As the global energy landscape shifts toward sustainability, the rapid proliferation of clean energy sources such as wind and solar has been accompanied by the increasing adoption of new energy vehicles. These vehicles rely heavily on power batteries, which inevitably degrade over time and must be retired when their capacity falls to 70%–80% of the initial value. However, these retired batteries still retain significant value and can be repurposed for grid-scale battery energy storage system applications. This practice, known as cascade utilization, extends the lifecycle of power batteries and enhances resource efficiency. In this study, I develop a comprehensive framework integrating life cycle assessment and life cycle cost analysis to evaluate both the social and economic benefits of a cascade utilization battery energy storage system. The system under investigation is a 10 MW / 30 MWh storage facility that uses retired lithium iron phosphate batteries, intended to support wind farm operations by smoothing output and reducing penalties under grid regulations.
I first construct a life cycle assessment model covering five main stages: raw material production, first use in electric vehicles, battery reassembly, second use in stationary energy storage, and end-of-life recycling. The functional unit is defined as one kilowatt-hour of nominal battery capacity. The system boundaries include all material and energy inputs and outputs from cradle to grave. For production, I collect inventory data from a lithium battery manufacturing project environmental report, including the consumption of cathode materials, anode materials, electrolyte, separators, and assembly energy. The first use stage accounts for electricity losses due to battery efficiency and the weight penalty of the battery pack. I use Chinese national average electric vehicle parameters: a driving range of 469 km, curb mass of 1836 kg, battery mass of 419 kg, and energy consumption rate of 17 kWh per 100 km. The energy loss during the first life is calculated as:
$$E_e = F_v \times D_v \times (1 – \eta_{k1})$$
$$E_m = k \times F_v \times D_v \times (m / m_T)$$
where $F_v$ is the energy consumption limit, $D_v$ the total mileage, $\eta_{k1}$ the average round-trip efficiency (assumed 90%), $k$ the mass-related coefficient (0.49), $m$ the battery mass, and $m_T$ the vehicle curb mass.
During the reassembly stage, the retired batteries are dismantled, inspected, and repackaged with new cables, enclosures, and battery management systems. The second use stage applies the repurposed batteries to a battery energy storage system for wind or photovoltaic power smoothing. The capacity degradation over cycles is modeled using a semi-empirical formula:
$$\xi = A \times e^{-(E_a / R t)} \times L^z$$
where $A = 0.1825$, $E_a/R t = 4.443$ K, and $z = 0.5878$ for LFP batteries. The second life is defined from 80% to 60% of initial capacity. Electricity losses in this stage arise from battery internal resistance, inverter efficiency, station auxiliary loads, and self-discharge. The round-trip loss per cycle is:
$$E_{\text{round-trip}} = \sum_{l=1}^{L} E \times (1 – \xi) \times DoD \times (1 – \eta_T \times \eta_{k2})$$
where $DoD$ is depth of discharge, $\eta_T$ transmission efficiency, $\eta_{k2}$ charging/discharging efficiency. Station operation loss is $E_{op} = L \times P_{con} \times t / f$, and self-discharge loss is $E_{self} = E \times \eta_{self} \times t_{stor}$. Two recycling technologies are considered: hydrometallurgical recovery (acid leaching to recover lithium chloride) and direct physical recycling (mechanical separation of components).
I then apply the ReCiPe 2016 midpoint method using SimaPro software to calculate five environmental impact indicators: global warming potential, fine particulate matter formation, terrestrial acidification, marine eutrophication potential, and fossil resource scarcity. Four scenarios are defined: (1) second use for wind + hydrometallurgical recycling, (2) wind + physical recycling, (3) photovoltaic + hydrometallurgical, (4) photovoltaic + physical. The results for GWP are shown in Table 1.
| Scenario | 1 (Wind+Wet) | 2 (Wind+Physical) | 3 (PV+Wet) | 4 (PV+Physical) |
|---|---|---|---|---|
| Production | 97.5 | 97.5 | 97.5 | 97.5 |
| First use | 122.0 | 122.0 | 122.0 | 122.0 |
| Reassembly | 2.7 | 2.7 | 2.7 | 2.7 |
| Second use | 9.8 | 9.8 | 27.8 | 27.8 |
| Recycling | -38.0 | 19.0 | -38.0 | 19.0 |
| Net GWP | 194.0 | 251.0 | 212.0 | 268.0 |
Scenario 1 (wind energy storage with hydrometallurgical recycling) yields the lowest GWP of 194 kg CO₂-eq. The production stage contributes 97.5, primarily from electricity consumption (32%) and cathode material (24%). The first use stage, even though smaller in total energy throughput, has a high impact because it draws from the grid (mainly fossil-based). In contrast, the second use stage benefits from clean energy, with wind second use having half the impact of photovoltaic. Hydrometallurgical recycling provides significant credits due to recovered lithium chloride, whereas physical recycling produces fewer credits. Other indicators follow similar trends. Table 2 summarizes the results for fine particulate matter formation, terrestrial acidification, marine eutrophication, and fossil resource scarcity.
