In the context of global energy transition and environmental protection, the rapid development of new energy vehicles has brought the recycling of retired batteries into sharp focus. The growing market penetration of electric vehicles inevitably generates a huge number of end-of-life high-voltage battery packs. These packs still retain 70–80% of their initial capacity after five to six years of service, making their disassembly and subsequent cell recovery or repurposing particularly valuable. However, a major obstacle in this process is the removal of the polyurethane structural adhesive used in battery module assembly. This adhesive ensures mechanical integrity, vibration resistance, thermal conduction, and electrical insulation of the battery cells. When the battery reaches its end of life, the polyurethane adhesive must be removed efficiently without damaging the battery cells. Conventional methods such as incineration, mechanical scraping, or chemical dissolution have serious limitations in terms of safety, environmental impact, or efficiency. Therefore, developing a mild, green, and effective method to degrade polyurethane structural adhesive is of utmost importance for the sustainable recycling of high-voltage batteries.
My research focused on the chemical degradation of polyurethane structural adhesives under comparatively mild conditions. I first investigated the conventional alkali-catalyzed alcoholysis using sodium hydroxide as the catalyst and diethylene glycol as the cleavage agent. Then, to lower the degradation temperature and improve the efficiency, I synthesized a series of ionic liquids and used them as catalysts in the alcoholysis of polyurethane. Throughout this work, I systematically examined the influence of reaction parameters such as temperature, time, catalyst dosage, the type of degradation agent, and the structure of the ionic liquid catalyst. The ultimate purpose was to propose a practical strategy for removing polyurethane structural adhesive from high-voltage battery modules, thus facilitating efficient and safe battery recycling.
This article presents my complete investigation, starting from the background and motivation, followed by experimental details and characterization, and ending with the most significant findings and conclusions. In particular, I would like to emphasize that the introduction of ionic liquids not only enhanced the degradation efficiency but also shifted the required temperature to a more acceptable level for battery applications. Moreover, the ionic liquids demonstrated remarkable reusability, which makes the process more cost-effective. I believe that this work provides a solid foundation for the industrial-scale removal of structural adhesives from high-voltage battery systems.
Background and Significance
Polyurethane is a versatile polymer synthesized from polyols and polyisocyanates. Its outstanding bonding strength, toughness, insulation properties, and resistance to vibration make it the dominant adhesive material in high-voltage battery assemblies. During the assembly process, polyurethane structural adhesives are applied on metal plates to bond individual battery cells into a module. The adhesive guarantees that the cells remain stable under mechanical stress, thermal cycling, and vibration, thus directly contributing to the safety of the battery system. Furthermore, the adhesive serves as a sealing layer protecting cells from moisture and dust. Given the increasing production and adoption of electric vehicles, the annual consumption of polyurethane structural adhesives in high-voltage battery manufacturing has grown considerably. Nevertheless, when a high-voltage battery pack is retired and needs to be dismantled for recycling, the same adhesive that was so valuable during the assembly becomes a nuisance. Efficiently removing the adhesive without harming the battery cells is a critical task to enable reuse of the remaining capacity and recovery of valuable metals such as nickel, cobalt, and lithium.
Among the known techniques for removing crosslinked polyurethane, chemical recycling routes are attractive because they can selectively cleave the carbamate linkages under controlled conditions. Degradation methods such as hydrolysis, aminolysis, alcoholysis, and glycolysis have been extensively reported. Alcoholysis, in particular, is regarded as a mature and effective approach, using small-molecular diols or polyols to break the urethane bonds and regenerate polyol-like products. In many cases, a catalyst is necessary to speed up the reaction and lower the activation energy. Traditional catalysts are alkali hydroxides, metal salts, or organic bases. However, with the demand for lower operating temperatures in high-voltage battery components, more selective and stable catalysts are needed.
