Effects of Water and Air on the Reaction Pathways in the Hydrothermal Liquefaction of Plastics

Abstract

This study establishes the role of water in the hydrothermal liquefaction (HTL) of plastics. 1‐dodecene is employed as a model compound for polyolefin fragments resulting from chain cracking. The presence of a terminal double bond enables potential hydration and oxidation reactions to be assessed. Comparative experiments are conducted under identical conditions of temperature using pyrolysis and HTL in the presence and absence of air. The resulting product distributions provide critical insight into the operative reaction pathways, specifically highlighting the function of water as a reaction medium. Importantly, although hydrogen exchange and hydrogen abstraction are well‐established, water does not appear to react oxidatively with the hydrocarbon substrate, as no incorporation of oxygen is observed when operating under an argon atmosphere. When operating under air, oxygenated compounds are detected in the products, particularly as light alcohols, aldehydes, and acids, consistent with the established supercritical water oxidation reaction pathways.

Is water an innocent solvent for the hydrothermal liquefaction (HTL) of plastics? Does it also react with plastic under supercritical conditions? Only in the absence of oxygen does water appear to be the former. Despite observations that residual oxygen under HTL conditions is the sole source of oxidized reaction products, several aspects of the reaction pathways remain to be fully understood.

1. Introduction

The extraordinary growth of the plastics industry since the mid‐20^th century has established plastic materials as a ubiquitous feature of modern life.^{[ 1 ]} This growth is a trend that is very likely to continue, as replacing plastics in their many uses is proving to be difficult.^{[ 1} , ² , ^{3 ]} This presents two major problems. First, the production of new, virgin plastics in the main requires nonrenewable fossil fuels as a feedstock. Second, most plastics in use are nonbiodegradable, and so are not readily decomposed in the environment.^{[ 1} , ^{4 ]} Ultraviolet light‐induced decomposition provides one of the main pathways for plastic breakdown into smaller particles, often referred to as “microplastics.” There is a growing body of research describing the potentially toxic impacts of these on humans and the broader environment, particularly in the oceans.^{[ 5} , ⁶ , ^{7 ]} Transitioning to a circular economy in which new plastic stocks are produced by recycling existing end‐of‐life plastics is a more sustainable way to manage the plastics problem.^{[ 4 ]} Unfortunately, current rates of plastic recycling are poor, with only 22% of the total plastic waste in 2020 being recycled and only 14% of the feedstock for new plastic products derived from recycled plastic.^{[ 8 ]} There are some major limitations regarding the repeated mechanical recycling of plastics, such as the need to separate the waste plastic feedstock by color and type to a high degree of purity, decreasing polymer chain lengths with each recycling iteration, thermal decomposition of polymer chains, and unwanted side reactions involving polymer additives.^{[ 9} , ¹⁰ , ¹¹ , ^{12 ]} Some success has been achieved in recycling polyethylene terephthalate, where the process has been refined with chemical additives that can reform the polymer chains to approximate their original lengths.^{[ 11 ]} Currently, for many other plastics, sending waste to a landfill is simply more economical than investing in the infrastructure and labor required for sustainable recycling pathways.^{[ 13 ]}

Chemical recycling, in which the polymers are depolymerized into potential monomer sources, is proposed as a more effective means of recycling complex mixtures of different plastics and of plastics thought to be otherwise unrecyclable.^{[ 14 ]} The major processes of chemical recycling are solvolysis, pyrolysis, and hydrothermal liquefaction (HTL).^{[ 14 ]} HTL is advantageous over these other processes due to its greater tolerance for mixed, wet, and contaminated waste streams; reduced formation of char or coke; and producing higher yields of liquid oil.^{[ 14} , ^{15 ]} HTL uses hot‐compressed water as a solvent to depolymerize a substrate, typically with water in the supercritical state (minimum temperature of 647 K (374 °C) and pressure of 22.1 MPa (221 bar)) for the recycling of polyolefins.^{[ 16 ]} The properties of water near and above the critical point vary significantly. They can be controlled by careful selection of the reaction temperature and pressure, with the latter controlled by the water density/reactor fill ratio.^{[ 17} , ¹⁸ , ¹⁹ , ²⁰ , ^{21 ]} In particular, the static dielectric constant and ionization product of water can exhibit significant changes around the critical point. These are illustrated in Figure S1, Supporting Information. The properties of water can be tuned by selecting conditions that will favor ionic chemistry (low pK_w) or radical chemistry (high pK_w).^{[ 18} , ^{22 ]} HTL of plastics is typically conducted under conditions with a high pK_w, to promote radical decomposition reactions.

