I began this study from a practical and urgent problem: the electric vehicle battery pack is becoming one of the most important engineered systems in modern transportation, yet its end-of-life treatment is still limited by the difficulty of separating bonded components. In an electric vehicle battery pack, many cells are fixed to metal modules, cooling plates, and load-bearing frames with polyurethane structural adhesives. These adhesives provide high strength, toughness, electrical insulation, vibration damping, sealing, and thermal management. They help keep the electric vehicle battery pack stable under vibration, impact, temperature cycling, and moisture exposure. However, the same performance that makes polyurethane structural adhesives useful during service also makes their removal difficult during dismantling, repair, and recycling. My objective was therefore to investigate chemical degradation methods that can remove polyurethane structural adhesives from an electric vehicle battery pack without destroying the battery components or creating unacceptable environmental risks.
The electric vehicle battery pack is not a simple collection of cells. It is a multi-material assembly in which electrochemical cells, busbars, thermal pads, cooling channels, structural plates, insulation films, and adhesives all interact. Polyurethane adhesives are widely used because they can bond dissimilar materials, tolerate movement, and maintain mechanical integrity over long periods. In many electric vehicle battery pack designs, the adhesive is applied as a two-component system, mixed, dispensed, and cured at ambient or slightly elevated temperature. The cured adhesive forms a crosslinked network containing urethane linkages, urea linkages, ether or ester segments, and sometimes fillers such as flame retardants or thermally conductive particles. This network is precisely why the electric vehicle battery pack can survive harsh operating conditions, but it also creates a recycling barrier.
When an electric vehicle battery pack reaches retirement, it may still retain a substantial fraction of its original capacity. In many cases, the residual capacity is high enough for second-life applications such as stationary storage, backup power, and low-speed electric vehicles. If the electric vehicle battery pack can be dismantled cleanly, the cells can be reused, remanufactured, or recycled more efficiently. If the adhesive cannot be removed, mechanical cutting and scraping may damage cells, release electrolytes, and contaminate the recovered material streams. Incineration can destroy organic adhesives, but it may also destroy valuable components and emit hazardous gases. Ultrasonic cleaning can be mild, but it requires sealing and may not be suitable for large electric vehicle battery pack modules. These limitations motivated me to focus on chemical degradation, especially alcoholysis, as a controllable route for adhesive removal.
I considered several chemical degradation pathways for polyurethane. The main routes include alcoholysis, aminolysis, hydrolysis, ammonolysis, alkali degradation, phosphorolysis, and thermal degradation. Each route has different reaction conditions, products, and practical limitations. In my work, I treated the degradation of the electric vehicle battery pack adhesive as a polymer depolymerization and de-crosslinking problem. The goal was not only to dissolve the adhesive but to break its covalent network into smaller, soluble, or dispersible fragments that can be separated from metal and cell surfaces. The general polyurethane linkage can be represented as a carbamate group:
$$ \mathrm{-O-C(=O)-NH-} $$
The formation of polyurethane can be simplified as the reaction between a polyol and a diisocyanate:
$$ \mathrm{HO-R_1-OH + O=C=N-R_2-N=C=O \rightarrow \cdots -O-R_1-O-C(=O)-NH-R_2-NH-C(=O)-O-\cdots} $$
During alcoholysis, a small diol attacks the carbamate linkage and causes chain scission through transesterification. A simplified alcoholysis reaction is:
$$ \mathrm{R-O-C(=O)-NH-R’ + HO-R”-OH \rightarrow R-OH + R”-O-C(=O)-NH-R’ + chain\ scission} $$
I evaluated the degradation process using mass loss, degradation rate, Fourier-transform infrared spectroscopy, rheology, and electrospray ionization mass spectrometry. The degradation rate was calculated from the initial and final masses of the polyurethane sample:
$$ D = \frac{m_1 – m_2}{m_1} \times 100\% $$
Here, \(m_1\) is the mass before degradation, and \(m_2\) is the mass after degradation. I also used swelling tests to estimate crosslink density and solvent compatibility. The swelling ratio was calculated as:
$$ S = \frac{m_2 – m_1}{m_1} \times 100\% $$
where \(m_1\) is the initial mass and \(m_2\) is the mass after swelling equilibrium. A higher swelling ratio generally indicates a looser network and easier solvent penetration. These measurements helped me relate adhesive structure to degradation behavior in the electric vehicle battery pack context.
