The reliable and safe operation of a Battery Energy Storage System (BESS) is paramount for integrating renewable energy and stabilizing the modern grid. Within such systems, lithium-ion batteries, particularly Lithium Iron Phosphate (LFP) chemistry, have become a dominant technology due to their inherent safety, long cycle life, and cost-effectiveness. A critical, yet often underexplored, mechanical phenomenon that directly impacts the performance, longevity, and structural integrity of these systems is the swelling force generated by battery cells during operation. This force arises from the reversible and irreversible volume changes of active materials as lithium ions shuttle between the cathode and anode during charge and discharge cycles. Understanding the evolution of these swelling forces throughout the entire lifecycle of a battery is essential for robust mechanical design, accurate state-of-health (SOH) estimation, and proactive safety management in large-scale Battery Energy Storage System deployments.
This article presents a detailed investigation into the swelling force characteristics of a high-capacity 280 Ah LFP prismatic cell and its behavior when constrained within a module assembly. We explore the fundamental mechanisms driving force generation, analyze its dependency on the state of charge (SOC), and, most importantly, track its evolution as the battery degrades over hundreds of cycles. Furthermore, we scale our analysis from single-cell behavior to module-level performance and utilize the empirical data to inform and validate structural simulation models. The insights gained aim to provide a foundational reference for engineers designing mechanical structures, thermal management systems, and safety protocols for next-generation Battery Energy Storage System solutions.
1. Experimental Methodology and Materials
The core of our study is a commercial 280 Ah prismatic LFP cell, a common building block for stationary storage. The key specifications for the cell and the modules assembled for testing are summarized in Table 1.
| Component | Parameter | Value | Unit |
|---|---|---|---|
| LFP Cell | Dimensions (T x H x W) | 72.0 x 207.2 x 173.7 | mm |
| Nominal Capacity | 280 | Ah | |
| Nominal Voltage | 3.2 | V | |
| AC Internal Resistance (1kHz) | ≤ 0.25 | mΩ | |
| Weight | ≈ 5.42 | kg | |
| Standard Charge/Discharge Power (0.5C) | 448 | W | |
| 1P8S Module (B) | Configuration | 8 cells in series | – |
| Voltage Range | 20.0 – 29.2 | V | |
| Nominal Energy | 7.168 | kWh | |
| Test Power (0.5P) | 3584 | W | |
| 1P12S Module (A) | Configuration | 12 cells in series | – |
| Voltage Range | 30.0 – 38.4 | V | |
| Nominal Energy | 10.752 | kWh | |
| Test Power (0.5P) | 5376 | W |
To simulate the constrained condition within a real Battery Energy Storage System module, cells were assembled into two module configurations: a 1P12S module (Module A) and a 1P8S module (Module B). These modules were then placed into a rigid steel fixture equipped with a high-accuracy force sensor aligned to measure the total swelling force exerted by the cell stack in the through-thickness direction. A controlled initial preload of 300 kgf was applied to ensure consistent mechanical contact and replicate the effect of module assembly bolts.
The cycling test protocol was designed to reflect typical energy storage duty cycles. Tests were conducted at an ambient temperature of 25 ± 2°C. The detailed cycling conditions are provided in Table 2.
| Step | Mode | Power (Module A / B) | Voltage Limits (per cell) | Duration / Next Step Trigger |
|---|---|---|---|---|
| 1 | Constant Power Charge | 5376 W / 3584 W | 3.65 V (max) | Until voltage limit |
| 2 | Rest | 0 W | – | 30 minutes |
| 3 | Constant Power Discharge | 5376 W / 3584 W | 2.50 V (min) | Until voltage limit |
| 4 | Rest | 0 W | – | 30 minutes |
Throughout the test, the force sensor recorded the swelling force synchronously with electrical data (voltage, current, capacity), allowing for a direct correlation between force, SOC, and cycle number.
2. Swelling Force Characteristics During a Single Cycle
The swelling force during the initial charge and discharge cycle reveals a complex, non-linear relationship with SOC, which is fundamental to the operation of any Battery Energy Storage System cell. As shown in the data, the force does not simply increase monotonically with SOC but exhibits distinct peaks and valleys.
During the charge process, starting from approximately 30% SOC, the force rises linearly to a first local peak (denoted as PC1) at around 30-40% SOC. Following this peak, the force undergoes a slight decline, reaching a minimum near 50-60% SOC. Subsequently, the force resumes a strong linear increase, culminating in the absolute maximum force (PC2) at 100% SOC. After a 30-minute rest at full charge, the force relaxes to a slightly lower value due to stress redistribution within the cell materials and fixture. Upon discharge, the force initially decreases from the rested high value. A notable local peak (PD1) appears again around 30-40% SOC during discharge, mirroring the charging peak. Finally, the force drops to its minimum value at 0% SOC, which is typically higher than the starting preload force due to initial irreversible expansion.
