the energy sector. Therefore, any technical solutions that can temporarily reduce the negative impact of coal-based power engineering on the environment should be considered. At the same time, the economic aspects of such solutions must be taken into account. That is why the eco-efficiency assessment method was chosen, which integrates economic and environmental aspects. The obtained results of the analyses indicate the occurrence of environmental and economic benefits resulting from the use of the battery in coal-fired power plants. It has been found that battery-based technology is more eco-efficient than technology without such a battery. A sensitivity analysis was carried out, which allowed the impact of individual computational variables on the eco-efficiency assessment result to be assessed. The results indicate that fuel prices (coal and heavy fuel oil—mazout) and CO₂ emission allowances have the greatest impact on the eco-efficiency of the analyzed technology. It was also found that the factors related to the battery, such as its efficiency, life span, decrease of the capacity after 10 years of operation, and construction cost, have a much smaller impact on the results.

1. Introduction

The growing share of renewable energy sources in the structure of electricity production and the threat of interruptions in its supply contribute to the growing interest in energy storage. Until now, technologies such as pumped storage power plants have been used on a large scale of 100 MW. In contrast, the application of lithium-ion batteries (LiB), despite the growing number of projects, has been limited to a small scale, ranging from kW to several MW [1,2]. However, currently implemented and announced projects show upcoming changes in the energy storage market. In 2017, the largest energy storage to date using 100 MW/129 MWh LiB was constructed at the Hornsdale Wind Farm in Australia by Tesla, Inc., and Neoen at a cost of US $ 50 million. According to press announcements [3,4,5,6,7], further large-scale lithium-ion energy storage facilities are to be built at the Moss Landing Gas Power Plant in California, the USA, for Pacific Gas and Electric Company (PG&E). Plants with a capacity of 183 MW/730 MWh (Tesla), 300 MW with a capacity of 1.2GWh (Vistra/Dynegy), 75 MW/300 MWh (Hummingbird Energy Storage LLC), and 10 MW/40 MWh (Micronoc Inc.) are planned. The AES Corporation has begun work on creating two energy storage facilities in Southern California: with a capacity of 400 MWh in Long Beach for the Alamitos Energy Center (AEC) and Strata Solar with a size of 100 MW/400 MWh in Oxnard. The largest energy storage in the world, the Manatee Energy Storage Center, with a size of 409 MW/900 MWh, belonging to NextEra, is to be built in Manatee County, Florida, the USA, in 2021. This storage will be powered by solar energy, and the Florida Power and Light’s plans are to install 30 million solar panels in this location to 2030. The center will replace two existing 1650 MW gas-fired power plants. The technology of energy storage in LiB is a development technology implemented on an increasing scale. In the literature, however, the environmental assessment of LiB is primarily limited to applications in electric cars [8,9,10]. In [11], a life cycle assessment (LCA) of the application of various electricity storage systems in the power grid was made: the Compressed Air Energy Storage (CAES), the Pumped Hydro, the PEMFC—proton exchange membrane fuel cell, the Sodium Sulfur batteries (NAS), and the battery systems (lithium-ion, lead-acid, and NaNiCl). For LiB, the authors of this paper based their work on the 2006 publication and assumed a relatively short life span of 5–10 years. A comparison of different energy storage systems using LCA is presented in the work [12]. The authors of this publication identified LiB as the most suitable for power grid applications. A review of the literature on LiB is presented in the paper [13], and the work [14] provides a review of the literature related to the application of LiB in the power grid. The authors of the work [14] emphasized the significant impact of the battery use phase on the overall assessment. In [15], the effect of using different batteries in the energy system to minimize costs and CO₂ emissions was analyzed. The results showed the cost-effectiveness of the system for the LiB and Vanadium Redox Flow Battery (VRFB).

The power industry in Poland is currently based on fossil fuels, mainly coal. This is associated with significant CO₂ emissions being released into the atmosphere. Currently, research is being carried out in Poland to reduce these emissions by injecting CO₂ into geological formations. However, this research is at an early stage of development. In addition, the possibilities of using this method of reducing CO₂ emissions in the energy sector are limited by geological considerations [16,17].