Water has several roles in the HTL process, acting as a catalyst, solvent, or reactant.^{[ 18 ]} As a catalyst, water facilitates acid‐ and base‐catalyzed reactions, particularly at conditions where the pK_w is lower than that of ambient water.^{[ 18 ]} As a solvent, water helps transfer energy to the reactants and can swell polymers and/or solvate organic molecules, inhibiting unwanted crosscoupling reactions.^{[ 15} , ^{18 ]} Yet, the role of water as a reactant is not as well understood, and limited information is available in the open literature other than for the case of hydrogen exchange based on studies of deuterated water.^{[ 23} , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ^{32 ]} In particular, there is considerable uncertainty and disagreement regarding the reaction pathways responsible for the incorporation of oxygen into liquid oil products.^{[ 14 ]} Several authors have reported the formation of oxygenated species such as alcohols, ketones, or aldehydes during the HTL of polyolefins and have used this to indicate that water reacts with the hydrocarbons present, marking a distinctive reaction pathway that is absent from pyrolysis.^{[ 14} , ³³ , ³⁴ , ³⁵ , ^{36 ]} However, other studies report that no oxygenated species were detected when conducting HTL processing.^{[ 15} , ²³ , ³⁷ , ³⁸ , ^{39 ]} It has been theorized that the presence of trace oxygen from air may be responsible for enabling a small degree of supercritical water oxidation (SCWO) reactions. Still, to our knowledge, no study has been reported in the literature that directly compares the effects of air and water on the polymer decomposition process.^{[ 32 ]} Two studies reported the formation of oxygenated species, yet make no mention of purging the reactor of air, and hence, SCWO reactions may occur.^{[ 33} , ^{34 ]} However, two additional studies conducted the reaction without oxygen and observed oxygenated compounds in the liquid oils.^{[ 35} , ^{36 ]} Wong et al.^{[ 36 ]} ascribed their result to oxygen remaining in their reactor, even after purging the atmosphere with inert gas, while Colnik et al.^{[ 35 ]} indicated that alcohols form in their reactor from the hydration of alkenes.

We report here a clarification of the role of water as a reactant in HTL processing of plastics. 1‐dodecene was selected as a model compound for cracked polyolefin fragments, as it is both capable of undergoing further cracking reactions and possesses a double bond that could potentially react with water or air in a hydration or oxidation reaction. Liquid products were generated by pyrolysis and by HTL at the same temperatures to elucidate some of the main reaction pathways elements. A reaction time of 15 min was selected for this study as polyolefins have previously been reported to undergo observable cracking in this time frame.^{[ 35} , ⁴⁰ , ⁴¹ , ^{42 ]} We therefore selected 1‐dodecene to model the behavior of these polyolefin fragment intermediates.

2. Results and Discussion

The following discussion considers four separate reaction conditions: pyrolysis in argon (PYR‐argon), “pyrolysis” in air (PYR‐air), HTL in argon (HTL‐argon), and HTL in air (HTL‐air). For all 4 conditions, the 1‐dodecene loading was 0.0066% v/v with respect to the reactor volume and the reaction temperature was 450 °C. This loading was selected for safety considerations and to allow for efficient mass transfer within the system. The working pressure of the reactors used in this study was calculated to be 267 bar, and the loading selected ensured that the system pressure would remain below this, even in the event of unexpected gas formation. Full calculations of the pressures are provided within the Supporting Information. For the HTL experiments, the water loading was 10% v/v, which was calculated to give a pressure of 236 bar at 450 °C, with an additional 7 bar pressure originating from the substrate and headspace atmosphere. Pyrolysis is defined by IUPAC as being performed in an inert atmosphere; in air some degree of combustion is also expected.^{[ 43 ]} However, once the oxygen is consumed, the atmosphere is effectively inert, and pyrolysis reactions can occur in this scenario. All experiments were conducted in triplicate. An overview of the reaction and analysis processes is shown in Figure  1 and full details are provided in the Experimental Section. After 15 min of reaction time, reactions under all four conditions showed a partial conversion of the starting material into smaller compounds. For the experiments in argon, conversion was 41 ± 4% for pyrolysis and 36 ± 6% for HTL. Replacing the atmosphere with air resulted in a higher conversion of 53 ± 2% for pyrolysis and 57 ± 4% for HTL. In all four cases, the remainder was unreacted 1‐dodecene. The increase in conversion for experiments performed in air is indicative of a reaction taking place between the oxygen in air and the 1‐dodecene substrate. As the reaction temperature is above the autoignition temperature of 1‐dodecene (255°C), it is also possible that a spontaneous partial combustion or oxidation reaction has occurred in these experiments,^{[ 44 ]} although there did not appear to be any significant difference in conversion between HTL and pyrolysis when carried out in the same conditions.