| Degradation Method | Typical Agent | Catalyst or Additive | Typical Temperature | Main Products | Practical Limitation |
|---|---|---|---|---|---|
| Alcoholysis | Diethylene glycol, propylene glycol, polyols | NaOH, KOH, ionic liquids | 125–250 °C | Polyols, carbamates, amine derivatives | High temperature for full degradation |
| Aminolysis | Alkyl amines, aromatic amines | None or basic additives | Lower than alcoholysis | Hydroxyl- and amine-containing fragments | Odor, toxicity, difficult product control |
| Hydrolysis | Water or steam | Acid or base | 200–300 °C | Polyols, polyamines, CO₂ | High pressure, energy intensive |
| Ammonolysis | Ammonia | High pressure | About 220 °C | Polyols, polyamines, ureas | Severe equipment requirements |
| Alkali degradation | Strong base | NaOH or KOH | About 160 °C | Polyols, aromatic amines, carbonates | Corrosion and high cost |
| Phosphorolysis | Phosphate esters | Heat | Moderate | Phosphorus-containing oligomers | Limited product market |
| Thermal degradation | Heat | Controlled oxygen | High | Isocyanates, polyols, gases | Emissions and poor selectivity |
Among these methods, alcoholysis attracted my attention because it can be performed with common diols such as diethylene glycol. Diethylene glycol has two hydroxyl groups, good polarity, and the ability to swell polyurethane. In my first experimental series, I used sodium hydroxide as a conventional basic catalyst and diethylene glycol as the degradation agent. I studied temperature, reaction time, degradation agent composition, and catalyst dosage. I also tested the method on several commercial two-component polyurethane adhesives to determine whether the formulation was generally applicable to the electric vehicle battery pack adhesive problem.
I first examined the effect of temperature on complete degradation. As the temperature increased, the time required for complete degradation decreased sharply. The reason is that higher temperature increases the fraction of activated molecules and the frequency of effective collisions. According to Arrhenius behavior:
$$ k = A e^{-E_a/(RT)} $$
where \(k\) is the rate constant, \(A\) is the pre-exponential factor, \(E_a\) is the activation energy, \(R\) is the gas constant, and \(T\) is the absolute temperature. A higher temperature increases \(k\), so the polyurethane network breaks down faster. In my tests, one adhesive required 7.43 h at 125 °C, while another required 8.10 h at the same temperature. At 245 °C, the complete degradation times dropped to 10 min and 5 min, respectively. This demonstrated that temperature is a dominant variable for adhesive removal from an electric vehicle battery pack.
| Temperature | Complete Degradation Time for Adhesive A | Complete Degradation Time for Adhesive B | Observation |
|---|---|---|---|
| 125 °C | 7.43 h | 8.10 h | Slow but possible |
| 165 °C | Much shorter | Shorter than at 125 °C | Practical laboratory condition |
| 245 °C | 10 min | 5 min | Very fast but too severe for cells |
I also observed that two adhesives with similar appearance behaved differently under identical conditions. To understand this, I performed swelling tests in diethylene glycol, diethanolamine, and water. Diethylene glycol produced the highest swelling, followed by diethanolamine, and water produced the least swelling. One adhesive swelled more than the other. I interpreted this as evidence that the less-swollen adhesive had a higher crosslink density. A higher crosslink density restricts solvent penetration and slows degradation. This is important for the electric vehicle battery pack because different manufacturers may use different adhesive formulations, and a removal method must tolerate such variability.
| Solvent | Swelling Ratio of Adhesive A | Swelling Ratio of Adhesive B | Interpretation |
|---|---|---|---|
| Diethylene glycol | 2.30% | 1.30% | Best swelling, strong interaction with urethane groups |
| Diethanolamine | 1.70% | 1.10% | Moderate swelling, amine interactions |
| Water | 1.30% | 0.95% | Lowest swelling |
Next, I compared diethylene glycol, diethanolamine, and their mixtures as degradation agents. The mass loss changed non-monotonically as the diethylene glycol fraction increased. At low diethylene glycol content, diethanolamine contributed strong nucleophilic attack and caused relatively fast chain scission. As the diethylene glycol fraction increased, the mixture became richer in a glycol that could swell the network and participate in transesterification. Pure diethylene glycol gave a mass loss of 2.578 g, while pure diethanolamine gave 1.856 g under the same conditions. Therefore, I selected diethylene glycol as the main degradation agent for further optimization. The reason is that diethylene glycol has two chemically equivalent hydroxyl groups, high polarity, and good ability to attack the carbonyl carbon of the carbamate linkage. By contrast, diethanolamine contains an amine group that can promote crosslinking or urea formation, which may compete with degradation.
| Degradation Agent | Mass Loss after 1 h at 165 °C | Main Chemical Action | Selection |
|---|---|---|---|
| Diethylene glycol | 2.578 g | Transesterification and swelling | Preferred |
| Diethanolamine | 1.856 g | Nucleophilic amine attack, possible crosslinking | Less effective |
| Mixed glycol/amine | Intermediate and non-monotonic | Competing alcoholysis and aminolysis | Not optimal |
I then optimized the sodium hydroxide dosage. With increasing sodium hydroxide, the mass loss first increased and then decreased. The maximum occurred at 0.2 g. In this range, hydroxide ions deprotonate diethylene glycol and increase the electron density on the hydroxyl oxygen, making it more nucleophilic. The activated oxygen attacks the carbamate carbonyl carbon, leading to C–O bond cleavage and network breakdown. However, excess base can shift the equilibrium, promote side reactions, or alter the ionic environment, so the degradation rate decreases after the maximum. I therefore fixed the sodium hydroxide dosage at 0.2 g for the optimized sodium hydroxide system.