In recent years, ionic liquids have attracted considerable interest as eco-friendly solvents and catalysts because of their negligible vapor pressure, high thermal stability, tunable structures, and excellent solubility for many organic compounds. Ionic liquids have been successfully employed in the degradation of polyesters such as polycarbonate, poly(ethylene terephthalate), and poly(lactic acid). Their catalytic activity often originates from the coordination between the cation or anion of the ionic liquid and the carbonyl oxygen of the ester or carbamate group, facilitating nucleophilic attack and bond cleavage. Encouraged by these studies, I decided to apply ionic liquids to the alcoholysis of polyurethane structural adhesives used in high-voltage batteries.
Experimental Section
Materials and Reagents
The experiments were carried out with commercially available polyurethane structural adhesives from different brands. I obtained three types of polyurethane AB adhesives and two types of filled one-component adhesives to cover a wide range of compositions. The degradation agents included diethylene glycol and diethanolamine. Sodium hydroxide was used as a conventional catalyst, while several ionic liquids were synthesized from 1-butyl-3-methylimidazolium chloride, 1-allyl-3-methylimidazolium chloride, 1-hexyl-3-methylimidazolium chloride, and anhydrous metal chlorides such as iron(III) chloride, cobalt(II) chloride, and zinc(II) chloride. All chemicals were of analytical grade and used without further purification.
Preparation of Ionic Liquids
I prepared ionic liquids with different anions by mixing the corresponding imidazolium chloride salt and the anhydrous metal chloride in a defined molar ratio. The mixture was heated to 50 °C and stirred for 5 hours to ensure complete reaction. After that, the resulting liquid was dried under vacuum at 60 °C for 24 hours to remove residual water. In this way, ionic liquids with different molar fractions of the anionic species were obtained. For example, the molar fraction \(x\) of tetrachloroferrate(III) in the ionic liquid was defined as:
$$
x = \frac{n(\mathrm{FeCl}_3)}{n(\mathrm{FeCl}_3) + n([\mathrm{Bmim}]\mathrm{Cl})}
$$
where \(n(\mathrm{FeCl}_3)\) is the amount of anhydrous iron chloride and \(n([\mathrm{Bmim}]\mathrm{Cl})\) is the amount of 1-butyl-3-methylimidazolium chloride. According to this definition, when the molar ratio of ionic liquid precursor to metal chloride is 1:1, the anion molar fraction is 0.5. I also synthesized ionic liquids containing zinc and cobalt to compare the effect of the Lewis acidity of the anion.
Alcoholysis Procedure
The degradation reactions were conducted in a three-necked flask equipped with a reflux condenser and a magnetic stirrer. In a typical run, a piece of polyurethane adhesive was weighed and placed into the flask. Then a known volume of degradation agent and a specified amount of catalyst were added. The mixture was heated to the desired temperature under continuous stirring. After the reaction time had elapsed, the flask was cooled to room temperature. The remaining solid polyurethane, if any, was recovered, rinsed with ethanol, dried, and weighed. The mass loss was calculated as \(m_1 – m_2\), where \(m_1\) is the initial mass and \(m_2\) is the remaining mass. For a clearer comparison, the degradation rate was defined as:
$$
\text{Degradation rate} = \frac{m_1 – m_2}{m_1} \times 100\%.
$$
Each experiment was repeated at least twice to verify reproducibility. I also performed swelling experiments to evaluate the crosslink density of different adhesives. In the swelling test, small cuboid specimens were immersed in water, diethylene glycol, or diethanolamine at 30 °C. After seven days, the samples reached swelling equilibrium and were weighed after removing the surface liquid. The swelling ratio was calculated as:
$$
\text{Swelling ratio} = \frac{m_{\mathrm{eq}} – m_0}{m_0} \times 100\%,
$$
where \(m_0\) is the initial mass and \(m_{\mathrm{eq}}\) is the mass at equilibrium.
Characterization
The synthesized ionic liquids were characterized by electrospray ionization mass spectrometry in both positive and negative ion modes. The degradation products were analyzed by Fourier-transform infrared spectroscopy and viscosity measurements. The viscosity was measured as a function of shear rate using a rotational rheometer at a constant temperature. These analyses helped me understand the chemical structure of the degradation products and the catalytic role of the ionic liquids.