At the same time, the energy sector based on renewable energy sources is developing—to the greatest extent, wind power. A negative feature of wind power is the instability of the electricity supply and the need to provide power reserves in the national energy system in the form of electricity sources based on fossil fuels. The introduction of electricity from renewable energy sources into the power grid, combined with changes in demand from consumers, means that conventional power plants operate with a high volatility—both annually and daily. This results in a decrease in the efficiency of electricity production. One way to stabilize the operation of a conventional power plant in such conditions is to apply a large-capacity battery. Such an accumulator could be used to introduce additional amounts of electricity into the power grid during hours of increasing demand and to store surpluses of generated electricity during times of falling demand.

This article presents the results of an eco-efficiency analysis on the cooperation of power units of a professional coal-fired power plant with a large-scale rechargeable battery. It was assumed that the energy storage has a system function in the power plant, which results in balancing the energy produced and the energy received. The direct benefit of using such energy storage may be optimization of the power unit operation, and thus a reduction of the unit costs of electricity production and decrease of the specific emissions of pollutants associated with its production. In addition, the optimization and stabilization of power unit work will result in a longer life expectancy and lower expenses for ongoing repairs and maintenance. Due to the availability of data, the article is limited to estimating the environmental and financial benefits that result from reducing the amount of fuel consumed (coal and mazout) and gas emissions released into the atmosphere, in particular CO₂, as well as the materials, media, and waste released in the field of environmental analysis.

2. Materials and Methods

2.1. Materials

In order to assess the eco-efficiency of electricity storage technology in a large-scale battery, actual operating data for 2018 were used. The data were obtained from a 350 MW coal-fired power plant operating in Poland. The analyzed power plant has six dust boilers with a power input in the fuel of 560 MW for each boiler. Steam from the boilers feeds six condensing turbines in a block system, with rated powers of 225 MW (five turbines) and 220 MW (one turbine). The turbine rotors are connected by means of coupling with the GTHW-type power generators. The turbines cooperate with three cooling towers.

For the analyzes, the following data provided by the power plant operator were used:

• Performance characteristics of the six exploited power units;
• Working time and volume of electricity production in an hourly cycle in 2018;
• Fuel consumption by individual power units in 2018;
• The calorific value of coal and auxiliary fuels used in the power plant in the analyzed period;
• Balance of inputs and outputs related to the production of electricity, including raw materials and fuels used, products manufactured (electricity and gypsum), gases emitted, and solid and liquid waste generated.

Table 1. Summary of the input and output data values per 1 MWh of the net generated electricity—average values from 2018.

Item	Unit	Value
INPUTS
Coal	kg/kWh	525.4
Other fuels (mazout)	kg/kWh	2.1
Limestone	kg/kWh	13.6
Water from the water supply network	m3/kWh	0.2
Water from own surface water intake	m3/kWh	3.2
OUTPUTS
Net electricity	GWh/year	4423.3
Carbon dioxide	kg/kWh	1047.4
Sulfur dioxide discharged into the atmosphere after flue gas cleaning processes	kg/kWh	0.5
Nitrogen oxides discharged into the atmosphere after flue gas cleaning processes	kg/kWh	0.8
Dust from the combustion of fuels discharged into the atmosphere after flue gas cleaning processes	kg/kWh	0.0
Ashes	kg/kWh	40.8
Industrial wastewater treated	m3/kWh	1.1
Gypsum	kg/kWh	24.6
Fly ashes from coal	kg/kWh	40.8
Slags, bottom ash, and boiler dust	kg/kWh	22.3
Sludges from on-site effluent treatment	kg/kWh	1.8
Sludges from water decarbonization	kg/kWh	0.9

summarizes the values of input and output data per 1 MWh of net electricity generated. These are the average values from 2018.

In order to estimate the benefits resulting from optimization of the power unit operation of the analyzed coal-fired power plant, the operational characteristics were used. These characteristics show the dependence between the gross fuel (coal) consumption per unit of electricity produced and the power unit load. This relationship for the individual power units is presented in Figure 1. It results from the dependency that the optimal operating point of the power units is near the nominal load of 220 MW.

In real power plant operating conditions, the load of individual power units is characterized by a very high variability. Very rarely, power units work at an optimal load. Due to the changing demand for electricity from consumers during the day, there are significant fluctuations in the hourly amounts of electricity produced. In 2018, the maximum hourly electricity production values reached 106% of the daily average value. The consequence was an uneven operation of the individual power units, as well as periodical stops of some of them.

The use of a large-scale battery will have the additional benefit of reducing the number of power unit stops. As a consequence, the number of restarts and the fuel consumption for these start-ups (coal and mazout) will decrease. In 2018, the total number of start-ups of stopped power units was over 300. The additional fuel consumption for these start-ups was about 9.1 thousand tons of mazout and about 32.7 thousand tons of coal.