| NaOH Dosage | Mass Loss after 1 h at 165 °C | Trend | Best Condition |
|---|---|---|---|
| Low | Lower | Insufficient activation | No |
| 0.2 g | 2.578 g | Maximum | Yes |
| Higher than 0.2 g | Decreased | Side reactions or equilibrium shift | No |
The effect of degradation time was straightforward. Longer time produced greater mass loss. At 20 min, the mass loss was 0.891 g; at 60 min, it was 2.578 g. The fastest change occurred between 30 and 40 min. For practical removal from an electric vehicle battery pack, I selected 40 min as a reasonable balance between degradation extent and energy consumption. A longer time can improve removal, but it also increases thermal exposure and cost. This balance is critical when the target is an electric vehicle battery pack that contains temperature-sensitive cells and polymers.
| Time | Mass Loss at 165 °C | Interpretation |
|---|---|---|
| 20 min | 0.891 g | Initial rapid swelling and chain scission |
| 30–40 min | Fast increase | Most efficient degradation window |
| 60 min | 2.578 g | High degradation but longer energy input |
To test general applicability, I applied the sodium hydroxide/diethylene glycol system to three different commercial polyurethane adhesives. The complete degradation times differed, but all samples could be degraded. Some adhesives degraded completely at lower temperatures than others. The differences were consistent with variations in crosslink density and chemical composition. This was encouraging because an electric vehicle battery pack may contain adhesives from different suppliers and different production batches. A robust removal method should not depend on one exact adhesive formulation.
| Adhesive Type | Temperature Where Degradation Accelerated | Complete Degradation Behavior | Conclusion |
|---|---|---|---|
| Polyurethane encapsulating AB adhesive | About 165 °C | Complete degradation | Compatible |
| Polyurethane PU adhesive | About 155 °C | Complete degradation | Compatible |
| Polyurethane resin AB adhesive | About 155 °C | Complete degradation | Compatible |
I analyzed the degradation products by infrared spectroscopy. The spectra showed broad absorption between 3200 and 3600 cm⁻¹, corresponding to O–H stretching. They also showed absorption between 2800 and 3000 cm⁻¹ from C–H stretching and strong absorption between 1000 and 1100 cm⁻¹ from C–O stretching. These features are consistent with hydroxyl-containing and ether-containing degradation products. Importantly, the peak positions remained similar across different degradation temperatures, which suggests that the main functional groups of the products did not change drastically with temperature. However, the intensities and the viscosity changed, indicating a decrease in molecular weight at higher temperature.
I used rheology to examine the degradation products. The products behaved as Newtonian fluids under the tested conditions because their viscosity did not change significantly with shear rate. As the degradation temperature increased, the product viscosity decreased. This is consistent with more extensive chain scission and a higher fraction of low-molecular-weight species. The viscosity trend supported the conclusion that higher temperature produces more complete degradation. At the same time, the high temperature required for complete degradation of the electric vehicle battery pack adhesive was a serious drawback. A typical electric vehicle battery pack cannot be heated to 165 °C or above without risking damage to cells, separators, seals, and electronic components. Therefore, I moved to a more efficient catalytic system based on ionic liquids.
| Characterization | Observation | Interpretation |
|---|---|---|
| FTIR O–H stretching | 3200–3600 cm⁻¹ | Hydroxyl-containing products |
| FTIR C–H stretching | 2800–3000 cm⁻¹ | Saturated hydrocarbon groups |
| FTIR C–O stretching | 1000–1100 cm⁻¹ | Alcohol, ether, or ester structures |
| Rheology | Newtonian behavior | Low-molecular-weight liquid products |
| Viscosity vs temperature | Viscosity decreased at higher temperature | More extensive chain scission |
Ionic liquids are attractive because they combine low vapor pressure, high thermal stability, tunable Lewis acidity, and good solubility for many organic and inorganic species. In my second experimental series, I synthesized several ionic liquids by reacting metal chlorides with imidazolium chlorides. The metal chlorides included iron(III) chloride, cobalt(II) chloride, and zinc(II) chloride. The imidazolium chlorides included allyl-methylimidazolium chloride, butyl-methylimidazolium chloride, and hexyl-methylimidazolium chloride. The synthesis was performed under reflux at 50 °C for 5 h, followed by vacuum drying at 60 °C for 24 h. The resulting ionic liquids included [Amim][FeCl₄], [Bmim][FeCl₄], [Hmim][FeCl₄], [Amim][ZnCl₃], and [Amim][CoCl₃].