Figure 1.

Overview of thermal decomposition reactions and analysis protocols.

2.1. GC‐MS Analysis

Representative gas chromatography‐mass spectrometry (GC‐MS) chromatograms from each set of conditions are overlaid in Figure  2 . The chromatograms are nearly identical, with the same set of peaks and compounds observed in all four cases – the major identifiable peaks are listed in Table  1 , and a full list of peaks is presented in Table S1 in the Supporting Information. In all four experiments, the major peaks (excluding the internal standard and remaining starting material) were matched using Agilent MassHunter and the NIST19 database as alkenes, with 1‐alkenes between 1‐octene and 1‐undecene observed. There was a small amount of tridecene detected, which could indicate a small degree of cross coupling took place between a molecule of 1‐dodecene or a shorter‐chain intermediate and a hydrocarbon radical. No peaks in any chromatogram were attributable to any oxygenated compound. The GC‐MS could detect only C₈+ hydrocarbons, suggesting that the starting material and early reaction intermediates do not react directly with water or oxygen.

Figure 2.

GC‐MS chromatograms of acetone soluble products obtained after 1‐dodecene reactions at 450 °C and 15 min under the following reaction conditions: a) HTL‐air, b) PYR‐air, c) HTL‐argon, and d) PYR‐argon.

Table 1.

Identification of key compounds from GC‐MS chromatograms of 1‐dodecene decomposition.

Retention time [min]	Compound identification	Retention time [min]	Compound identification
4.25	1‐octene	12.30	1‐undecene
6.56	1‐nonene	15.20	1‐dodecene (starting material)
9.35	1‐decene	17.97	1‐tridecene

This indicates two major findings. First, in the presence of water, alkenes do not show any evidence of undergoing a hydration reaction at 450 °C and 243 bar. Second, the major reaction pathways appear to be the same for both HTL and pyrolysis. The lack of any observed hydration reactions is at first surprising; however, sub and supercritical water has been known to act as a solvent for dehydrating alcohols.^{[ 45} , ⁴⁶ , ^{47 ]} Above 150 °C, the equilibrium between alkene, water, and alcohol shifts to favor the alkene, with rapid conversion from alcohol to alkene observed above 200 °C.^{[ 46 ]} Under conditions where the pK_w of water is low, water can act as an acid catalyst due to the increased extent of autoionization.^{[ 46 ]} The reverse reaction, where an alkene is hydrated to the alcohol, is observed to an extent as part of the equilibrium of this reaction, but this is reported for conditions in which the pK_w of water is low enough to support acid catalysis. The reactions in this study are conducted in a more gas‐like supercritical water phase, with a pK_w of about 18, so it is possible that the hydration reaction pathways remain inaccessible when there is no significant capacity for the solvent to participate as an acid catalyst. Our observations also rule out a radical mechanism for primary alkene hydration under the conditions of this study.

2.2. ¹H NMR Analysis

Water‐suppressed ¹H NMR spectra were obtained for each of the aqueous extractions of the reactions. No‐D NMR was first obtained (Figures S2, S4, S6, and S8 in the Supporting Information) in an attempt to detect any labile protons present in the sample mixture. 10 vol% of D₂O was then added to collect the spectra shown in Figure  3 . There were no significant peaks present in the no‐D NMR spectra that disappeared upon addition of D₂O for all reaction conditions. The addition of D₂O to the NMR samples resulted in spectra that were more readily analyzed than the no‐D spectra due to the increased signal‐to‐noise ratio (Figures S3, S5, S7, and S9, Supporting Information). Variation across several HTL‐argon runs was observed as very low intensity resonances in the 3.3–3.8 ppm spectral region of the ¹H NMR. We attribute these small signals to the effect of adventitious oxygen.

Figure 3.

¹H‐NMR (water suppressed) of the aqueous phase (10% D₂O) obtained after 1‐dodecene reactions at 450 °C for 15 min under the following reaction conditions: a) HTL‐air, b) PYR‐air, c) HTL‐argon, and d) PYR‐argon.