For the purposes of life cycle costing (LCC), a unit cost of battery construction equal to USD 200/kWh based on literature data [18], and the average fuel costs (coal and mazout) made available by the power plant operator for 2019, were adopted.

2.2. Methods

A comparison of two variants of electricity generation in a 350 MW coal-fired power plant was made: without the electricity storage and with the electricity storage in the form of a lithium-ion battery. In all analyzes, a functional unit equal to 1 MWh of the electricity transferred to the power grid was adopted (the net electricity). The battery installation life cycle was assumed to be 10 years and the constant efficiency was assumed to be 86%. The decrease in battery capacity and efficiency during its operation was not modeled. The impact of the battery life, efficiency, and decrease in capacity on the results of calculations was examined in the sensitivity analysis.

2.2.1. Environmental Impact

The use of energy storage to optimize the operation of a coal-fired power plant results in an increase in the efficiency of electricity generation and, consequently, a reduction in the use of fuel (hard coal). The effect is a reduction in exhaust emissions released into the atmosphere, a reduction in the amount of waste produced, less water consumption, and a smaller amount of wastewater, which results in a reduction of the negative impact on the environment in the electricity generation process at the power plant. On the other hand, the production of LiB for energy storage causes an additional burden on the environment. Therefore, reducing the environmental impact at a power plant comes at the expense of additional negative environmental impacts elsewhere in the supply chain. Life cycle assessment is a method that allows this phenomenon to be taken into account and prevents pollution from moving outside the company gates by implementing only seemingly pro-environmental measures. Therefore, this method was used to assess the potential environmental benefits of using energy storage at a coal-fired power plant, taking into account the environmental costs associated with battery manufacture.

The assessment was carried out using the ReCiPe 2016 method [19]. It is a method that allows the conversion of environmental interventions in the form of emissions, the use of mineral resources, and fossil fuels into potential environmental impacts in various impact categories (midpoint) or in the form of damage to human health, ecosystem quality, and the availability of non-renewable resources (endpoint). For this reason, the ReCiPe method uses the conversion factors, which, on the one hand, take into account complex environmental mechanisms, but on the other, assume the necessary averaging and simplification. For calculations, the Hierarchist perspective (the default perspective assuming a compromise between the long- and short-term negative effects on the environment) was used. The ReCiPe method allows the amount of potential environmental effects of processes and products in individual impact categories to be estimated, as well as the estimated total damage in the form of one number to be calculated, which in turn is used to calculate the eco-efficiency. Compared with the previous version of the ReCiPe 2008 method [20], the new version of ReCiPe 2016 takes into account the depletion of lithium resources, which is important for estimating the environmental impact of lithium-ion battery production.

Table 2. Impact categories included in ReCiPe 2016 [19].

Impact Category	Unit
Global warming	kg CO2 eq
Stratospheric ozone depletion	kg CFC11 eq
Ionizing radiation	kBq Co-60 eq
Ozone formation, human health	kg NOx eq
Fine particulate matter formation	kg PM2.5 eq
Ozone formation, terrestrial ecosystems	kg NOx eq
Terrestrial acidification	kg SO2 eq
Freshwater eutrophication	kg P eq
Marine eutrophication	kg N eq
Terrestrial ecotoxicity	kg 1,4-DCB *
Freshwater ecotoxicity	kg 1,4-DCB *
Marine ecotoxicity	kg 1,4-DCB *
Human carcinogenic toxicity	kg 1,4-DCB *
Human non-carcinogenic toxicity	kg 1,4-DCB *
Land use	m2a crop eq
Mineral resource scarcity	kg Cu eq
Fossil resource scarcity	kg oil eq
Water consumption	m3

* 1,4-dichlorobenzene.

lists the impact categories included in the ReCiPe method.

The results of the endpoint analysis, after normalization, weighting, and aggregation into one single score, are expressed in point units (Pt). A higher value is related to a higher negative potential impact of the process on the environment.

The data for the electricity production process at the power plant were obtained directly from the operator of the analyzed coal-fired power plant. The secondary data were used for other processes. Namely, the data for the upstream and downstream processes were taken from the ecoinvent database: coal mining in Poland, road and rail transport of coal over a distance of 100 km, limestone production, water decarbonization, wastewater treatment, management of sludge from decarbonization and wastewater treatment processes, and management of waste from burning.