The anion mole fraction in the ionic liquid was calculated from the molar amounts of metal chloride and imidazolium chloride. For example, for an iron-containing ionic liquid:
$$ x_{\mathrm{FeCl_3}} = \frac{n(\mathrm{FeCl_3})}{n(\mathrm{FeCl_3}) + n([\mathrm{Bmim}]\mathrm{Cl})} $$
I varied this mole fraction to study how the cation-to-anion balance affects catalytic activity. I also used electrospray ionization mass spectrometry to confirm the ionic liquid composition. In positive ion mode, strong peaks appeared at m/z 123.9 and 139.1, corresponding to the imidazolium cations. In negative ion mode, a peak appeared at m/z 197.8, corresponding to [FeCl₄]⁻. A fragment at m/z 162.3 was also observed, likely from [FeCl₃]⁻ formed by reduction or collision-induced dissociation. These results confirmed that the target ionic liquids were successfully synthesized.
| Ionic Liquid | Cation | Anion | Key ESI-MS Signal | Expected Lewis Acidity |
|---|---|---|---|---|
| [Amim][FeCl₄] | Allyl-methylimidazolium | [FeCl₄]⁻ | Positive m/z 123.9; negative m/z 197.8 | Strong |
| [Bmim][FeCl₄] | Butyl-methylimidazolium | [FeCl₄]⁻ | Positive m/z 139.1; negative m/z 197.8 | Strong |
| [Hmim][FeCl₄] | Hexyl-methylimidazolium | [FeCl₄]⁻ | Cation-related signals | Strong |
| [Amim][ZnCl₃] | Allyl-methylimidazolium | [ZnCl₃]⁻ | Cation-related signals | Moderate to weak |
| [Amim][CoCl₃] | Allyl-methylimidazolium | [CoCl₃]⁻ | Cation-related signals | Moderate |
I compared sodium hydroxide and ionic liquids as catalysts under otherwise identical conditions. Using diethylene glycol as the degradation agent at 170 °C for 40 min, sodium hydroxide gave a degradation rate of 26.7%. In contrast, [Bmim][FeCl₄] gave 67.4%, and [Amim][FeCl₄] gave 69.5%. This was a major improvement and showed that ionic liquids can significantly accelerate the degradation of polyurethane structural adhesive from an electric vehicle battery pack. The ionic liquid cations and anions act cooperatively. The cation interacts with the carbonyl oxygen of the carbamate group, increasing the positive character of the carbonyl carbon. The anion interacts with the hydroxyl hydrogen of diethylene glycol, increasing the electron density and nucleophilicity of the hydroxyl oxygen. This cooperative activation lowers the barrier for C–O bond cleavage.
| Catalyst | Degradation Rate at 170 °C for 40 min | Relative Performance |
|---|---|---|
| NaOH | 26.7% | Baseline |
| [Bmim][FeCl₄] | 67.4% | Much higher |
| [Amim][FeCl₄] | 69.5% | Best among tested iron ionic liquids |
I then studied the effect of the anion mole fraction in the ionic liquid. The degradation rate increased as the anion mole fraction increased from 0.4 to 0.5, reached a maximum at 0.5, and then decreased at 0.67 and 0.75. The best performance occurred when the cation and anion were present in a 1:1 ratio. This supports the idea that both ions must participate in the transition state. If the cation is in excess, there are too few anions to activate the alcohol. If the anion is in excess, there are too few cations to activate the carbonyl group. The balanced 1:1 composition gives the most efficient cooperative catalysis.
| Anion Mole Fraction | Degradation Rate | Proposed Condition |
|---|---|---|
| 0.40 | Lower | Cation excess, insufficient anion activation |
| 0.50 | 67.5% | Optimal cation/anion balance |
| 0.67 | Lower | Anion excess, insufficient cation activation |
| 0.75 | Lower | Strong anion excess |
The amount of ionic liquid also influenced the degradation rate. As the ionic liquid mass increased, the degradation rate first increased and then leveled off. More ionic liquid provides more catalytic sites, which increases the probability that polyurethane chains encounter an activated degradation agent. However, beyond a certain dosage, the system becomes saturated. Additional ionic liquid does not significantly increase the number of effective collisions because the available polyurethane surface or soluble polymer concentration becomes limiting. In my experiments, 2 g of ionic liquid was sufficient to reach the plateau. This dosage is practical for further development because it avoids unnecessary cost and waste.
| Ionic Liquid Mass | Degradation Rate Trend | Interpretation |
|---|---|---|
| Low | Increases with mass | More catalytic sites |
| 2 g | Approaches plateau | Practical optimum |
| High | No major further increase | Saturation and mass-transfer limitation |
I also compared degradation agents in the ionic liquid system. Diethylene glycol was clearly better than diethanolamine. At an anion mole fraction of 0.5, the degradation rate with diethylene glycol was about 27% higher than with diethanolamine. The reason is similar to that observed in the sodium hydroxide system. Diethylene glycol has two hydroxyl groups in similar chemical environments, and both can participate in transesterification. Diethanolamine contains an amine group that can undergo competing reactions and may promote crosslinking. Therefore, diethylene glycol remains the preferred degradation agent for polyurethane structural adhesive in an electric vehicle battery pack.
