Structure–property relationship in AESO–vanillin copolymers: Effect of vinyl group position on polymer rigidity

Materials

Vanillin, cinnamyl bromide (purity ≥97%), 4-vinylbenzyl chloride (purity ≥90%), Acrylate-Epoxidized Soybean Oil (AESO, 4000 ppm MEHQ as inhibitor), Sodium Borohydride (NaBH₄), Toluene, Potassium tert-Butoxide (purity ≥98%), tert-Butanol, silica gel (60 Å, mesh = 200-400), Magnesium Sulfate Heptahydrate (MgSO₄·7H₂O), inhibitor remover of hydroquinone and monomethyl ether hydroquinone, and Sodium Hydroxide (purity 98%) were purchased from Sigma-Aldrich. Ethyl alcohol, Tetrahydrofuran (THF), Hexanes, Ethyl Acetate in analytical degree were supplied by Fermont. Except for AESO, all the chemicals were used as received. AESO was passed through a packed column of inhibitor remover to remove MEHQ.

Characterization

1H-NMR spectra were recorded using a Bruker spectrometer operating at 300 MHz. Deuterated dimethyl sulfoxide (DMSO-d₆) was used as the solvent for vanillin derivatives, while deuterated chloroform (CDCl₃) was employed for AESO samples. Tetramethylsilane (TMS) served as the internal standard, and chemical shifts (δ) are reported in parts per million (ppm). The acrylate content in AESO was quantified following the method described in.^24,25

FTIR-HATR spectra were obtained using an IRPrestige-21 SHIMADZU spectrometer equipped with a Horizontal Attenuated Total Reflectance (HATR) accessory featuring a diamond crystal. Spectra were collected over the range of 4000–560 cm^-1 with a resolution of 4 cm^-1 and averaged over 64 scans.

Thermal analysis (DSC-TGA) was conducted using a PerkinElmer Simultaneous Thermal Analyzer (STA 8000) under a nitrogen purge (20 mL/min) at a heating rate of 10°C/min, covering a temperature range from 30 to 650°C.

DSC curves for the polymer samples were recorded using a TA Instruments Discovery 250 calorimeter under a nitrogen atmosphere. For glass transition temperature (Tg) determination, approximately 5 mg of each sample was first heated from 25 to 175°C at 10°C/min to erase thermal history. The samples were then cooled to −50°C at the same rate, followed by a second heating cycle from −50 to 175°C. T_g values were extracted from the second heating curve.

Swelling testing. Swelling experiments were performed by gravimetric method performed in three different solvents: THT, Toluene and DMF. Circular polymer specimens (2.5 cm in diameter, 2 mm thick) were prepared (as following recommendations from polymer testing standards, ²⁶ weighed (initial mass, (m_i), and immersed in solvent for 24 h. After immersion, the samples were gently dried to remove excess solvent and immediately re-weighed to record the final mass, (m_f).

Synthesis of vanillyl alcohol from vanillin

Vanillyl alcohol was synthesized following a microscale method ²⁷ with modifications ²⁸ to improve the yield (68.3%) and adapt the procedure to a larger scale. Vanillin (10 g, 0.066 mol) was dissolved in 20 mL of ethanol in a beaker. Subsequently, 9.6 mL of sodium borohydride solution (3.4 M in 1.0 M NaOH) was added dropwise under magnetic stirring. The addition was carried out slowly due to the exothermic nature of the reaction.

After complete addition of the NaBH₄ solution, the reaction mixture was placed in an ice bath and stirred for 20 min. It was then removed from the ice bath and stirred at room temperature for an additional 10 min to ensure completion of the reduction. The mixture was returned to the ice bath, and 6 M hydrochloric acid was added dropwise with continuous stirring until hydrogen gas evolution ceased, the pH became acidic, and a precipitate formed.

The resulting precipitate was collected by vacuum filtration, washed thoroughly with distilled water, and dried in a desiccator for at least 24 h. The crude vanillyl alcohol was then recrystallized from toluene to obtain the purified product, with a yield of 76%.

Synthesis of monomer vinylic aromatic derivatives of vanillin

The synthesis of vinylic aromatic monomers derived from vanillin (Figure 1) was carried out as follows. A two-neck round-bottom flask was charged with 0.058 mol (6.55 g) of potassium tert-butoxide, 65 mL of tetrahydrofuran (THF), and 30 mL of tert-butanol as solvents. The flask was placed in a high-frequency ultrasonic bath for 10 min to promote dissolution and activation. It was then equipped with a magnetic stirrer and transferred to an oil bath at reflux temperature.