Sodium Hydroxide Catalyzed Degradation of Polyurethane
Effect of Degradation Temperature
In the first set of experiments, I used diethanolamine as the degradation agent and sodium hydroxide as the catalyst to evaluate the influence of temperature on the complete degradation of two adhesives named black and white. Figure and data indicated that the complete degradation time decreased remarkably with increasing temperature. At 125 °C the black adhesive took 7.43 hours to completely disappear, while the white adhesive took 8.1 hours. When the temperature was increased to 245 °C, the time shortened to only 10 minutes for the black and 5 minutes for the white adhesive. This strong temperature dependence is consistent with the Arrhenius behavior, since an increase in temperature raises the fraction of activated molecules and thereby enhances the frequency of effective collisions between the adhesive and the degradation agent. The rate-limiting step in alcoholysis involves the nucleophilic attack on the carbamate carbon; higher temperatures accelerate this attack.
Interestingly, the white adhesive always required a longer time for complete degradation than the black adhesive under identical conditions. This observation led me to investigate the swelling behavior of both adhesives in different solvents. The results are summarized in the table below.
| Adhesive | Solvent | Initial mass (g) | Equilibrium mass (g) | Swelling ratio (%) |
|---|---|---|---|---|
| Black | Diethylene glycol | 0.298 | 0.305 | 2.30 |
| Black | Diethanolamine | 0.301 | 0.306 | 1.70 |
| Black | Water | 0.308 | 0.312 | 1.30 |
| White | Diethylene glycol | 0.469 | 0.475 | 1.30 |
| White | Diethanolamine | 0.536 | 0.542 | 1.10 |
| White | Water | 0.525 | 0.530 | 0.95 |
It is evident that diethylene glycol produced a higher swelling ratio than diethanolamine or water for both adhesives. The black adhesive swelled more than the white adhesive, which implies that the white adhesive has a higher crosslink density and therefore a more constrained network. As a result, solvent penetration and subsequent chain cleavage are more difficult in the white adhesive, explaining its longer degradation time. These findings highlight the importance of evaluating the actual adhesive composition when designing a removal process for high-voltage battery disassembly.
Selection of Degradation Agent
To improve the process, I compared the effectiveness of diethylene glycol, diethanolamine, and their mixtures. Using 0.2 g of sodium hydroxide at 165 °C for one hour, I measured the mass loss of the white adhesive. The results are plotted in terms of the diethylene glycol content in the mixed degradation agent. When pure diethanolamine was used, the mass difference was only 1.86 g, whereas pure diethylene glycol led to a larger mass difference of 2.58 g. Mixed solvents showed an interesting trend: as the proportion of diethylene glycol increased, the mass loss first decreased and then increased. This behavior suggests that the degradation mechanism changes with the composition. Diethanolamine contains two primary amine groups which are strong nucleophiles; however, its amino group can also react with the isocyanate groups to form urea linkages, leading to possible crosslinking and therefore a reduced degradation efficiency. In contrast, diethylene glycol has two primary hydroxyl groups with similar reactivity, and its ether oxygen enhances the electronegativity of the hydroxyl oxygen, making it more effective in attacking the carbamate carbonyl. Consequently, diethylene glycol was selected as the degradation agent for subsequent experiments.
Optimization of Sodium Hydroxide Alcoholysis
The influence of temperature was further evaluated using diethylene glycol as the degradation agent and a fixed amount of sodium hydroxide. The reaction time was one hour. At 135 °C the mass loss was only 0.45 g, while at 175 °C it reached 2.60 g. Because the difference between 165 °C and 175 °C was not significant, I chose 165 °C as the preferred temperature, balancing high degradation efficiency and lower energy consumption.
