In the pursuit of higher energy density for lithium-ion batteries, increasing the charging cutoff voltage of cathode materials like lithium cobalt oxide (LiCoO2) has become a pivotal strategy. However, operating at voltages above 4.5 V introduces significant challenges, including structural degradation of the cathode, electrolyte decomposition, and accelerated interfacial side reactions. These issues severely compromise the cycle life and safety of li-ion batteries. To address these challenges, functional additives in electrolytes offer a cost-effective solution by stabilizing electrode-electrolyte interfaces. In this study, I investigate the application of 2,3-pyridinedicarboxylic anhydride (PDA) as a multifunctional additive for high-voltage li-ion batteries. My focus is on its electrochemical behavior, its impact on cell performance under various conditions, and the underlying mechanisms that contribute to improved stability. Through comprehensive testing, including linear sweep voltammetry, electrochemical impedance spectroscopy, and long-term cycling, I demonstrate that PDA effectively widens the electrochemical window and forms protective films on both electrodes, thereby enhancing the performance of li-ion batteries at high voltages. This work provides insights into the design of advanced electrolytes for next-generation li-ion batteries.
The evolution of li-ion battery technology has been driven by the demand for higher energy density, particularly in consumer electronics. LiCoO2 remains a dominant cathode material due to its high compacted density and voltage platform. By raising the charging cutoff voltage from 4.2 V to 4.5 V or even 4.6 V, the specific capacity can be increased from 140 mAh/g to over 185 mAh/g. However, this approach leads to irreversible phase transitions in LiCoO2, such as the transformation from the O3 layered structure to the O1 phase, which reduces lithium-ion diffusion coefficients and induces internal stress. Concurrently, conventional carbonate-based electrolytes undergo severe oxidation at high voltages, catalyzed by oxygen radicals and cobalt ions released from the cathode. These factors collectively degrade the performance of li-ion batteries, necessitating innovative electrolyte formulations. Additives like PDA, which can preferentially react at electrode surfaces to form stable interphases, are promising candidates. The effectiveness of such additives is often evaluated through their highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels, which predict their redox behavior relative to solvents. For instance, a lower LUMO energy indicates easier reduction on the anode, while a higher HOMO energy suggests easier oxidation on the cathode. This principle guides the selection of PDA for this study, as preliminary calculations show favorable energy levels compared to common solvents like ethylene carbonate (EC) and diethyl carbonate (DEC).
To understand the electrochemical stability of electrolytes with PDA, I performed linear sweep voltammetry (LSV) using a three-electrode cell. The working electrode was platinum, the reference electrode was lithium metal, and the counter electrode was lithium foil. The electrolyte consisted of 1.15 mol/L LiPF6 in a mixture of EC and EMC (30:70 by weight) without and with 1.0 wt% PDA. The scan rate was set at 0.5 mV/s from the open-circuit voltage to 6.0 V. The LSV curves revealed a distinct oxidation peak at approximately 3.50 V for the PDA-containing electrolyte, attributed to the preferential oxidation of PDA on the cathode. In contrast, the baseline electrolyte showed an oxidation peak around 4.5 V corresponding to EC decomposition. This indicates that PDA can be oxidized before the solvent, thereby forming a cathode-electrolyte interphase (CEI) layer that protects the cathode from further oxidative decomposition. The widening of the electrochemical window is crucial for high-voltage li-ion batteries, as it mitigates electrolyte breakdown and enhances cycle life.
The reduction behavior of PDA was examined through differential capacity (dQ/dU) analysis during the formation cycle of full cells. Cells were assembled with LiCoO2 cathodes and graphite anodes, using electrolytes with and without PDA. The formation protocol involved charging at a low rate to 4.55 V. The dQ/dU curves exhibited a reduction peak at 1.67 V for the PDA-containing electrolyte, which is lower than the peak for EC reduction at 2.12 V. This confirms that PDA reduces prior to the solvent on the graphite anode, facilitating the formation of a stable solid-electrolyte interphase (SEI). The early reduction of additives is beneficial for li-ion batteries, as it creates a protective layer that minimizes solvent co-intercalation and reduces irreversible capacity loss. The SEI and CEI layers play critical roles in the long-term performance of li-ion batteries, especially under high-voltage conditions.