The process of energy storage was assumed to occur in a battery consisting of lithium-ion cells with a total capacity of 400 MWh. Since there is a lack of data on large-scale installations, the cell production data were based on literature for smaller cells, using the aspect ratios. This approximation is justified for technological reasons because high-capacity batteries are constructed of smaller unit cells. For example, the large-scale energy storage planned by Tesla at the Moss Landing Power Plant with a capacity of 183 MW/730 MWh will be assembled from several hundred Tesla Megapack lithium-ion batteries, with a capacity of 3 MWh each. Despite this, due to the large change of scale in our analyzes, one should take into account the error associated with scaling the production process, primarily related to the lack of a linear proportion for the housing, temperature control system, battery management system, and electrical system.

For energy storage, this analysis adopts the Li-ion battery production process and the assumptions described in [21] for batteries weighing 253 kg and having a capacity of 26.6 kWh, due to the high quality and reliability of the source data. The application of the NCM (lithium-nickel-cobalt-manganese-oxide) cell was adopted, in which the use of an aluminum collector covered with lithium-nickel-cobalt-manganese oxide with stoichiometry Li(Ni_1/3Co_1/3Mn_1/3)O₂ as the cathode and the anode in the form of a copper collector covered with graphite was assumed. The scale conversion was conducted on the basis of the cell capacity in kWh. The system boundaries used in the analysis are presented in Figure 2. The analysis considered the phase of production and the use of the battery.

The system boundary of the process of battery production can be found in the source material [21].

2.2.2. Eco-Efficiency

The eco-efficiency was defined, for the first time, by the World Business Council for Sustainable Development (WBCSD) as providing products and services at a competitive price that meet human needs and improve their quality of life, reducing the environmental impact and resource consumption throughout the entire life cycle. Eco-efficiency assessment is a method of assessing technology, products, and process efficiency. It allows for an integrated economic and environmental assessment with the life cycle of a technology or product taken into account. Eco-efficiency assessment aims to select a product or technology with the lowest environmental impact and the lowest cost over the full life cycle of the product or technology. The basic goals of increasing eco-efficiency are as follows: to obtain the highest possible added value with the least impact on the environment, and to increase the economic efficiency whilst reducing the impact on the environment. Eco-efficiency links the main goals of the enterprise, which are the profit and profitability of production, with the environmental approach, thanks to which people who are decision makers in the enterprise have the opportunity to create innovative products and technologies which also meet environmental criteria [22,23].

The main goal of eco-efficiency assessment is to conduct a comparative analysis of various solutions integrating economic and environmental aspects. The applicable terminology and methodological framework for the eco-efficiency assessment are included in the ISO 14045: 2012 standard “Environmental management—Eco-efficiency assessment of product systems—Principles, requirements and guidelines” [24]. According to this standard, the environmental performance rating for the eco-efficiency should be obtained based on the life cycle assessment (LCA) technique. This method allows an assessment of the material and energy consumption, as well as the environmental impacts throughout the entire life cycle. Based on the LCA, it is possible to assess the impact of technology on the specific categories of damage, such as human health, ecosystem quality, and resource depletion. One can also specify which elements and stages of a process generate the greatest environmental load. According to the above-mentioned standard, the life cycle costing (LCC) method was also used to assess the eco-efficiency. The life cycle costs were assessed using the dynamic unit cost index (DGC-Dynamic Generation Cost) [25,26]. The DGC is equal to the price that allows the achievement of discounted revenues equal to discounted costs. The DGC shows the technical cost of obtaining a product unit [27].

In order to refer the results of calculations covering the entire life cycle to the adopted functional unit, in the first stage of the eco-efficiency analysis, Equations (1) and (2) were used:

$$LCA=\frac{{\displaystyle \sum _{t=0}^n\frac{LC{A}_t}{{(1+i)}^t}}}{{\displaystyle \sum _{t=0}^n\frac{P_t}{{(1+i)}^t}}},$$

(1)

where

LCA_t—results of the LCA analysis carried out using the ReCiPe method in a given year of construction, operation, or decommissioning of the installation;

P_t—electricity production volume in a given year;

i—discount rate;

t—year, which takes values from 0 to n, where 0 is the year in which the first environmental impacts arise and in which the first costs are born (the first year of construction), while n is the year of decommissioning of the installation.

$$DGC=\frac{{\displaystyle \sum _{t=0}^n\frac{LC{C}_t}{{(1+i)}^t}}}{{\displaystyle \sum _{t=0}^n\frac{P_t}{{(1+i)}^t}}},$$

(2)

where

LCC_t—results of the LCC analysis for a given year of construction, operation, or decommissioning of the battery installation; other explanations as in Equation (1).