Next, I varied the amount of sodium hydroxide from 0.05 g to 0.35 g at 165 °C. The results are listed in Table 2. The mass loss increased with catalyst addition up to 0.2 g and then decreased. The optimum sodium hydroxide amount was therefore 0.2 g. Too much sodium hydroxide might shift the equilibrium and reduce the rate of alcoholysis, possibly due to the increased ionic strength and the salting-out effect of the degradation agent.
| NaOH amount (g) | Mass loss (g) |
|---|---|
| 0.05 | 1.48 |
| 0.10 | 1.92 |
| 0.15 | 2.31 |
| 0.20 | 2.58 |
| 0.25 | 2.36 |
| 0.30 | 2.02 |
| 0.35 | 1.65 |
The study of degradation time showed that the mass loss increased nearly linearly with time in the initial stage. At 20 minutes, the mass loss was 0.89 g, and at 60 minutes it reached 2.58 g. The maximum reaction rate was observed between 30 and 40 minutes. Therefore, I set 40 minutes as an appropriate degradation time for the subsequent ionic liquid studies.
Universality of the Sodium Hydroxide System
To test whether the selected recipe could handle different commercial adhesives, I prepared samples from three different polyurethane AB adhesives obtained from different manufacturers. Each adhesive was mixed in a 1:1 weight ratio of component A and component B, degassed, and allowed to cure for 24 hours. The cured adhesives were then cut into similar pieces and subjected to diethylene glycol with sodium hydroxide at different temperatures. Complete degradation times were recorded. The results are presented in Table 3.
| Temperature (°C) | PU potting AB (min) | PU general AB (min) | PU resin AB (min) |
|---|---|---|---|
| 125 | 245 | 205 | 270 |
| 135 | 180 | 135 | 170 |
| 145 | 115 | 70 | 95 |
| 155 | 70 | 35 | 50 |
| 165 | 30 | 18 | 25 |
All three adhesives were successfully degraded with the same formula, though their degradation times differed because of variations in chemical composition and crosslink density. The PU general AB adhesive degraded faster than the others at every temperature, indicating a lower crosslink density or a more accessible carbamate linkage. Since the high-voltage battery structural adhesives are often based on a similar polyether-polyurethane chemistry, this universal behavior demonstrates that the diethylene glycol and sodium hydroxide system is a reasonable baseline method for many commercial products.
Product Analysis
I analyzed the degradation products obtained at different temperatures using Fourier-transform infrared spectroscopy. The infrared spectra showed a broad absorption band around 3200–3600 cm⁻¹, which corresponds to the O–H stretching vibration of alcohols and probably also residual moisture. The bands at 2800–3000 cm⁻¹ were assigned to C–H stretching vibrations of methylene and methyl groups. A strong absorption around 1000–1100 cm⁻¹ was attributed to C–O stretching vibrations, indicating that the degradation products were mainly hydroxyl-terminated oligomers. Importantly, the positions of the absorption peaks did not change with temperature, meaning that the chemical nature of the products was the same at all temperatures. However, the viscosity of the degradation products decreased as the degradation temperature increased. This is a clear indication that the molecular weight of the products is lower when the degradation is carried out at a higher temperature, which is consistent with deeper chain cleavage.
Ionic Liquid Catalyzed Degradation of Polyurethane
The sodium hydroxide system required a relatively high temperature for complete degradation, which is problematic for high-voltage battery modules that cannot be heated above their thermal safety limit. To solve this problem, I synthesized ionic liquids containing iron, zinc, and cobalt anions and tested their catalytic performance. Ionic liquids can simultaneously activate both the nucleophile and the electrophile through their anion and cation, making them excellent catalysts for the alcoholysis of carbamates and esters.