When comparing the spectra, several notable features can be observed. First, there are more compounds present when the reaction is performed under air rather than under argon. Second, HTL and pyrolysis conditions produce very similar spectra, with greater differences due to the presence or absence of oxygen, rather than that of water. There were many signals that could not be unambiguously assigned to any specific molecular structures. However, some extra signals in the spectra associated with conversion in air can be attributed to oxygenated compounds. For example, there are two signals; a triplet at 9.67 ppm and a quartet at 9.66 ppm, attributable to propionaldehyde and acetaldehyde, respectively. The triplet at ≈1.17 ppm and quartet at 3.65 ppm, plus the singlet at 3.34 ppm are attributable to ethanol and methanol, respectively. The singlet at 2.08 ppm is attributable to acetic acid. The presence of other alcohols is also indicated in the reactions run under air. The complicated sets of multiplets evident in the chemical shift range 0.85–1.25 ppm were assigned to the terminal methyl groups of these alcohols. The inset displaying the resonances from 5.0–8.0 ppm is also markedly different in the reactions run under air compared with those conducted under argon. The latter being featureless, while the former contains many weak signals attributed to a range of oxidation products. Taken together, the differences between sets of conditions evident in these spectra are consistent with water acting as a solvent only.

2.3. HPLC‐UV Analysis

High‐performance liquid chromatography‐Ultraviolet detection (HPLC‐UV) chromatographic analyses were performed on the aqueous phases of the two supercritical water (HTL‐argon and HTL‐air) and the pyrolysis‐in‐air experiments to detect any light organic acids, if present (Figure  4 ). Multiple peaks were observed in the chromatograms of both experiments conducted in air. Using a reference standard, the peak at a retention time of 23.7 min was identified as acetic acid, consistent with the resonance observed in the ¹H NMR. Other peaks in both chromatograms from reactions in air were not matched, though this wavelength of UV absorbance suggests they belong to carbonyl‐containing molecules. In contrast, the chromatogram for HTL‐argon does not contain any peaks, indicating the absence of organic acids. The HPLC trace for PYR‐argon is similarly featureless (see Figure S10, Supporting Information) as is expected due to the lack of oxidized species, consistent with a lack of oxygen in the reaction environment.

Figure 4.

HPLC‐UV/DAD chromatograms detecting absorbance at 210 nm for a) HTL‐air, b) PYR‐air, and c) HTL‐argon.

2.4. The Effect of Water

The results of this study show that there is very little difference observed between the products of the pyrolysis and HTL reactions for the model compound 1‐dodecene when conducting the reaction under argon. Equally, the GC‐MS chromatograms shown in Figure 2 indicate that there is no significant difference between pyrolysis and HTL products obtained under the same atmosphere. All major peaks correspond to the same compounds, which are largely 1‐alkenes that have shorter chain lengths than the 1‐dodecene starting material. This would then indicate that the homolytic cracking reactions that are occurring are a common reaction pathway between the pyrolysis and HTL regimes. If water is not participating as a reactant, then it is fulfilling the role of a solvent, and the HTL reaction is effectively occurring as “pyrolysis in supercritical water”. The role of the water is to dissolve (for light hydrocarbons) or swell (for plastics) the hydrocarbon feedstock and, as a diluent, control the radical decomposition reaction by reducing the likelihood of radical coupling reactions occurring between two hydrocarbon species.^{[ 15 ]}

Within that paradigm, an additional role for the water is to react with the organic radicals formed by reversible hydrogen abstraction (hydrogen scrambling), as proven through deuterium exchange studies, leading to the temporary formation of hydroxy radicals that then abstract hydrogen from the organic molecules present, thereby “occupying” the organic radicals in such a manner as to hinder crosslinking between them.^{[ 23} , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ^{32 ]}

2.5. The Effect of Air

In contrast to the effect of adding water to the system, replacing the argon atmosphere with air results in a significant change in both the overall conversion and in the product slate of light compounds formed. An increase of conversion of ≈20% with respect to the decomposition conducted in argon indicates that the oxygen in air is reacting with 1‐dodecene to form light compounds that are not detected by the GC‐MS. This could include combustion products (CO₂ or CO) and other light hydrocarbon species. The liquid reaction products detectable via GC‐MS are unaffected between air and argon atmospheres, which indicates that the basic pyrolysis mechanisms are a function of the high temperatures used, providing enough energy to overcome the homolysis activation energy. Only the lightweight product fraction is affected by the presence or absence of oxygen, indicative of a reaction scheme involving both pyrolysis and oxidation. Analysis of the ¹H NMR and HPLC‐UV shows water‐soluble oxygenated species that are present only when air is used as the reaction atmosphere. When argon is used as the reaction atmosphere, all peaks detected in GC‐MS and ¹H NMR correspond to hydrocarbon species, and the HPLC‐UV trace has no peaks. Hence, it appears that oxygen from air is responsible for forming oxygenated reaction products and not water. The formation of oxygenated compounds occurs via an oxidation (or partial oxidation) reaction – under HTL conditions, a partial SCWO process is likely to result in oxygenated products.