The LCC analysis was carried out at constant prices. The analysis period covers the years 2020–2030, i.e., one year of investment implementation and 10 years of battery exploitation. In order to compare both variants of the electricity production (with and without a battery) using the DGC indicators, the following cost items were included in their calculation:

• Investment outlays for building a battery (in the variant with a battery);
• Fuel expenditure (coal);
• Fuel expenditure (mazout);
• Expenses for the purchase of CO₂ emission allowances.

Other costs by type occurring in the electricity production are the same or very similar for both analyzed variants. Therefore, they do not affect the results of the comparative analysis. In addition, due to their confidential nature, they were not made available by the operator of the analyzed coal-fired power plant. Therefore, they were omitted in the calculations. The costs associated with the disposal of the battery after the end of its exploitation were adopted as equal to 0. It was assumed that they will be financed from the sale of materials recovered in the process of demolition and utilization. The changes in prices of fuel and CO₂ emission allowances were forecast in the period covered by the analysis on the basis of available reports in this field [28,29].

In accordance with the definition contained in the ISO 14045: 2012 standard, eco-efficiency is a measure that refers to the results of the environmental assessment of a product system in which the value of the system is analyzed. The value of the product system was defined as the result of the economic assessment related to the functional unit. Since the results of both environmental and economic assessment express the negative aspects of the technology analyzed (the higher the value of indicators, the lower the eco-efficiency of technology), it was assumed that the value of the eco-efficiency indicator is inversely proportional to the value of the environmental and economic assessment indicators. Based on such a definition and assumptions, and on the basis of Equations (1) and (2), the following formula was adopted for calculating the eco-efficiency (Equation (3)):

$$E{E}_K=\frac{{\left({\displaystyle \sum _{t=0}^n\frac{P_t}{{(1+i)}^t}}\right)}^2}{{\displaystyle \sum _{t=0}^n\frac{LC{A}_t}{{(1+i)}^t}}\cdot {\displaystyle \sum _{t=0}^n\frac{LC{C}_t}{{(1+i)}^t}}},$$

(3)

where the explanations are the same as in Equations (1) and (2).

Equation (3) integrates the results of two analyses: the LCA performed by the ReCiPe method and the LCC expressed by the DGC indicator. From the eco-efficiency assessment calculated using Equation (3), the following relationship was obtained: the lower the value of the eco-efficiency index is, the less eco-efficient the technology is.

2.2.3. Sensitivity Analysis

The obtained results of the eco-efficiency analysis were subjected to a sensitivity analysis. The purpose of a sensitivity analysis is to determine the impact of changes in selected input variables on the level of the economic efficiency indicators of the technology analyzed; in this case, a lithium-ion battery was applied to optimize the operation of power units at a coal-fired power plant. In the first step, the expected values of these indicators are calculated, which are the most realistic in the given conditions of investment uncertainty. Then, changes are made to the values of successively selected variables, and the strength and direction of the impact of these variables on the level of efficiency are examined. Each of the input variables may be changed by a certain number of percentage points above or below the expected value while maintaining the other conditions so that they are unchanged. In addition, for each of these changed values, compared to the baseline scenario, a new value of the economic efficiency indicator is calculated. The scope of the analysis will be limited to the variables having the greatest impact on the result, i.e., the value of the economic efficiency indicator. They are the so-called critical variables [30].

The following critical variables identified for the analyzed technology of the application of a large-scale lithium-ion battery were analyzed:

• Changes in the capital expenditures (battery construction cost): deviations of ±10%, ±30%, and ±50%;
• Changes in fuel prices (coal and mazout): deviations of ±10%, ±30%, and ±50%;
• Changes in prices of CO₂ emission allowances: deviations of ±10%, ±30%, and ±50%;
• Changes in the battery efficiency: deviations of ±6.98%, ±4.65%, and ±2.33%, which correspond to a battery efficiency ranging from 80% to 92%;
• Changes in the battery life cycle: deviations of −20%, + 20%, + 40%, + 60%, + 80%, and + 100%, which correspond to a lifetime of 8 to 20 years;
• Changes in the battery capacity: decrease of the capacity by 0%, 10%, 20%, and 30% after 10 years of operation, which correspond to a battery capacity of 400 to 280 MWh.