This characteristic “double-peak” profile in both charge and discharge is not an artifact but is intrinsically linked to the phase behavior of the graphite anode material. The intercalation of lithium into graphite occurs in distinct stages, each corresponding to specific atomic arrangements and lattice spacings. The overall electrochemical reactions are:
Cathode: $$ \text{LiFePO}_4 \rightleftharpoons \text{Li}_{1-x}\text{FePO}_4 + x\text{Li}^+ + x\text{e}^- $$
Anode: $$ x\text{Li}^+ + x\text{e}^- + 6\text{C} \rightleftharpoons \text{Li}_x\text{C}_6 $$
Overall: $$ \text{LiFePO}_4 + 6\text{C} \rightleftharpoons \text{Li}_{1-x}\text{FePO}_4 + \text{Li}_x\text{C}_6 $$
The staging phenomena in graphite are key. Lithium intercalation progresses through stages IV, III, II(L), II, and finally I, where the stage number indicates the number of graphene layers between two adjacent lithium layers. Each stage transition involves a change in the c-axis lattice parameter (interlayer spacing). Crucially, the volume change (and thus the stress generated) is not constant per inserted lithium ion. The transition from stage IV to stage III/II involves a significant initial lattice expansion. This correlates with the first force peak PC1 at lower SOC. The plateau region where force decreases slightly coincides with the transition to a dilute stage II, where lattice expansion is minimal. The final, sharp rise to peak PC2 at full charge is driven by the transition to stage I (LiC6), where the graphite lattice is fully expanded, resulting in maximum volume change.
This explanation, however, focuses solely on the anode. The LFP cathode also undergoes volume change during (de)lithiation, but in the opposite direction: it contracts upon delithiation (charging). The coupled volume change of both electrodes modulates the observed swelling force. The slight force dip in the mid-SOC range can be partially attributed to this opposing effect, where significant anode expansion is temporarily subdued while the cathode is contracting. The net volumetric strain $\epsilon_{net}(SOC)$ can be conceptually described as a coupling function:
$$ \epsilon_{net}(SOC) = f_v\left( \Theta_{anode}(SOC), \Theta_{cathode}(SOC) \right) $$
where $\Theta$ represents the composition or lithiation degree of each electrode. The resulting stress $\sigma$ in a constrained cell is related to this strain and the effective stiffness of the cell stack.
| Material & Phase | Key Lattice Parameter (c-axis for Graphite, a/b/c for LFP) | Approx. Volumetric Change vs. Pristine |
|---|---|---|
| Graphite (C6) | c ≈ 3.355 Å | 0% (Reference) |
| Graphite (Stage II, LiC18) | c ≈ 3.509 Å | +5.6% |
| Graphite (Stage I, LiC6) | c ≈ 3.706 Å | +12.4% |
| LiFePO4 (Full Lithiation) | a=6.01 Å, b=10.33 Å, c=4.69 Å | +6.5% vs. FePO4 |
| FePO4 (Full Delithiation) | a=5.79 Å, b=9.82 Å, c=4.79 Å | -6.5% vs. LiFePO4 |
3. Evolution of Swelling Force Over the Battery Lifetime
For the long-term durability of a Battery Energy Storage System, understanding how mechanical stresses evolve with aging is as critical as understanding the initial electrical performance. Our long-term cycling tests reveal a significant and systematic evolution of the swelling force profile.
The most striking observation is the changing hierarchy of the force peaks. In early life (Beginning of Life, BOL), the peak at 100% SOC (PC2) is the global maximum. However, as cycling progresses and the cell’s SOH declines, the peak at ~30% SOC during charge (PC1) grows at a faster rate. Typically, after several hundred cycles (corresponding to an SOH of approximately 92-94%), PC1 surpasses PC2 to become the new maximum swelling force experienced by the module. Concurrently, the discharge peak PD1 also grows, eventually exceeding the faded PC2 as well. This transformation is consistent across modules with different series counts.
This evolution can be attributed to the loss of active lithium and the growth of the Solid Electrolyte Interphase (SEI) on the graphite anode. As cycling proceeds, irreversible side reactions consume lithium ions, reducing the total cyclable lithium inventory. Therefore, at 100% SOC in an aged cell, the graphite anode is not fully lithiated to the Stage I (LiC6) composition as it was at BOL. Instead, its lithiation degree corresponds to a lower SOC point from the BOL reference. Since the most dramatic volume expansion occurs in the final stage transition, the attenuation of this final transition leads to a reduced PC2.
Conversely, the mechanisms driving the early SOC peak (PC1) remain active or are even exacerbated. SEI growth and possible particle cracking increase the irreversible thickness growth of the electrode stack. The mechanical stress associated with the phase transitions in the mid-SOC range becomes the dominant contributor to the overall constrained force. Consequently, the swelling force profile effectively “compresses” along the SOC axis as SOH decreases, elevating the significance of the mid-SOC peaks.
Analyzing the maximum swelling force in each cycle against SOH reveals a highly valuable correlation. After the initial nonlinear aging period (down to ~90% SOH), the maximum swelling force exhibits a strong linear relationship with SOH. This linear correlation holds true for both module configurations tested. The relationship can be expressed as:
$$ F_{max} = k \cdot \text{SOH} + F_0 $$
where $F_{max}$ is the maximum swelling force in a cycle, SOH is expressed as a fraction (e.g., 0.9 for 90%), $k$ is a negative constant (slope) representing the force increase per unit decrease in SOH, and $F_0$ is a constant offset.