Characterization of Ionic Liquids
Figure with the electrospray ionization mass spectra confirmed the successful formation of the desired ionic liquids. In the positive ion mode, signals at mass-to-charge ratios 124 and 139 appeared for [Amim]⁺ and [Bmim]⁺, respectively. In the negative ion mode, a strong peak at m/z 197.8 was attributed to the [FeCl₄]⁻ anion. Another peak at m/z 162.3 could arise either from the fragmentation of [FeCl₄]⁻ or from the reduction product [FeCl₃]⁻ in the presence of diethylene glycol. These data confirmed that the synthesized compounds were indeed ionic liquids with the expected cations and anions.
Comparison of Catalysts
I first evaluated the degradation performance of NaOH and two iron-containing ionic liquids under the same conditions: 40 mL of diethylene glycol, 0.2 g or 2 g of catalyst (for comparison, the ionic liquid mass was kept at 2 g), 170 °C, and 40 minutes. The degradation rate with sodium hydroxide was only 26.7%, whereas [Bmim][FeCl₄] and [Amim][FeCl₄] achieved degradation rates of 67.4% and 69.5%, respectively. The significantly higher degradation rate achieved by the ionic liquids indicates that the ion-pair synergistic catalysis is more efficient than the simple base catalysis. The mechanism is illustrated schematically by the following equation:
$$
\ce{R1-NH-CO-O-R2 + HO-R3-OH ->[\text{ionic liquid}] R1-NH2 + HO-R2 + HO-R3-OH}
$$
Actually, the precise mechanism involves the cation of the ionic liquid coordinating to the carbonyl oxygen, thus increasing the partial positive charge on the carbonyl carbon. Simultaneously, the anion of the ionic liquid interacts with the hydroxyl hydrogen of the degradation agent, increasing the electron density on the hydroxyl oxygen and enhancing its nucleophilicity. The two partners join together to form a six-membered transition state, which subsequently breaks to give the cleavage products.
Effect of Anion Molar Fraction
Using [Bmim][FeCl₄] as the catalyst, I prepared samples with different molar fractions of [FeCl₄]⁻, namely 0.4, 0.5, 0.67, and 0.75. The degradation rate of polyurethane was measured at 170 °C for 40 minutes. The results are illustrated in Table 4.
| [FeCl₄]⁻ molar fraction | Degradation rate (%) |
|---|---|
| 0.40 | 51.2 |
| 0.50 | 67.4 |
| 0.67 | 58.3 |
| 0.75 | 49.7 |
The optimal degradation rate occurred at a molar fraction of 0.5, which corresponds to a stoichiometric cation-to-anion ratio of 1:1 in the ionic liquid. This outcome strongly supports the proposed mechanism in which one cation and one anion are needed to form the cyclic transition state. The rates decrease when the ratio deviates from unity, because an excess of either cation or anion may break the balanced interaction.
Optimization of Ionic Liquid Concentration
I also investigated the influence of ionic liquid dosage on the degradation rate. The quantity of [Amim][FeCl₄] was varied from 0.5 g to 4 g, while maintaining the degradation agent volume at 40 mL and the temperature at 170 °C. The degradation rate initially increased sharply with the amount of catalyst, reflecting the larger number of active ion pairs available to coordinate with the polyurethane. When the ionic liquid amount exceeded 2 g, the degradation rate leveled off, suggesting that all readily accessible urethane groups had already been activated, and the excess ionic liquid did not further enhance the reaction. Therefore, I chose 2 g as the standard catalyst amount for subsequent experiments.
Influence of Degradation Agent
I compared diethylene glycol and diethanolamine using the same amount of ionic liquid as the catalyst. Diethylene glycol offered a much higher degradation rate than diethanolamine. For example, with [Amim][FeCl₄] at 170 °C, the degradation rates were 69.5% and 42% for diethylene glycol and diethanolamine, respectively. The higher efficacy of diethylene glycol is again explained by its oxygen-rich structure and the reduced tendency to form secondary crosslinks compared to the amine-containing agent. The presence of the amine group in diethanolamine can lead to the formation of urea bonds, which are even more stable than the original carbamate bonds and thus prevent further degradation. In contrast, diethylene glycol simply replaces the alcohol segment of the polyurethane without producing side crosslinks.