OPEN IN VIEWER

Figure 1.

Chemical structure of vinylic aromatic derivates of vanillin.

Separately, 0.019 mol (3 g) of vanillyl alcohol was dissolved in 20 mL of THF. This solution was added slowly to the potassium tert-butoxide mixture under constant stirring. A precipitate formed due to the deprotonation of vanillyl alcohol. After 20 min of stirring, 0.043 mol (7.3 g) of 4-vinylbenzyl chloride or 0.043 mol (8.7) g of cinnamyl bromide were added to produce VAVE and VAVC, respectively (Figure 1). The reaction mixture was stirred at reflux for 26 hr.

After completion, the reaction was cooled to 50°C and concentrated under reduced pressure. Then, 90 mL of purified water was added, and the aqueous phase was extracted with ethyl acetate (3 × 100 mL). The combined organic layers were dried over magnesium sulfate heptahydrate (MgSO₄·7H₂O) and concentrated under vacuum.

The crude product was purified by column chromatography using silica gel and a toluene/hexane mixture (70:30 v/v) as the eluent. VAVE yield was 65%; while VAVC was recrystallized using a hexane/ethyl acetate mixture (90:10 v/v) to obtain it with a yield of 63%.

Synthesis of polymers

Polymer mixtures were prepared according to the formulations listed in Table 1. Each mixture was manually stirred with benzoyl peroxide (PBO) as a free-radical initiator and dissolved in a minimal volume of tetrahydrofuran (THF) until complete homogenization was achieved. Subsequently, THF was removed under vacuum, and the resulting viscous mixtures were poured and uniformly spread onto glass substrates (2 × 2 cm).

OPEN IN VIEWER

Table 1.

Composition of polymer mixtures and DSC data: initial (T_i), final (T_f), and maximum (T_max) temperatures of the polymerization exotherm.

Sample	AESO mol	VAVE mol	VAVC mol	PBO mol %	Ti-Tf°C	Tmax°C
PAESO	1	0	0	0.05	84.7-145.6	101.5
PVAVE	0	1	0	0.03	111.7-149	125.7
AESO-CO-VAVE	1	1.5	0	0.05	107.6-148	124.2
AESO-CO-VAVC	1	0	1.5	0.05	107.2-146	128.7
VAVE-CO-VAVC	0	0.7	0.3	0.05	Not defined	Not defined

Differential Scanning Calorimetry (DSC) analyses were performed on both the individual monomers and their mixtures to obtain information regarding the polymerization exotherm. With exception of mixture VAVE-VAVC, in all other cases, an exothermic peak was observed. The initial (T_i) and final (T_f) temperatures, as well as the maximum temperature (T_max) of the polymerization exotherm, are summarized in Table 1. As shown, 120°C was selected as the polymerization temperature since all curves include this value and it lies near the peak of the exothermic transition. Therefore, curing of the samples deposited on the glass substrates was carried out at 120°C for varying times depending on the formulation. Additionally, AESO-co-VAVE and AESO-co-VAVC samples underwent a post-curing step at 140°C for 1 h to enhance DB conversion and improve thermal stability.

Vanillin and vanillyl alcohol characterization results

Vanillyl alcohol was synthesized following previously reported procedures and characterized using FTIR-HATR, ¹H NMR, ¹³C NMR (DEPT-135 analysis), and DSC techniques.

The FTIR spectrum of vanillyl alcohol was analyzed, with band assignments based on literature references.^29,30 The characteristic carbonyl stretching vibration of the aldehyde group from vanillin at 1660 cm^-1 was absent in the spectrum of vanillyl alcohol. Additionally, vanillyl alcohol exhibited a distinct characteristic band at 3445 cm^-1, attributed to free hydroxyl groups. Signals associated with the methylene group in alkyl alcohols (R–CH₂–OH) were detected at 1430 cm^-1. Furthermore, two bands at 1033 and 995 cm^-1 were observed, corresponding to the stretching vibrations of CH₂–OH bonds. These spectral features confirm the conversion of vanillin to vanillyl alcohol via reduction.

Vanillyl alcohol (VA) was also characterized by ¹H NMR spectroscopy and peak assignments were made according to literature.^29,31 The ¹H NMR spectrum of vanillyl alcohol showed singlet signals at 4.98 and 4.38 ppm, attributed to the benzylic methylene group (s, 2H, Ar–CH ₂ –OH) and its associated hydroxyl proton (s, 1H, Ar–CH₂–OH), respectively. Aromatic protons appeared as multiplets at 6.69 and 6.87 ppm (m, 3H, Ar–H).