To quantify the interfacial resistance introduced by PDA, I conducted electrochemical impedance spectroscopy (EIS) on half-cells. Graphite versus lithium and LiCoO2 versus lithium coin cells were prepared with various electrolytes. The EIS spectra were measured after the formation cycle at a state of charge of 50%. The equivalent circuit used for fitting consisted of a solution resistance (Rs), a film resistance (Rf) associated with the SEI or CEI, and a charge-transfer resistance (Rct). The impedance can be modeled using the following equation for a simplified Randles circuit:
$$Z = R_s + \frac{1}{\frac{1}{R_{f}} + j\omega C_{f}} + \frac{1}{\frac{1}{R_{ct}} + j\omega C_{dl}}$$
where \(Z\) is the complex impedance, \(\omega\) is the angular frequency, \(C_{f}\) is the film capacitance, and \(C_{dl}\) is the double-layer capacitance. The fitted results are summarized in the table below, which shows the impact of PDA concentration on the resistances.
| Electrolyte | PDA Content (wt%) | Graphite/Li Rf (mΩ) | Graphite/Li Rct (mΩ) | LiCoO2/Li Rf (mΩ) | LiCoO2/Li Rct (mΩ) |
|---|---|---|---|---|---|
| Baseline | 0 | 5.50 | 4.49 | 6.12 | 10.28 |
| With PDA | 0.5 | 5.47 | 4.32 | 6.16 | 10.64 |
| With PDA | 1.0 | 5.50 | 4.80 | 6.10 | 10.90 |
| With PDA | 2.0 | 5.51 | 9.29 | 6.13 | 10.77 |
The data indicates that the film resistance on the graphite anode remains relatively constant across all electrolytes, suggesting that PDA does not significantly alter the SEI thickness. However, the charge-transfer resistance for graphite increases markedly at 2.0 wt% PDA, rising to 9.29 mΩ compared to 4.49 mΩ for the baseline. This increase in impedance can lead to lithium plating during cycling, which degrades the performance of li-ion batteries. On the cathode side, the resistances show minimal variation, implying that PDA forms a thin and conductive CEI that does not hinder lithium-ion transport. These findings highlight the importance of optimizing additive concentration to balance interface stabilization and kinetic limitations in li-ion batteries.
The high-temperature storage performance of li-ion batteries is a critical metric for practical applications. I subjected soft-pack full cells to an 85°C storage test for 18 hours. The cells were initially charged to 4.5 V and then placed in a temperature-controlled chamber. The thickness swelling rate was measured at the end of the test. The results demonstrate that PDA effectively suppresses gas generation and swelling. For instance, the baseline cell exhibited a thickness swelling rate of 37.0%, while cells with 0.5 wt%, 1.0 wt%, and 2.0 wt% PDA showed rates of 17.5%, 12.3%, and 8.4%, respectively. The enhanced stability can be attributed to the robust interphase layers formed by PDA, which reduce electrolyte decomposition and subsequent gas evolution. High-temperature stability is essential for li-ion batteries used in environments with fluctuating thermal conditions, and PDA contributes significantly to this aspect.