Under SCWO conditions, when the system is provided with enough oxygen, energy, and time, all hydrocarbon molecules will be converted into carbon dioxide and water.^{[ 48 ]} The major intermediate species observed for the degradation of hydrocarbon species is acetic acid, which is attributed to the higher activation energy for oxidizing acetic acid as compared to other hydrocarbons.^{[ 49} , ^{50 ]} Under conditions when total oxidation is not achieved, significant concentrations of acetic acid are observed.^{[ 48} , ⁴⁹ , ^{50 ]} Light alcohols, such as ethanol and methanol, also have a higher oxidation activation energy than other hydrocarbons and could be considered intermediate species using the same logic.^{[ 49 ]} However, in practice, alcohols are not observed as SCWO intermediates, and it has been posited that the kinetics of their oxidation reactions are much faster than for acetic acid, and accordingly, the rate limiting step is the oxidation of acetic acid to carbon dioxide.^{[ 49 ]} The combustion of hydrocarbons is also observed to produce carbonyl containing compounds, especially carboxylic acids.^{[ 51} , ^{52 ]} The conditions used in this study, however, are not completely comparable to the reaction setup for a SCWO reaction. As the oxidant is just the residual oxygen from the air atmosphere in a batch reactor, the conditions in the reactors used for this study would only be capable of performing partial oxidation. Complete combustion of 1‐dodecene would require a molar ratio of O₂ to dodecene of 18, but the conditions used in this work are only a ratio of ≈0.28. Hence, intermediates, such as methanol, that would typically oxidize further under SCWO conditions may be trapped and unable to oxidize further due to insufficient oxygen as per the conditions used in this study.

3. Conclusion

Hydrocarbon compounds will readily decompose homolytically at high temperatures, under both pyrolysis and HTL conditions. Using 1‐dodecene as a model compound, the effects that water and oxygen exert on this decomposition reaction have been observed. Water does not appear to react with the hydrocarbon substrate outside of the transfer or exchange of hydrogen from water molecules to hydrocarbon radicals via a hydrogen abstraction reaction, which does not change the molecular structure.^{[ 18} , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ^{32 ]} No incorporation of oxygen is observed when water is present under an argon atmosphere. When oxygen is added to the system, oxygenated compounds are detected in the products, particularly as light alcohols, aldehydes, and acids. This clearly indicates that it is molecular oxygen in the reactor and not water that is reacting with the hydrocarbon substrate under HTL conditions. To match this insight with some reports in the literature that observe some oxidation, it is suggested that those systems may not been degassed sufficiently well to avoid the presence of adventitious oxygen.

Thus, supercritical water plays a significant role in controlling this reaction as a solvent, where the water molecules can dissolve light hydrocarbons or swell larger polymer molecules.^{[ 15 ]} This reduces the likelihood of hydrocarbon radicals reacting with each other and works in unison with reversible hydrogen abstraction involving organic radicals and water, together serving to prevent the formation of undesirable solids. One noticeable limitation of this study is that the reactor design does not allow for gas sampling. Future reactor designs will incorporate a means for gas sampling to assist in the determination of a full carbon balance and analysis of any gaseous products.

4. Experimental Section

4.1.

4.1.1.

Materials

1‐dodecene was obtained from Tokyo Chemical Industry. Acetone (analytical grade) and mesitylene were obtained from Merck. Deuterium oxide was obtained from Sigma‐Aldrich. Tetramethylsilane (TMS) was obtained from Fluka. All reagents were used as received without further purification. Milli‐Q water was freshly obtained from a Merck‐Millipore Milli‐Q EQ 7000 with a conductivity of 18.2 μS m⁻¹.