Minimum Degradation Temperature
To further reduce the thermal impact on high-voltage battery modules, it is important to know the lowest possible temperature at which the ionic liquid still catalyzes degradation. I performed experiments at temperatures ranging from 90 °C to 180 °C. At 90 °C, the degradation rate was only about 8% with [Bmim][FeCl₄], whereas at 120 °C it increased to about 20%. At 150 °C the rate reached roughly 45%, and at 180 °C it was close to 75%. Although the ionic liquid catalysts clearly improve the low-temperature performance compared to sodium hydroxide, their activity at very low temperatures remains limited. A more practical approach is to use the ionic liquid as a softening agent at temperatures around 50 °C, as will be described later in the metal plate experiment.
Influence of the Cation
To understand the role of the cation, I synthesized [Amim][FeCl₄], [Bmim][FeCl₄], and [Hmim][FeCl₄] while keeping the same anion [FeCl₄]⁻. The degradation rates were measured at four different temperatures (150, 160, 170, and 180 °C). The results are presented in Table 5.
| Ionic liquid | 150 °C | 160 °C | 170 °C | 180 °C |
|---|---|---|---|---|
| [Amim][FeCl₄] | 42.5 | 56.8 | 69.5 | 79.1 |
| [Bmim][FeCl₄] | 38.2 | 51.4 | 67.4 | 76.8 |
| [Hmim][FeCl₄] | 33.7 | 46.9 | 61.0 | 72.3 |
It is clear that [Amim]⁺, which contains an allyl group with a carbon–carbon double bond, provides the highest catalytic activity. The double bond has a high electron density and creates an uneven charge distribution on the cation, strengthening the interaction between the cation and the carbonyl oxygen of the polyurethane. [Bmim]⁺ with a butyl side chain performs slightly worse than [Amim]⁺, and [Hmim]⁺ with a hexyl chain performs the worst. This trend is due to the increasing steric hindrance in the longer alkyl chain, which shields the positive charge and impedes the access of the carbonyl group to the active site. Consequently, for a fixed anion, the best cation is the one with an unsaturated electron-rich substituent and the smallest steric hindrance.
Influence of the Anion
I further fixed the cation as [Amim]⁺ and synthesized [Amim][ZnCl₃], [Amim][CoCl₃], and [Amim][FeCl₄] to investigate the effect of the anionic Lewis acidity. The degradation rates at four different temperatures are reported in Table 6.
| Ionic liquid | 150 °C | 160 °C | 170 °C | 180 °C |
|---|---|---|---|---|
| [Amim][ZnCl₃] | 47.5 | 61.3 | 73.8 | 84.9 |
| [Amim][CoCl₃] | 44.1 | 57.7 | 70.4 | 80.2 |
| [Amim][FeCl₄] | 42.5 | 56.8 | 69.5 | 79.1 |
The catalytic activity follows the order [ZnCl₃]⁻ > [CoCl₃]⁻ > [FeCl₄]⁻ under all tested temperatures. Interestingly, this order is the reverse of the Lewis acidity strength of the anions. The iron-containing anion [FeCl₄]⁻ has the strongest Lewis acidity because Fe³⁺ has a high oxidation state and a strong electron-accepting tendency. The cobalt anion has intermediate Lewis acidity, while the zinc anion has the weakest Lewis acidity. In the degradation mechanism, the anion should interact with the hydrogen atom of the hydroxyl group of the degradation agent to increase the nucleophilicity of the oxygen. An extremely strong Lewis acid may complex too tightly with the oxygen-containing species, preventing the timely transfer of electrons and therefore slowing down the reaction. The moderate coordination ability of [ZnCl₃]⁻ strikes the optimal balance between activating the hydroxyl group and releasing the degradation products. Thus, in the design of an ionic liquid for polyurethane alcoholysis, the anion should be carefully chosen so that its Lewis acidity remains moderate.