| Degradation Agent | Relative Degradation Rate with Ionic Liquid | Chemical Explanation |
|---|---|---|
| Diethylene glycol | Higher by about 27% | Efficient transesterification and swelling |
| Diethanolamine | Lower | Competing amine reactions and possible crosslinking |
Temperature remained a strong factor in the ionic liquid system. As temperature increased, the degradation rate increased. Higher temperature provides more energy for bond cleavage and increases the mobility of the polymer chains and degradation agent. However, because ionic liquids reduce the activation barrier, the required temperature can be lower than in the sodium hydroxide system. In my experiments, 170 °C for 40 min gave strong degradation. I also found that at a much lower temperature, the adhesive could be softened enough to be scraped from a metal plate. This was important because an electric vehicle battery pack may not tolerate full degradation temperature everywhere, but local softening and mechanical removal may be feasible in some dismantling scenarios.
| Temperature | Degradation Behavior with Ionic Liquid | Practical Meaning for Electric Vehicle Battery Pack |
|---|---|---|
| Lower temperature | Softening and partial degradation | Possible mechanical removal after swelling |
| Moderate temperature | Increasing degradation rate | Balance between speed and safety |
| 170 °C | High degradation in 40 min | Effective but still challenging for cells |
I then investigated how the cation structure affects catalytic performance. With the same anion [FeCl₄]⁻, the degradation efficiency followed the order [Amim][FeCl₄] > [Bmim][FeCl₄] > [Hmim][FeCl₄]. The allyl cation contains a carbon–carbon double bond, which creates an uneven charge distribution and a higher electron density in certain regions. This enhances the interaction between the cation and the carbonyl oxygen of the polyurethane. The butyl and hexyl cations are saturated and have more uniform charge distribution. In addition, the hexyl chain is longer than the butyl chain, so it creates greater steric hindrance. Steric hindrance reduces the frequency of effective collisions between the ionic liquid, the degradation agent, and the polyurethane chain. Therefore, the cation with the allyl group and the shortest alkyl chain gave the best performance among the tested cations.
| Ionic Liquid | Cation Structure Feature | Charge Distribution | Steric Hindrance | Degradation Order |
|---|---|---|---|---|
| [Amim][FeCl₄] | Allyl group with C=C | Uneven, polarized | Low | Best |
| [Bmim][FeCl₄] | Butyl chain | More uniform | Moderate | Intermediate |
| [Hmim][FeCl₄] | Hexyl chain | More uniform | High | Lowest |
I also studied the anion effect while keeping the cation [Amim]⁺ constant. The degradation efficiency followed the order [Amim][ZnCl₃] > [Amim][CoCl₃] > [Amim][FeCl₄]. This order is the reverse of the Lewis acidity order. In other words, the strongest Lewis acid did not produce the best catalytic result. The Lewis acidity order is approximately:
$$ \mathrm{ZnCl_3^- < CoCl_3^- < FeCl_4^-} $$
The zinc-containing anion has the weakest Lewis acidity, the cobalt-containing anion is intermediate, and the iron-containing anion is the strongest. A moderate or weaker Lewis acid can interact with the carbonyl oxygen without over-stabilizing the transition state. If the Lewis acid is too strong, it may bind the carbonyl oxygen too tightly and hinder the subsequent nucleophilic attack by diethylene glycol. Therefore, an ionic liquid with a relatively low Lewis acidity can provide better overall catalytic performance for polyurethane alcoholysis. This finding is useful for designing ionic liquids for removal of adhesive from an electric vehicle battery pack because it shows that catalytic activity is not simply a matter of acid strength.
| Ionic Liquid | Anion | Lewis Acidity | Degradation Performance | Proposed Reason |
|---|---|---|---|---|
| [Amim][ZnCl₃] | [ZnCl₃]⁻ | Weakest | Best | Balanced activation without over-stabilization |
| [Amim][CoCl₃] | [CoCl₃]⁻ | Intermediate | Intermediate | Moderate interaction |
| [Amim][FeCl₄] | [FeCl₄]⁻ | Strongest | Lowest among these three | Too strong binding may hinder attack |
I proposed a cooperative mechanism for ionic liquid-catalyzed alcoholysis. First, the cation interacts with the carbonyl oxygen of the carbamate group:
$$ \mathrm{PU-C(=O)-NH- + IL^+ \rightleftharpoons PU-C(=O^+\cdots IL)-NH-} $$
This interaction withdraws electron density from the carbonyl oxygen and increases the positive character of the carbonyl carbon. Second, the anion interacts with the hydroxyl hydrogen of diethylene glycol:
$$ \mathrm{HO-(CH_2CH_2O)_2-H + IL^- \rightleftharpoons IL\cdots H-O-(CH_2CH_2O)_2-H} $$
This interaction increases the electron density on the hydroxyl oxygen and makes it more nucleophilic. Third, the activated hydroxyl oxygen attacks the activated carbonyl carbon, forming a six-membered transition state. The C–O bond of the carbamate then breaks, and the ionic liquid is released unchanged. A simplified catalytic cycle is:
$$ \mathrm{PU + DGA \xrightarrow{IL} PU\ fragments + DGA\ derivatives + IL} $$
The ionic liquid acts as a catalyst rather than a reactant. This is consistent with the observation that the ionic liquid can be recovered and reused multiple times. It also explains why the cation and anion must be present in a balanced ratio: both are needed to activate the two reacting partners simultaneously.