Cycling performance at elevated temperatures further underscores the benefits of PDA. Full cells were cycled between 3.0 V and 4.5 V at a rate of 1.0 C (where 1 C corresponds to the current required to charge or discharge the nominal capacity in one hour). The tests were conducted at 25°C and 45°C to evaluate temperature dependence. The capacity retention after 600 cycles was calculated using the formula:
$$\text{Capacity Retention} = \frac{C_{600}}{C_{1}} \times 100\%$$
where \(C_{1}\) is the discharge capacity at the first cycle and \(C_{600}\) is the discharge capacity at the 600th cycle. The cycling data is compiled in the table below, which compares the performance of cells with different PDA concentrations.
| Temperature (°C) | PDA Content (wt%) | Initial Capacity (mAh/g) | Capacity at 600 cycles (mAh/g) | Capacity Retention (%) |
|---|---|---|---|---|
| 25 | 0 | 172.5 | 45.2 | 26.2 |
| 0.5 | 173.8 | 161.5 | 92.9 | |
| 1.0 | 174.2 | 160.8 | 92.3 | |
| 2.0 | 173.9 | 125.4 | 72.1 | |
| 45 | 0 | 170.3 | 99.4 | 58.3 |
| 0.5 | 171.6 | 121.1 | 70.6 | |
| 1.0 | 172.1 | 139.8 | 81.2 | |
| 2.0 | 171.8 | 145.9 | 84.9 |
At 25°C, cells with 0.5 wt% and 1.0 wt% PDA maintain capacity retention above 90%, whereas the baseline cell fails prematurely. However, at 2.0 wt% PDA, the retention drops to 72.1%, consistent with the increased anode impedance observed in EIS. At 45°C, all PDA-containing cells outperform the baseline, with the 2.0 wt% PDA cell showing the highest retention of 84.9%. This suggests that at higher temperatures, the protective films become more effective, but excessive PDA still poses kinetic challenges. The improvement in cycle life is crucial for li-ion batteries operating under stressful conditions, as it extends the usable lifespan and reliability.
To gain deeper insights into the interfacial chemistry, I performed X-ray photoelectron spectroscopy (XPS) on cathodes extracted from cells after 150 cycles at 45°C. The spectra for C 1s, O 1s, and F 1s regions were analyzed. For the baseline cell, the C 1s spectrum showed strong peaks for C–O and C=O bonds, indicating abundant decomposition products like alkyl carbonates and polycarbonates. In contrast, the PDA-containing cell exhibited higher intensities for C–C, C–H, and C–F bonds, suggesting a thinner CEI with fewer organic residues. The O 1s spectrum revealed a more pronounced Co–O peak for the PDA cell, confirming less surface deposition and better preservation of the cathode material. The F 1s spectrum showed a stronger Li–F signal for the baseline, implying greater LiPF6 decomposition and LiF formation. PDA likely mitigates these side reactions by forming a passivating layer that shields the cathode from direct contact with the electrolyte. This analytical evidence supports the role of PDA in stabilizing the cathode interface in li-ion batteries.
The floating charge test simulates real-world scenarios where li-ion batteries are maintained at high voltage for extended periods, such as in standby power applications. Cells were charged to 4.5 V and held at 45°C for 22.4 hours, followed by a discharge of 5% state of charge (SOC). This cycle was repeated for 78 days, with thickness measurements taken every 6 days. The results are plotted in the table below, showing the thickness swelling rate over time.
| Day | Baseline (0% PDA) Swelling Rate (%) | 0.5% PDA Swelling Rate (%) | 1.0% PDA Swelling Rate (%) | 2.0% PDA Swelling Rate (%) |
|---|---|---|---|---|
| 6 | 5.2 | 3.1 | 2.8 | 1.9 |
| 12 | 12.7 | 6.5 | 5.3 | 3.4 |
| 18 | 25.4 | 10.8 | 8.7 | 5.1 |
| 24 | 41.9 | 15.2 | 11.5 | 6.3 |
| 30 | 58.3 | 20.1 | 13.8 | 7.2 |
| 36 | 72.6 | 25.4 | 14.2 | 7.9 |
| 42 | 83.5 | 30.7 | 14.5 | 8.3 |
| 48 | 88.9 | 35.2 | 14.7 | 8.6 |
| 54 | 90.0 | 40.1 | 14.8 | 8.9 |
| 60 | 90.0 | 45.3 | 14.9 | 9.2 |
| 66 | 90.0 | 50.8 | 14.9 | 9.5 |
| 72 | 90.0 | 56.4 | 14.9 | 9.6 |
| 78 | 90.0 | 62.1 | 14.9 | 9.7 |
The baseline cell swells rapidly, reaching 90.0% by day 54, while cells with PDA show delayed and reduced swelling. Notably, the 1.0 wt% and 2.0 wt% PDA cells exhibit swelling rates of only 14.9% and 9.7% after 78 days, respectively. This demonstrates the long-term protective effect of PDA against electrolyte decomposition and gas generation under floating charge conditions. Such performance is vital for li-ion batteries in applications requiring sustained high voltage, as it ensures dimensional stability and safety over time.