Degradation of a Polyurethane Layer on a Metal Plate
To simulate a more realistic scenario encountered in high-voltage battery disassembly, I applied a polyurethane structural adhesive layer onto a metal plate and then immersed the coated plate in a mixture of diethylene glycol and an ionic liquid at only 50 °C for 40 minutes. After this treatment, the adhesive layer was noticeably softened, and it could be readily scraped off with a hard plastic blade. The before-and-after appearances are represented in the results, which demonstrate that the ionic liquid mixture can effectively loosen the polyurethane layer even at low temperatures without damaging the metal substrate. This procedure offers a gentle and practical route to removing structural adhesives from high-voltage battery modules that are sensitive to heat.
Reusability of the Ionic Liquid Catalyst
One of the greatest advantages of ionic liquids is their recyclability. I therefore tested the repeated use of the same ionic liquid and degradation solution by adding a fresh piece of polyurethane after each run. The mass loss of the polyurethane was measured for 25 consecutive cycles. The data showed that the mass loss remained almost constant during the first 20 cycles, indicating that the ionic liquid retained its catalytic activity. A slight decrease was observed after the 20th cycle, which I attribute to the accumulation of degradation products in the reaction mixture. These products may dilute the catalytic system, increase the viscosity, and obstruct the interaction between the ionic liquid and the polyurethane. When the reaction solution became saturated with degradation products, the ionic liquid began to lose its apparent activity. This behavior was confirmed by the fact that adding fresh diethylene glycol to the reaction mixture partially restored the degradation efficiency, since the accumulated products were effectively diluted. After about 25 cycles, the catalyst activity declined significantly despite replenishment, signaling the need for purification or catalyst recovery.
The catalytic cycle of the ionic liquid can be understood through the following simplified mechanism. Let the ionic liquid be represented by C⁺A⁻. In the alcoholysis of a carbamate bond with a diol, the cation C⁺ coordinates to the carbonyl oxygen:
$$
\ce{C=O + C+ -> C-O+}
$$
The anion A⁻ forms a hydrogen bond with a hydroxyl group of the diol, increasing the negative charge on the alcohol oxygen. Then the alcohol oxygen attacks the activated carbonyl carbon, leading to a tetrahedral intermediate. The subsequent proton transfer and bond cleavage regenerate the ionic liquid and release two smaller fragments containing hydroxyl and amine end groups. Because the ionic liquid appears unchanged at the end of the cycle, it is truly a catalyst in this process.
Comparison between NaOH and Ionic Liquid Systems
Throughout my research, I observed two systems for the removal of polyurethane adhesive from high-voltage battery components. The NaOH system is simple and cost-effective, but it requires temperatures above 165 °C for a practical degradation rate. Such high temperatures are not compatible with high-voltage battery modules because they could permanently damage the battery cells, separators, or current collectors. In contrast, the ionic liquid system offers a higher degradation rate at a lower temperature and can even be operated at 50 °C for softening the adhesive, making it much more suitable for the sensitive components of a high-voltage battery. The reusability of ionic liquids further reduces the amount of chemical waste and operational cost.
Table 7 provides a direct comparison between the optimum conditions of both catalytic systems.
| Parameter | NaOH system | Ionic liquid system |
|---|---|---|
| Degradation agent | Diethylene glycol | Diethylene glycol |
| Catalyst amount | 0.2 g NaOH per 60 mL | 2 g ionic liquid per 40 mL |
| Reaction temperature | 165 °C (complete degradation) | 170 °C (degradation), 50 °C (softening) |
| Reaction time | 40 min (mass loss test) | 40 min |
| Degradation rate | 26.7% (170 °C, 40 min) | 67.4–73.8% (170 °C, 40 min) |
| Minimum temperature for complete degradation | 125 °C | around 100 °C (lower if softening only) |
| Reusability | Not applicable | Up to 20 cycles |
| Compatibility with high-voltage battery components | Limited due to high temperature | Excellent, especially when used as softening agent |
The mass loss formula used in both systems is:
$$
\Delta m = m_1 – m_2.
$$
The degradation rate can also be expressed on a percentage basis:
$$
R_d = \frac{m_1 – m_2}{m_1} \times 100\%.
$$
When the reaction temperature is not sufficiently high for complete dissolution, the remaining polyurethane often appears swollen and softened. This softened state is beneficial because the adhesive can then be mechanically peeled from the metal substrate. This observation is particularly valuable for the low-temperature treatment of high-voltage battery packs that cannot tolerate full dissolution temperatures.