I tested the ionic liquid system on a metal plate carrying cured polyurethane adhesive. Using diethylene glycol and [Amim][ZnCl₃] at 50 °C for 40 min, the adhesive softened significantly. After treatment, I could remove it with a hard plastic scraper. This was a low-temperature demonstration of adhesive removal relevant to an electric vehicle battery pack. It showed that ionic liquids can reduce the mechanical difficulty of dismantling even when full polymer degradation is not performed. The softened adhesive no longer held the metal plate strongly, and mechanical separation became possible without damaging the metal. This hybrid chemical–mechanical approach may be valuable for repairing or recycling an electric vehicle battery pack.
| Treatment | Condition | Result | Implication |
|---|---|---|---|
| Ionic liquid plus diethylene glycol | 50 °C, 40 min | Adhesive softened | Low-temperature removal by scraping |
| Mechanical scraping | Hard plastic tool | Adhesive removed from metal plate | Reduced damage and easier dismantling |
I studied the recycle behavior of the ionic liquid catalyst. The ionic liquid remained active for many cycles. However, after about 20 cycles, the mass loss of polyurethane began to decrease. The degradation efficiency continued to decline, and by about 25 cycles, the performance was poor. I attributed this to the accumulation of degradation products in the reaction mixture. These products dilute the ionic liquid, interfere with interactions between the ionic liquid and the polyurethane surface, and reduce the effective concentration of the active catalyst. When I added fresh diethylene glycol, the degradation performance recovered partially. This suggests that the ionic liquid itself was not completely decomposed but that the reaction medium became contaminated. For practical use in an electric vehicle battery pack recycling line, the degradation products would need to be separated or the solvent refreshed periodically.
| Cycle Number | Observed Performance | Interpretation | Action |
|---|---|---|---|
| 1–20 | Good degradation | Catalyst remains active | Continue use |
| After 20 | Mass loss decreases | Product accumulation and dilution | Refresh solvent or separate products |
| Around 25 | Poor performance | Catalytic medium severely contaminated | Recover or replace ionic liquid |
The overall comparison between sodium hydroxide and ionic liquids shows a clear advantage for ionic liquids in terms of degradation rate at the same temperature and time. Sodium hydroxide is inexpensive and simple, but it requires higher temperatures for complete degradation. Ionic liquids are more expensive and require synthesis, but they provide much higher degradation rates and can be reused. For an electric vehicle battery pack, the ideal process would combine a low-cost pre-swelling step, a highly active ionic liquid catalyst, and a mechanical separation step. This would reduce thermal exposure and improve material recovery.
| Feature | NaOH Catalyst | Ionic Liquid Catalyst |
|---|---|---|
| Cost | Low | Higher |
| Degradation rate at 170 °C, 40 min | 26.7% | 67.4–69.5% |
| Temperature for complete degradation | High | Lower than NaOH system |
| Reusability | Limited | Good up to about 20 cycles |
| Mechanism | Base activation of glycol | Cooperative cation/anion activation |
| Application to electric vehicle battery pack | Effective but thermally demanding | Promising for lower-temperature removal |
I also summarized the optimized conditions from my experiments. For the sodium hydroxide system, the best conditions were diethylene glycol as the degradation agent, 0.2 g NaOH, 165 °C, and about 40 min. For the ionic liquid system, the best conditions were diethylene glycol as the degradation agent, an anion mole fraction of 0.5, 2 g ionic liquid, 170 °C, and 40 min. The ionic liquid system gave a much higher degradation rate under comparable conditions. When [Amim][ZnCl₃] was used, the adhesive could also be softened at 50 °C and removed mechanically from a metal plate.
| Parameter | NaOH System | Ionic Liquid System |
|---|---|---|
| Degradation agent | Diethylene glycol | Diethylene glycol |
| Catalyst | NaOH, 0.2 g | Ionic liquid, 2 g |
| Anion mole fraction | Not applicable | 0.5 |
| Temperature | 165 °C | 170 °C |
| Time | 40 min | 40 min |
| Degradation rate | 26.7% under comparison condition | 67.4–69.5% |
| Low-temperature removal | Not demonstrated | Softening at 50 °C with [Amim][ZnCl₃] |
The degradation products from both systems contained hydroxyl, ether, and hydrocarbon groups according to infrared spectroscopy. The products were liquid-like and showed Newtonian behavior. The viscosity decreased as degradation temperature increased, indicating that higher temperature produced shorter chain fragments. In the ionic liquid system, the catalyst remained structurally intact during the reaction, which is consistent with a catalytic cycle rather than a stoichiometric reaction. This is important for reducing cost and environmental impact because the ionic liquid can potentially be recovered and reused.