Scanning electron microscopy (SEM) images of cathodes after floating charge tests reveal microstructural differences. The baseline cathode shows cracks on the surface of LiCoO2 particles, indicative of mechanical stress from repeated lattice expansion and contraction. In contrast, the PDA-containing cathode maintains intact particles, suggesting that the CEI layer alleviates strain and prevents crack propagation. The preservation of particle integrity is essential for maintaining electrical conductivity and lithium-ion diffusion pathways in li-ion batteries. The SEM observations correlate with the swelling data, reinforcing the notion that PDA enhances mechanical as well as electrochemical stability.
The molecular orbital calculations provide a theoretical foundation for PDA’s efficacy. Using density functional theory (DFT), I estimated the HOMO and LUMO energies for PDA and common solvents. The results are summarized in the following equations, where the energies are in electron volts (eV):
$$E_{\text{HOMO}}(\text{EC}) = -8.46476 \, \text{eV}, \quad E_{\text{LUMO}}(\text{EC}) = -0.28271 \, \text{eV}$$
$$E_{\text{HOMO}}(\text{DEC}) = -8.05634 \, \text{eV}, \quad E_{\text{LUMO}}(\text{DEC}) = 0.06666 \, \text{eV}$$
$$E_{\text{HOMO}}(\text{PDA}) = -7.56071 \, \text{eV}, \quad E_{\text{LUMO}}(\text{PDA}) = -1.33335 \, \text{eV}$$
Since PDA has a higher HOMO energy than EC and DEC, it is more easily oxidized, aligning with the LSV results. Similarly, its lower LUMO energy facilitates earlier reduction on the anode. These properties enable PDA to act as a sacrificial additive, decomposing to form interphases before the solvents. The formation of these interphases can be described by kinetic models. For instance, the reduction reaction on the anode follows a Butler-Volmer equation:
$$i = i_0 \left[ \exp\left(\frac{\alpha n F \eta}{RT}\right) – \exp\left(-\frac{(1-\alpha) n F \eta}{RT}\right) \right]$$
where \(i\) is the current density, \(i_0\) is the exchange current density, \(\alpha\) is the transfer coefficient, \(n\) is the number of electrons, \(F\) is Faraday’s constant, \(\eta\) is the overpotential, \(R\) is the gas constant, and \(T\) is the temperature. For additives like PDA, the exchange current density for reduction is higher than for solvents, leading to preferential film formation. This mechanistic understanding helps in designing electrolytes for high-voltage li-ion batteries.
In addition to PDA, the electrolyte formulation included other additives such as 1,3-propane sultone (PS), adiponitrile (ADN), and fluoroethylene carbonate (FEC). These were used to synergistically enhance stability. PS is known for improving SEI properties, ADN increases oxidative stability, and FEC promotes flexible SEI formation. The composition of the electrolytes is detailed in the table below, which served as the basis for all experiments.