Conclusion and Future Perspectives
In this research, I have systematically investigated the chemical degradation of polyurethane structural adhesives that are typically used in high-voltage battery modules. The main conclusions can be summarized as follows.
First, diethylene glycol is a highly effective degradation agent for polyurethane adhesives, outperforming diethanolamine because it promotes carbamate cleavage without causing unintended crosslinking. Second, sodium hydroxide can catalyze the degradation effectively, but it generally requires a temperature of at least 125 °C for complete degradation and around 165 °C to obtain a reasonably fast reaction. This high-temperature requirement will damage heat-sensitive components in high-voltage batteries, so safer catalytic alternatives must be considered. Third, I successfully synthesized a series of ionic liquids such as [Bmim][FeCl₄], [Amim][FeCl₄], [Hmim][FeCl₄], [Amim][ZnCl₃], and [Amim][CoCl₃]. The electrospray mass spectrometry confirmed their expected structures. These ionic liquids significantly enhanced the degradation rate of polyurethane in diethylene glycol. The degradation rate improved from 26.7% with sodium hydroxide to around 67–74% with selected ionic liquids under identical conditions of 170 °C and 40 minutes. Fourth, the activity of the ionic liquid depends strongly on its structural components. The optimal molar fraction of the anionic species is 0.5, corresponding to a 1:1 cation-to-anion ratio. Cations with an unsaturated substituent such as [Amim]⁺ and a smaller steric bulk are better than saturated, longer-chain cations. Anions with relatively weak Lewis acidity, such as [ZnCl₃]⁻, are more effective than the strongly acidic [FeCl₄]⁻. These insights enable the design of even more efficient catalysts for polyurethane alcoholysis. Fifth, the ionic liquid system can be used to soften the structural adhesive at a very mild temperature of 50 °C, allowing easy removal from a metal plate. This is a critical step toward industrial application in the disassembly of high-voltage batteries. Finally, the ionic liquid is a genuine catalyst, since it can be recycled more than 20 times without losing its effectiveness. The slow deactivation after 20 cycles is caused by the accumulation of degradation products, which can be partly overcome by replenishing the degradation agent.
Despite the promising results, several challenges remain before the technology can be scaled up to commercial use in the recycling of high-voltage batteries. The viscosity of the reaction mixture increases as the polymer chain fragments accumulate, and the separation and purification of the degradation products need further optimization. The ionic liquids I used contain transition metals, which might be a concern if they remain in the recovered battery materials. Even though the metal is strongly coordinated in the anion, a thorough rinsing and purification protocol must be established. In future work, I plan to explore ionic liquids with lower metal content or even metal-free ionic liquids, and to develop continuous-flow degradation processes that are better suited for industrial throughput. Another direction is the integration of the degradation step with the subsequent hydrometallurgical extraction of valuable metals from the high-voltage battery electrodes. If the degradation of polyurethane structural adhesive and the leaching of cathode materials can be combined in a single process, the overall recycling efficiency and economic feasibility would be greatly improved.
In conclusion, this work has demonstrated that polyurethane structural adhesives, which present a major hurdle in high-voltage battery recycling, can be degraded efficiently by ionic liquid catalyzed alcoholysis. The low-temperature operating window and the reusability of the catalyst provide a promising path for the development of safe and sustainable battery disassembly technologies. My findings are expected to contribute to the circular economy of high-voltage battery materials and to support the ambitious environmental goals of reducing waste and conserving valuable resources.