From a practical perspective, removing polyurethane structural adhesive from an electric vehicle battery pack requires more than a high degradation rate. The process must be compatible with the battery materials, safe for workers, and scalable. High temperature can accelerate degradation but may damage cells. Strong acids or bases can corrode metals and electronics. Organic solvents can create fire and health hazards. Ionic liquids offer a tunable platform because their cations and anions can be selected to balance catalytic activity, viscosity, solubility, and safety. My results show that cation charge distribution and steric hindrance are important design parameters, and that anion Lewis acidity should be moderate rather than extreme. A cation with an allyl group and a relatively small size, combined with a weakly Lewis acidic anion such as [ZnCl₃]⁻, gave the best performance among the systems I tested.
I also considered the environmental impact. Polyurethane adhesives are crosslinked and thermoset-like, so they do not melt and flow like thermoplastics. Chemical alcoholysis converts them into smaller molecules that can be separated from metal, glass, and cell components. If the degradation products can be recovered as polyols or other intermediates, they may be reused in new polymer formulations. If they cannot be reused, they can at least be concentrated and treated more safely than the intact adhesive. The use of diethylene glycol as the main degradation agent is favorable because it is less volatile than many organic solvents and can participate directly in transesterification. The use of ionic liquids as catalysts can reduce the required temperature compared with simple base catalysis, which may reduce energy consumption and protect the electric vehicle battery pack components.
Nevertheless, several challenges remain. First, full degradation still required 170 °C in my ionic liquid experiments. This temperature may be acceptable for loose adhesive samples or separated metal plates, but it may not be acceptable for an assembled electric vehicle battery pack containing live or residual cells. Second, the ionic liquid cost is higher than that of sodium hydroxide. Third, product accumulation limits the reuse of the ionic liquid after about 20 cycles. Fourth, the degradation products are complex mixtures, and their separation and purification may be difficult. Fifth, the process must be tested on larger electric vehicle battery pack modules under realistic conditions, including mixed materials, residual electrolyte, and mechanical constraints.
For future work, I would focus on lowering the degradation temperature further while maintaining high adhesive removal efficiency. One approach is to combine ionic liquids with physical pretreatment such as controlled swelling, ultrasound, or microwave heating. Another approach is to design task-specific ionic liquids with even better cation–anion cooperativity. For example, a cation with strong carbonyl activation and an anion with moderate hydrogen-bond acceptance could lower the activation energy more effectively. A third approach is to integrate chemical degradation with mechanical separation so that complete depolymerization is not required. If the adhesive can be softened and debonded at 50–80 °C, the electric vehicle battery pack can be dismantled with much less thermal risk. My low-temperature metal plate test with [Amim][ZnCl₃] is an early demonstration of this hybrid strategy.
I also recommend that future studies quantify the kinetics of the ionic liquid system more precisely. The Arrhenius equation can be fitted to degradation rates at multiple temperatures:
$$ \ln k = \ln A – \frac{E_a}{RT} $$
By plotting \(\ln k\) against \(1/T\), the apparent activation energy can be obtained. This would allow a direct comparison between sodium hydroxide and ionic liquid catalysts. It would also help identify whether the ionic liquid changes the mechanism or simply increases the frequency of successful collisions. In addition, the diffusion of diethylene glycol into the adhesive network should be studied. Swelling is often the first step in degradation, and if swelling is slow, the overall process will be slow even if the intrinsic reaction is fast. A diffusion–reaction model could be written as:
$$ \frac{\partial C}{\partial t} = D \frac{\partial^2 C}{\partial x^2} – k C C_{PU} $$
where \(C\) is the concentration of the degradation agent, \(D\) is the diffusion coefficient, \(k\) is the reaction rate constant, and \(C_{PU}\) represents the reactive sites in the polyurethane. Such a model would help optimize the balance between solvent penetration and chemical bond cleavage in an electric vehicle battery pack adhesive layer.
Overall, I found that chemical degradation is a viable route for removing polyurethane structural adhesive from an electric vehicle battery pack. Sodium hydroxide with diethylene glycol can degrade several commercial adhesives, but it requires relatively high temperature. Ionic liquids, especially [Amim][FeCl₄] and [Amim][ZnCl₃], greatly improve the degradation rate and provide a tunable catalytic platform. The cation and anion work together to activate the carbonyl group and the alcohol, respectively, forming a transition state that promotes carbamate cleavage. The best cation has uneven charge distribution and low steric hindrance, while the best anion has moderate or low Lewis acidity. The ionic liquid can be reused for many cycles, although product accumulation eventually reduces performance. Low-temperature softening and mechanical removal are also possible, which may be the most practical route for an assembled electric vehicle battery pack.