Procedure

To understand the role of water and air on reactions of 1‐dodecene under high‐temperature (450 °C) conditions, experiments were conducted under four reaction conditions: a) 1‐dodecene only in argon (PYR‐argon), b) 1‐dodecene and water in argon (HTL‐argon), c) 1‐dodecene only in air (PYR‐air), and d) 1‐dodecene and water in air (HTL‐air). The batch reactors used for this work were made of 316 stainless steel and custom built from commercially available components (Swagelok Australia). The body of the reactor was constructed from a 0.5 in. outer diameter tube with a wall thickness of 0.065 in. that was sealed with a cap at one end, and sealed using a gasket and plug at the other end. The internal volume of the reactor was ≈7 mL. The loading of the materials and sealing of the reactors for the PYR‐air and HTL‐air decomposition reactions was conducted under atmospheric conditions. The loading of materials and sealing of the reactors for the PYR‐argon and HTL‐argon decomposition reactions was conducted in a wet‐argon glove box (i.e., a glove box that is purged to exclude oxygen but is not explicitly water‐free, with a typical humidity below 30%), and reagents were degassed before use. 1‐dodecene (46 μL) and milli‐Q water (0.7 mL) were loaded into the reactor by micropipette and the reactor sealed for the HTL‐argon and HTL‐air experiments. For the PYR‐air and PYR‐argon experiments, 1‐dodecene (46 μL) was loaded into the reactor which was then sealed. The volume of air or argon present in each reactor was the remaining headspace of the reactor, at 1 atm of pressure (≈7 mL for pyrolysis or ≈ 6.3 mL for HTL). The reaction was initiated by rapidly submerging the reactor into a fluidized sand bath controlled at the reaction temperature prior to the experiment. Reaction temperature (determined separately) was typically reached within 3 min using this setup and maintained with an error of 2°C. After 15 min of reaction time, the reactor was removed from the sand bath and quenched in a bucket of cool water to prevent further reaction. The reaction time was defined as the time from when the reactor was first placed into the sand bath until the reactor was quenched in the water bath. After the reaction was quenched, the reactor was opened, and the reactor contents were decanted for the HTL‐air and HTL‐argon experiments. The liquid that was decanted at this point was termed the aqueous phase for further analysis. For analysis via GC, the organic liquid contents in the reactor were collected by washing with acetone (3 × 2 mL) and this extract was combined with the previously decanted contents (for HTL‐air and HTL‐argon conditions) to produce one single phase for GC‐MS and GC‐FID analysis as detailed below. For 1H NMR and HPLC‐UV analyses, aqueous phase from HTL‐air and HTL‐argon experiments was decanted from the reactor as described above and used directly. For the PYR‐air and PYR‐argon conditions, the contents of the reactor were washed with water (0.7 mL) and this aqueous phase used for NMR and HPLC analyses. To all four NMR samples, after an initial measurement was taken, deuterium oxide (70 μL) and TMS (≈10 μL) were added, and a new measurement was recorded.

Quantification was performed on an Agilent GC 7890 gas chromatograph, fitted with an Agilent column (HP‐5, 30 m × 320 μm × 0.25 μm) and an FID detector. 46 μL of mesitylene were added to the combined acetone and aqueous phase of each sample as an internal standard, and conversions were calculated from the peak area ratio. The injector temperature was set to 250 °C, and the FID temperature was set to 350 °C. The heating profile of the column oven was 30 °C for 3 min, then increased at 5°C min⁻¹ to 300°C, which was held for 10 min. GC‐MS measurements were recorded on an Agilent 8890 gas chromatograph, fitted with an Agilent column (HP‐5MS, 30 m × 250 μm × 0.25 μm) and connected to an Agilent 5977B mass spectrometer. The injector temperature was set to 250°C, and the injection volume was 1 μL. The heating profile of the column oven was 30°C for 3 min, then increased at 5°C min⁻¹ to 300°C, which was held for 10 min. The mass spectrometer detected masses between m/z = 45 and 550. HPLC‐UV measurements were recorded on an Agilent 1260 Infinity equipped with a Hi‐Plex H column and a DAD UV detector at 210 nm. Injection volume was 50 μL, and the column oven was operated at 40°C with a flow rate of 0.4 mL min⁻¹ of 0.1 M H₂SO₄ mobile phase. ¹H NMR measurements were recorded on a Bruker Avance NEO 500 MHz spectrometer at room temperature.

Supporting Information

The authors have cited additional references within the Supporting Information.^{[ 53} , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ^{60 ]}

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supplementary Material

Steel Jack, Barnett Christopher, Yuen Alexander K. L., Masters Anthony F., Montoya Alejandro, Maschmeyer Thomas, Aida Taku M.. ChemSusChem. 2025; 18, e202501269. 10.1002/cssc.202501269

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.