3. Results and Discussion

3.1. Environmental Impact

Table 3. Calculated environmental impact per 1 MWh of electricity (ReCiPe2016 Midpoint, H).

Impact Category	Unit	Coal-Fired Power Plant without Battery	Coal Power-Fired Plant with Battery
Global warming	kg CO2 eq	1249.738	1226.611
Stratospheric ozone depletion	kg CFC11 eq	0.000	0.000
Ionizing radiation	kBq Co-60 eq	18.787	19.076
Ozone formation, human health	kg NOx eq	1.027	1.009
Fine particulate matter formation	kg PM2.5 eq	0.170	0.173
Ozone formation, Terrestrial ecosystems	kg NOx eq	1.030	1.012
Terrestrial acidification	kg SO2 eq	0.524	0.531
Freshwater eutrophication	kg P eq	0.266	0.264
Marine eutrophication	kg N eq	0.016	0.017
Terrestrial ecotoxicity	kg 1,4-DCB *	67.235	127.897
Freshwater ecotoxicity	kg 1,4-DCB *	11.982	12.197
Marine ecotoxicity	kg 1,4-DCB *	16.609	16.933
Human carcinogenic toxicity	kg 1,4-DCB *	29.017	28.646
Human non-carcinogenic toxicity	kg 1,4-DCB *	292.023	302.644
Land use	m2a crop eq	17.691	17.371
Mineral resource scarcity	kg Cu eq	0.205	0.289
Fossil resource scarcity	kg oil eq	314.870	308.101
Water consumption	m3	225.827	233.267

* 1,4-dichlorobenzene.

presents the environmental impact values of the individual impact categories calculated by the life cycle assessment (LCA) method.

A comparison of the environmental damage associated with 1 MWh of electricity transferred to the grid without using the energy storage and after applying it is presented in

Table 4. Calculated damages to the environment per 1 MWh of electricity (ReCiPe 2016 Endpoint).

Damage Category	Unit	Coal-Fired Power Plant without Battery	Coal-Fired Power Plant with Battery
ReCiPe2016 H/A	Pt	36.729	36.702
incl.:
Human health	Pt	32.539	32.501
Ecosystems	Pt	3.986	4.004
Resources	Pt	0.204	0.197
ReCiPe 2016 E/A		208.902	210.506
ReCiPe 2016 I/A		30.853	31.596

.

The use of energy storage does not significantly affect the environmental damage resulting from the electricity production process in a coal-fired power plant (Table 4). The analysis of results in the specific impact categories (Table 3) showed that the use of energy storage contributes to a two-fold increase in impact in the terrestrial ecotoxicity category and a significant increase in the mineral resource scarcity category. It is associated with the use of metals for the production of lithium-ion cells.

The negative environmental impact in the Global Warming category is associated with more than 84% CO₂ emissions in the exhaust gases; with 14% methane emissions during coal mining; and to a lesser extent, with the use of electricity in the mine and coal transport. Therefore, the reduction of CO₂ emissions and consumption of coal per 1 MWh of electricity, obtained through the use of energy storage, reduces the impact in this category.

A similar effect is expected in the fine particulate matter formation category because, as a result of energy storage, the dust emissions from power plants have also decreased. However, contrary to expectations, an increase in the environmental impact was recorded in this category. This is related to the battery life cycle and dust emissions during nickel production and the extraction of nickel-containing ore for the production of Li-ion batteries.

Cell production mostly takes place in South Korea, China, and Japan. Therefore, the life cycle assessment took into account their transport by sea, hence the increase in the marine ecotoxicity and marine eutrophication impact categories.