In conclusion, my study contributes a systematic comparison of base-catalyzed and ionic-liquid-catalyzed alcoholysis for polyurethane structural adhesives. It shows how temperature, time, catalyst dosage, degradation agent, cation structure, anion structure, and anion mole fraction affect degradation. It also provides a mechanistic explanation based on cooperative cation–anion activation. For the electric vehicle battery pack recycling industry, these findings support the development of safer, more efficient, and more sustainable adhesive removal processes. The transition from high-temperature sodium hydroxide degradation to tunable ionic liquid catalysis is a meaningful step toward practical dismantling, second-life reuse, and material recovery. With further optimization, ionic liquid-assisted alcoholysis could become a key enabling technology for the circular economy of electric vehicle battery pack systems.
| Research Question | Key Finding | Implication for Electric Vehicle Battery Pack Recycling |
|---|---|---|
| Can NaOH catalyze adhesive degradation? | Yes, with diethylene glycol at 165 °C and 0.2 g NaOH | Useful for bulk adhesive removal but thermally demanding |
| Does the method apply to different adhesives? | Yes, three commercial adhesives were degraded | Potential broad applicability |
| Do ionic liquids improve degradation? | Yes, degradation rate increased from 26.7% to 67.4–69.5% | Higher efficiency at comparable conditions |
| What is the best ionic liquid composition? | 1:1 cation/anion ratio, 2 g dosage | Design rule for catalytic adhesive removal |
| Which cation performs best? | Allyl-containing cation with low steric hindrance | Improves carbonyl activation and collision frequency |
| Which anion performs best? | Weak to moderate Lewis acid, such as [ZnCl₃]⁻ | Avoids over-stabilization of the transition state |
| Can the ionic liquid be reused? | Yes, for about 20 cycles before product accumulation | Supports a circular catalytic process |
| Can removal occur at low temperature? | Softening at 50 °C and scraping from metal plate | Hybrid chemical–mechanical dismantling is promising |
My work leaves several open questions that are important for scaling. How will residual electrolyte, metal ions, and cell debris affect ionic liquid performance in a real electric vehicle battery pack? Can the ionic liquid be separated from the degradation products by filtration, extraction, or membrane processes? What is the energy balance of heating an entire electric vehicle battery pack module compared with local heating of adhesive joints? How can the process be automated for safe industrial dismantling? Answering these questions will require collaboration among polymer chemists, battery engineers, process designers, and recycling operators. The electric vehicle battery pack is a complex product, and its end-of-life treatment cannot be solved by a single chemical reaction alone. It requires an integrated strategy that combines material design, adhesive formulation, dismantling technology, and recycling infrastructure.
I believe the most promising near-term path is to use ionic liquids as low-temperature swelling and softening agents rather than relying on complete alcoholysis of the entire adhesive layer. If the adhesive can be weakened enough for mechanical separation, the electric vehicle battery pack can be opened with less energy and less risk. In parallel, high-temperature alcoholysis can be used for separated metal plates or adhesive-rich waste streams where cells are no longer present. This two-stage approach would separate sensitive components first and then treat the adhesive residue under more aggressive conditions. Such a strategy aligns with the principles of selective dismantling and material-specific recycling.
From a materials design perspective, the results also suggest that adhesive manufacturers could consider cleavable linkages or reversible crosslinks in future electric vehicle battery pack adhesives. If the adhesive contains bonds that respond to a specific chemical trigger, dismantling could become much easier. However, this must be balanced against the need for long-term durability, crash safety, and thermal stability. The trade-off between adhesive performance and recyclability is a central challenge. My study on alcoholysis and ionic liquid catalysis provides data that can inform this trade-off. It shows that polyurethane networks can be broken down efficiently when the right catalyst and degradation agent are combined, but it also shows that high temperature remains a limiting factor for complete degradation.
In summary, I have developed and compared two chemical routes for removing polyurethane structural adhesive from an electric vehicle battery pack. The base-catalyzed route is simple and broadly effective but requires high temperature. The ionic liquid route is more efficient and tunable and can operate through cooperative cation–anion catalysis. The best ionic liquid systems combine an allyl-functionalized imidazolium cation with an anion of moderate or weak Lewis acidity. The degradation rate at 170 °C for 40 min reached about 69.5%, which is much higher than the 26.7% obtained with sodium hydroxide under comparable conditions. The ionic liquid can be reused for about 20 cycles, and low-temperature softening enables mechanical removal from metal surfaces. These findings provide a foundation for safer and more sustainable dismantling of electric vehicle battery pack systems, and they point toward future work on process integration, catalyst recovery, and low-temperature debonding.