| Indicator | Scenario 1 | Scenario 2 | Scenario 3 | Scenario 4 |
|---|---|---|---|---|
| FPMF (kg PM₂.₅-eq) | 0.394 | 0.491 | 0.405 | 0.502 |
| TA (kg SO₂-eq) | 0.943 | 1.132 | 0.928 | 1.116 |
| MEP (kg P-eq) | -0.011 | -0.008 | -0.010 | -0.009 |
| FRS (kg oil-eq) | 43.7 | 57.7 | 49.5 | 63.4 |
Scenario 1 consistently performs best across all indicators except terrestrial acidification, where Scenario 3 (photovoltaic with wet recycling) is slightly better due to differences in acid gas emissions from secondary use. The sensitivity analysis for GWP in Scenario 1 reveals that the first-life charging efficiency has the greatest impact: a 10% increase in $\eta_{k1}$ reduces GWP by 19 units, while a 10% increase in production energy consumption only adds 3 units. This emphasizes the importance of improving battery efficiency during the primary automotive phase.
Moving to economic analysis, I apply a life cycle cost model to the 10 MW/30 MWh battery energy storage system. The net present value is calculated as:
$$V_{NPV} = \sum_{n=0}^{N} C_{I,n}(1+r)^{-n} – \sum_{n=0}^{N} C_{O,n}(1+r)^{-n}$$
where $C_{I,n}$ and $C_{O,n}$ are cash inflows and outflows in year $n$, and $r=8\%$ is the discount rate. The total life cycle cost includes initial investment (10.691 million yuan for the 5-year system, covering retired battery purchase, BMS, power conversion, and balance-of-plant), annual O&M (0.574 million), charging costs (varying with electricity price), replacement costs (when the battery system needs renewal every 5 years), and salvage value (0.5% of initial investment). The levelized cost of electricity is defined as:
$$C_{LCOE} = \frac{C_{total}}{E_{total}}$$
where $E_{total}$ is the net present value of total discharged electricity over the system lifetime. The discharged electricity calculation accounts for self-discharge, depth of discharge, charging/discharging efficiency, and annual cycles. Table 3 presents the LCOE breakdown for system lifetimes of 5 and 15 years.
| Cost Component | 5-year system | 15-year system |
|---|---|---|
| Initial investment | 3.17 (87.4%) | 1.48 (60.8%) |
| O&M | 0.15 (4.2%) | 0.15 (6.3%) |
| Charging | 0.31 (8.7%) | 0.31 (12.9%) |
| Replacement | — | 0.49 (20.1%) |
| Salvage | -0.01 (-0.3%) | -0.00 (-0.1%) |
| Total LCOE | 3.63 | 2.44 |
For the 5-year system (no battery replacement), the LCOE is 3.63 yuan/kWh, with initial investment contributing 87%. When the system operates for 15 years, requiring two battery replacements, the LCOE drops to 2.44 yuan/kWh because the fixed initial investment is spread over more electricity output. However, the net present value for the system after 1000 cycles is -42.066 million yuan, indicating that current cost structures are still not profitable. Sensitivity analysis on LCOE is performed for state of charge, depth of discharge, charging electricity price, and system lifetime. Results shown in Table 4 highlight the most influential parameters.
| Parameter Variation | LCOE Change (yuan/kWh) |
|---|---|
| SOC increases from 80% to 95% | Decreases from 2.48 to 2.33 |
| DoD increases from 76% to 84% | Decreases from 2.47 to 2.42 |
| Charging price from 0.283 to 0.673 yuan/kWh | Increases from 2.33 to 2.87 |
| System lifetime from 5 to 15 years | Decreases from 3.63 to 2.44 |
The LCOE is most sensitive to charging electricity price and system lifetime. Using low-cost wind power for charging (0.283 yuan/kWh) can reduce LCOE to 2.33 yuan/kWh, while relying on photovoltaic generation (0.673 yuan/kWh) pushes it up to 2.87 yuan/kWh. Extending the system lifetime beyond 15 years would further lower LCOE, but battery degradation limits practical second-life duration.
In conclusion, the cascade utilization of retired power batteries in a battery energy storage system offers substantial environmental benefits when paired with clean energy sources and efficient recycling technologies. Scenario 1 (wind + hydrometallurgical) achieves a GWP of only 194 kg CO₂-eq per kWh capacity, outperforming other combinations. However, the economic viability remains challenging: the LCOE of 2.44 yuan/kWh (15-year lifetime) is still higher than that of new battery systems or other grid storage options, primarily due to high initial investment and replacement costs. As the technology matures and economies of scale improve, the costs of second-life battery energy storage systems are expected to decline. Policy support, such as subsidies for retired battery collection and tax incentives for green storage, could accelerate commercialization. This comprehensive analysis provides a quantitative foundation for decision-makers to optimize both social and economic outcomes of cascade battery energy storage systems, contributing to resource efficiency and environmental protection.