The increase of the negative environmental impact in the terrestrial ecotoxicity category is connected to the life cycle of the LiB. It mainly refers to copper and other metal emissions being released into the air and water during the production process of copper, which is used as raw material for the production of anode collectors applied in battery cells. In the mineral resource scarcity category, the highest share is associated with the use of iron and ferronickel ores for the production of steel, which is used, for example, in wagons and railroad tracks necessary for coal transport (40% impact) and in coal mines (17%). It also refers to the chromium steel used in the wastewater treatment process. Due to the fact that cell production is an energy-consuming process, the assumed country of production and the adopted national energy mix have a significant impact on the LCA results. Therefore, the level of impact in the category of the mineral resource scarcity is also influenced by the consumption of uranium ore associated with the participation of nuclear energy in the energy mix. The share of metal production for battery manufacturing in this category can be estimated at 31%, including 12% cobalt, 5% nickel, and 7% copper. The level of impact in this category is influenced by the adopted degrees of recovery and recycling of metals, including steel, copper, aluminum, manganese, and nickel, as well as cobalt, for which supplies are considered to be at risk. The use of recycled metals reduces the impact associated with the use of non-renewable raw materials and the penetration of metals such as cobalt, copper, nickel, and lead from the landfill to the environment and their ecotoxicity. While modeling the end of life, recycling can also affect the results of the analysis. The recycling levels established in Directive 2006/66/EC [31] assumed the collection of 45% by weight of the used batteries and the minimum recovery of 50% by weight of the battery, without energy recovery. Therefore, the recovery of metal from the battery is small and is primarily limited to iron and copper, and sometimes also cobalt and nickel due to their price [32]. The European Union has now established a new class of hazardous waste for LiB and new regulations related to recycling [33], which should improve the situation in the future. The Canadian and USA regulations still allow battery storage in landfills.

By carrying out further stages of life cycle assessment, i.e., normalization of the results obtained, their conversion to the category of damages and weighing allowed an aggregated result to be obtained (Table 4). It indicates that the negative effects associated with the increase in ecotoxicity and the use of mineral resources are compensated for by the reduction of greenhouse gas emissions and the use of fossil fuels. These stages assume simplifications and use expert indicators. Therefore, the value of damages is approximate and burdened with a high uncertainty. For comparison, the results obtained using the individualist perspective (only short-term impacts, future problems will be solved by technological development) and the egalitarian perspective (all impacts, including long-term, precautionary thinking) are presented, and the results show no significant differences between the analyzed variants.

3.2. Eco-Efficiency

Table 5. The results of the eco-efficiency analysis.

Technology	DGC, PLN/MWh	LCA ReCiPe, Pt/MWh	Eco-Efficiency, MWh2/PLN × Pt
Production of electricity in coal-fired power plant with battery	296.36	36.702	0.000092
Reference technology: production of electricity in coal-fired power plant without battery	297.87	36.729	0.000091

presents the results of the eco-efficiency analysis, along with the results of the LCA and LCC analyses that were used to calculate the eco-efficiency indicators.

The calculation results indicate that the production of electricity in a coal-fired power plant with the use of the battery is more eco-efficient than without using the battery. However, the difference is insignificant. This is illustrated by the graphic interpretation of the results presented in Figure 3.

The higher eco-efficiency of the electricity production in a coal-fired power plant using a battery is proven by the better results of the LCA and LCC analyzes, expressed as the DGC indicator. The results of the LCA analysis were discussed in the previous section. On the other hand, the positive result of the LCC analysis for electricity production technology using the battery was achieved by savings in the fuel expenditure and in the purchase of CO₂ emission allowances. They compensate for the significant capital expenditure required for building a battery. The use of a battery allows for the reduction of fuel consumption in two situations:

• For start-ups of the stopped power units,
• During normal operation of the power units.

The use of a battery will contribute to a reduction in the number of start-ups of the stopped power units due to sudden increases in the demand for electricity. Daily peaks in the demand for electricity will be quickly supplemented with the electricity stored in the battery, while during a reduced demand for electricity, the surpluses of its production will be discharged to the battery. This stabilizing power plant function may be used to optimize the operation of individual power units. It will be able to work most of the time in the optimal load range when the efficiency of electricity generation is the highest. Consequently, fuel consumption and CO₂ emissions per unit of produced electricity will decrease.

3.3. Sensitivity Analysis

Table 6. The results of the sensitivity analysis of electricity production technology in a coal-fired power plant using a battery.