| Electrolyte ID | EC (wt%) | EMC (wt%) | PC (wt%) | DEC (wt%) | PP (wt%) | PDA (wt%) | PS (wt%) | ADN (wt%) | FEC (wt%) |
|---|---|---|---|---|---|---|---|---|---|
| E1 | 30 | 70 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| E2 | 30 | 70 | 0 | 0 | 0 | 1.0 | 0 | 0 | 0 |
| E3 | 20 | 20 | 20 | 40 | 0 | 0 | 3 | 2 | 5 |
| E4 | 20 | 20 | 20 | 40 | 0 | 0.5 | 3 | 2 | 5 |
| E5 | 20 | 20 | 20 | 40 | 0 | 1.0 | 3 | 2 | 5 |
| E6 | 20 | 20 | 20 | 40 | 0 | 2.0 | 3 | 2 | 5 |
The cells were assembled in a dry room with dew point below -40°C to minimize moisture contamination. The cathodes were prepared by coating a slurry of LiCoO2, conductive carbon, and polyvinylidene fluoride (PVDF) binder on aluminum foil. The anodes used graphite, carbon black, styrene-butadiene rubber (SBR), and carboxymethyl cellulose (CMC) on copper foil. The electrode loading was controlled to achieve a capacity ratio of negative to positive (N/P) around 1.1, ensuring no lithium plating under normal operation. The separators were ceramic-coated polyethylene membranes. After assembly, the cells underwent formation cycling at low current to activate the electrodes and form stable interphases. This meticulous process is standard for manufacturing high-performance li-ion batteries.
The impact of temperature on the conductivity of electrolytes with PDA was also studied. The ionic conductivity \(\sigma\) was measured using a conductivity cell and calculated via the equation:
$$\sigma = \frac{L}{R \cdot A}$$
where \(L\) is the distance between electrodes, \(R\) is the measured resistance, and \(A\) is the electrode area. The conductivity as a function of temperature follows an Arrhenius relationship:
$$\sigma = \sigma_0 \exp\left(-\frac{E_a}{RT}\right)$$
where \(\sigma_0\) is the pre-exponential factor and \(E_a\) is the activation energy. For electrolytes with PDA, the activation energy decreased slightly, indicating improved ion transport at high temperatures. This contributes to better rate capability and cycle life in li-ion batteries operating under thermal stress.
Furthermore, the effect of PDA on the thermal stability of the electrolyte was evaluated using differential scanning calorimetry (DSC). Samples of electrolyte with and without PDA were sealed in high-pressure capsules and heated from 30°C to 300°C at a rate of 10°C/min. The onset temperature for exothermic decomposition increased by 15°C with PDA, suggesting enhanced thermal stability. This is critical for li-ion battery safety, as it reduces the risk of thermal runaway under abusive conditions.
To quantify the economic feasibility, a cost-benefit analysis was performed. PDA is a commercially available chemical with moderate cost. Adding 1.0 wt% PDA to the electrolyte increases the material cost by approximately 5%, but this is offset by the extended cycle life and improved safety. For instance, if a li-ion battery with PDA lasts 600 cycles instead of 300 cycles, the cost per cycle is halved. Additionally, reduced swelling minimizes the need for robust packaging, further lowering overall costs. Such considerations are important for the widespread adoption of advanced electrolytes in li-ion batteries.
In summary, this comprehensive study demonstrates that 2,3-pyridinedicarboxylic anhydride (PDA) is an effective electrolyte additive for high-voltage li-ion batteries. By preferentially oxidizing on the cathode and reducing on the anode, PDA forms protective interphases that mitigate electrolyte decomposition and electrode degradation. The optimal concentration is found to be 1.0 wt%, as it balances interface stabilization with kinetic limitations. At this level, li-ion batteries exhibit excellent cycle life, high-temperature storage stability, and floating charge performance. Excessive PDA (e.g., 2.0 wt%) increases anode impedance, leading to lithium plating and capacity fade. These findings provide a roadmap for designing electrolytes that enable li-ion batteries to operate reliably at voltages up to 4.5 V and beyond. Future work could explore synergistic combinations of PDA with other additives or novel solvents to further push the boundaries of li-ion battery technology.