For our specific modules, linear regression yielded:
For Module A (1P12S): $$ F_{max, A} \approx -7429 \times \text{SOH} + 7533 \quad (R^2 > 0.99) $$
For Module B (1P8S): $$ F_{max, B} \approx -7402 \times \text{SOH} + 7306 \quad (R^2 > 0.99) $$
The near-identical slopes indicate that the fundamental force growth trend per cell is independent of the number of cells in series within the module. The total force scales with the number of cells, but the rate of force increase with degradation is an intrinsic cell property. This linear model provides a potential pathway for a mechanical-based SOH estimation algorithm within a Battery Energy Storage System, supplementing traditional voltage-capacity methods.
4. Module-Level Analysis and Structural Simulation
Transitioning from cell behavior to module and pack design is a critical step in Battery Energy Storage System engineering. The measured swelling forces, especially their growth over life, impose significant demands on module components such as endplates, binding rods or straps, busbars, and the pack enclosure itself.
Our data shows that swelling force is additive. The total force measured on the 1P12S module was consistently higher than that on the 1P8S module under equivalent SOC and SOH conditions, roughly scaling with the number of cells in the stacking direction. This additive effect means that for long strings of cells, the cumulative force can reach very high levels. For instance, in our test, the 1P12S module exhibited a maximum force exceeding 2300 kgf at 70% SOH.
To assess whether standard module designs can withstand these lifecycle forces, we performed finite element analysis (FEA) simulations. Using the linear model extrapolated from test data, we predicted the maximum swelling force at a conservative end-of-life SOH of 60%. This force (minus the initial preload) was applied as a distributed pressure on the cell surfaces within a detailed 3D model of the module, which included endplates, steel binding straps, busbars, and insulating pads.
The simulation results are enlightening. The primary mode of deformation is the bending of the endplates. The maximum displacement occurred at the center of the endplate, with values on the order of 1.5 mm for the simulated force. While noticeable, this deflection is typically within allowable limits for pack integration. More critically, the stress analysis confirmed the structural adequacy of key components:
- Endplates: Maximum von Mises stress was found at reinforcement ribs, but remained safely below the yield strength of the aluminum alloy.
- Steel Straps & Bolts: Stresses in the binding straps and tension bolts were well within their elastic limits, confirming that the mechanical clamping system retains its integrity.
- Busbars: The simulation highlighted an important design consideration. The swelling-induced module expansion can displace cell terminals, applying tensile and bending loads on connecting busbars. Stresses in rigid busbars approached the material yield point. This underscores the necessity of using designed busbars with flexible sections (e.g., folded or arched shapes) to accommodate the relative displacement between cells without causing high stress or fracture, thereby preventing potential electrical failure in the Battery Energy Storage System.
The simulation, grounded in empirical swelling force data, validates that a well-designed module structure can mechanically withstand the predicted lifecycle forces. This process exemplifies a data-driven design approach essential for reliable Battery Energy Storage System development.
5. Conclusion and Implications for Battery Energy Storage System Design
This investigation into the swelling force characteristics of LFP battery modules provides critical insights for the engineering of safe and durable Battery Energy Storage Systems. The key findings are:
- SOC-Dependent Force Profile: Swelling force is non-monotonic with SOC, featuring distinct peaks near 30% SOC and 100% SOC due to phase transitions in the graphite anode, modulated by the opposing volume change of the LFP cathode.
- Force Evolution with Aging: The swelling force profile evolves significantly over the cell’s lifetime. The peak at 100% SOC diminishes relative to the peak at 30% SOC, which becomes the dominant source of mechanical stress after the cell degrades to approximately 92% SOH.
- Linear Correlation with SOH: After the initial aging period, the maximum swelling force in a cycle exhibits a strong linear correlation with SOH. This relationship is intrinsic to the cell and scales with the number of cells in series within a module.
- Additive Nature and Structural Impact: Swelling forces are additive in a constrained stack. While module components like endplates and straps can be designed to withstand these forces, special attention must be paid to busbar design to accommodate displacement and prevent stress-induced failure.
The implications for Battery Energy Storage System design and management are profound. First, mechanical design must account not just for initial forces but for significantly higher forces that develop at mid-life, using predictive models based on data like that presented here. Second, the linear force-SOH correlation opens the door for novel diagnostic and prognostic algorithms. An integrated force sensor in a module could provide a direct, mechanical signal for state-of-health estimation, offering a complementary data stream to electrical measurements. Finally, understanding the swelling force evolution aids in setting appropriate initial preloads during module assembly. An optimal preload must minimize initial interfacial resistance without over-constraining the cell, which can accelerate degradation, and must allow for the predictable force growth over time.
In summary, swelling force is not merely a byproduct of operation but a key state variable that bridges electrochemical behavior, mechanical integrity, and long-term reliability. A deep understanding and proactive management of swelling forces will be a cornerstone in designing the next generation of high-performance, long-lasting, and safe grid-scale Battery Energy Storage Systems.