Unit	Value
Deviations	LCA ReCiPe, Pt/MWh	DGC, zł/MWh	Eco-Efficiency, MWh2/PLN × Pt
Changes of Capital Expenditures—Battery Construction Cost
−50%	36.702	292.31	0.0000932
−30%	36.702	293.93	0.0000927
−10%	36.702	295.55	0.0000922
0%	36.702	296.36	0.0000919
10%	36.702	297.17	0.0000917
30%	36.702	298.78	0.0000912
50%	36.702	300.40	0.0000907
Changes of Fuel Prices (Coal and Mazout)
−50%	36.702	212.45	0.0001282
−30%	36.702	246.01	0.0001108
−10%	36.702	279.58	0.0000975
0%	36.702	296.36	0.0000919
10%	36.702	313.14	0.0000870
30%	36.702	346.70	0.0000786
50%	36.702	380.26	0.0000717
Changes of Prices of CO2 Emission Allowances
−50%	36.702	236.13	0.0001154
−30%	36.702	260.22	0.0001047
−10%	36.702	284.31	0.0000958
0%	36.702	296.36	0.0000919
10%	36.702	308.40	0.0000883
30%	36.702	332.49	0.0000819
50%	36.702	356.59	0.0000764
Changes of Battery Efficiency
−6.98%	36.937	298.01	0.0000908
−4.65%	36.855	297.43	0.0000912
−2.33%	36.777	297.45	0.0000914
0.00%	36.702	296.36	0.0000919
2.33%	36.631	295.86	0.0000923
4.65%	36.562	295.87	0.0000924
6.98%	36.497	295.39	0.0000928
Changes of Battery Life Cycle
−20%	36.888	301.88	0.0000898
0%	36.702	296.36	0.0000919
20%	36.578	293.45	0.0000932
40%	36.490	291.35	0.0000941
60%	36.423	289.75	0.0000948
80%	36.372	288.51	0.0000953
100%	36.330	287.52	0.0000957
Changes of Battery Capacity
−30%	36.813	299.66	0.0000906
−20%	36.776	298.56	0.0000911
−10%	36.739	297.46	0.0000915
0%	36.702	296.36	0.0000919

presents the results of the sensitivity analysis. They show how the identified critical variables affect the obtained values of the LCA, DGC, and eco-efficiency indicators.

In order to interpret the results of the sensitivity analysis and to determine the hierarchy of impact of the individual critical variables on the eco-efficiency of the electricity production technology in a coal-fired power plant using a battery, a graphic interpretation was produced (Figure 4).

The results of the sensitivity analysis allow the determination of the following hierarchy of impact of the individual determinants on the eco-efficiency index value for the electricity production technology in a coal-fired power plant using a battery (from the highest to the lowest):

• Fuel prices—coal and mazout;
• Prices of CO₂ emission allowances;
• Battery efficiency;
• Battery life cycle;
• Decrease of the battery capacity after 10 years of operation;
• The capital expenditures of the battery.

The above hierarchy results from the fact that the largest deviations from the initial value occur in the case of the DGC indicator for changes in fuel prices and CO₂ emission allowances. In the structure of the costs analyzed, the construction of the battery accounts for only approx. 22% of the annual expenditure on the purchase of fuels and CO₂ emission allowances. That is why, in the case of an analysis covering 10 years of battery life, even significant fluctuations in the cost of its construction have a very small impact on the result of the analysis. When studying the sensitivity to changes in the battery performance and lifetime, slight deviations from the baseline occur for the LCA indicator, in combination with slight deviations of the DGC indicator. Therefore, the impact of these two critical variables on the result of the eco-efficiency analysis is also small.

4. Conclusions

From the whole life cycle perspective, the use of LiB to stabilize coal-fired boilers has a dual impact on the environment. On the one hand, the reduction in specific carbon dioxide emissions and consumption of coal achieved through energy storage reduces the impact in the global warming category. However, it also contributes to adverse effects associated with ecotoxicity and the consumption of mineral resources in connection with the use of metals for the production of lithium-ion cells. The aggregation of results obtained by their normalization and weighing has shown that these activities are leveling out.

In the cost analysis of electricity production technology in a coal-fired power plant, the economic benefit of using a battery to optimize the operation of power units should be emphasized.

It was found that technology of electricity production in a coal-fired power plant using a rechargeable battery is more eco-efficient than technology without using a battery.

The analysis of the eco-efficiency sensitivity of electricity production technology in a coal-fired power plant using a rechargeable battery allows the impact of individual variables on the eco-efficiency assessment result to be assessed. Thanks to this, it is possible to indicate decision criteria for the profitability of development projects related to the analyzed use of large-scale electricity storage on the basis of LiB.

The prices of fuels used for electricity production, followed by prices of CO₂ emission allowances, have the greatest impact on the eco-efficiency of electricity production technology in a coal-fired power plant using a rechargeable battery. Other variables affecting the eco-efficiency result are factors related to the battery: its efficiency, life span, decrease of the capacity after 10 years of operation, and construction cost.