The chemical structures of vanillin and vanillyl alcohol were further confirmed by ¹³C NMR using DEPT-135 analysis. The methine carbon adjacent to the aldehyde group in vanillin appeared at 129.05 ppm in the positive phase. Upon reduction, this carbon was converted into a methylene group in vanillyl alcohol, shifting upfield to 63.44 ppm and appearing in the negative phase, consistent with its new chemical environment as part of an alkyl alcohol.

The vanillyl alcohol exhibited a melting point of 115°C, aligning with the literature range of 110–117°C, ²⁷ and the measured enthalpy was 30.08 kJ/mol, closely matching the literature value of 30.74 kJ/mol. ³² Minor discrepancies in melting point and enthalpy may be attributed to variations in sample purity, crystallinity, or experimental conditions.

Characterization of vinylic aromatic derivatives of vanillin

The synthesis of vinylic aromatic derivatives of vanillin was carried out as previously described, and its characterization was performed using FTIR-HATR spectroscopy, ¹H and ¹³C NMR, and thermal analysis by DSC-TGA.

Figure 2 shows the FTIR spectra of the vinylic aromatic derivatives of vanillin: VAVE, which contains a terminal vinyl group, and VAVC, which features a central double bond. As with the spectra of vanillin and vanillyl alcohol, all spectra were normalized to the aromatic skeletal vibration band at 1510 cm^-1 for comparative analysis.

OPEN IN VIEWER

Figure 2.

FTIR spectra of vinylic aromatic derivatives of vanillin: VAVC.

The disappearance of hydroxyl group signals in the spectra of VA derivatives confirms their successful conversion into vinyl functionalities. In place of these signals, weak bands appear in the region of 3140–3000 cm^-1, corresponding to the stretching vibrations of = CH and = CH ₂ groups associated with vinyl moieties. Additional bands between 3000 and 2800 cm^-1 are attributed to C–H stretching vibrations from methyl and methylene groups.

Characteristic signals related to the vinyl double bonds are highlighted in the spectra: a band at 975 cm^-1 in the VAVC spectrum is attributed to the trans C = C stretching vibration, ³³ while VAVE exhibits bands at 990 cm^-1 and 914 cm^-1. The signal at 990 cm^-1 corresponds to in-phase trans C–H wagging, and the band at 914 cm^-1 is assigned to = CH ₂ wagging vibrations. ³³ Both spectra also display a signal at 1375 cm^-1, commonly referred to as the “umbrella mode.” In the region between 1350 and 1000 cm^-1, bending and symmetric stretching vibrations are observed, associated with aromatic C–H, C–O, C–O–C, and CH ₂ –O–CH ₂ groups. Below 900 cm^-1, additional bands correspond to C–C stretching vibrations of the aryl ring, along with C–H wagging and rocking modes.

Figure 3 presents the ¹H NMR spectra of the vinylic aromatic derivatives of vanillin: VAVC and VAVE. For VAVC, a singlet corresponding to the methyl protons (CH₃, 3H) appears at 3.79 ppm, while for VAVE it is observed at 3.77 ppm. The methylene protons from aryl, alkyl ether (s, 2H, Ar–CH ₂ –O-R) are detected at 4.46 ppm for VAVC and at 5.07 ppm for VAVE; whereas methylene protons from diaryl ether (s, 4H, Ar–CH ₂ –O–Ar) appear at δ 4.49 ppm and δ 4.44 ppm, respectively.

OPEN IN VIEWER

Figure 3.

¹H NMR spectra of vinylic aromatic derivatives of vanillin: VAVC and VAVE.

Vinylic protons in VAVC (m, 4H, –CH = CH–) are observed in the range 6.80–6.35 ppm. For VAVE (m, 6H, CH = CH ₂ ), signals appear at 7.01–6.83 ppm and 7.51–7.31 ppm. The splitting of signals in these regions is attributed to the presence of rotamers, which result from restricted rotation around the vinyl group. Finally, aromatic protons Ar-H) appear downfield in the range 7.55–6.94 ppm, corresponding to 13 protons in VAVC and 11 protons in VAVE, both as multiplets.

The thermal properties of the vinylic aromatic derivatives of vanillin were evaluated using DSC-TGA, and the results are summarized in Table 2. According to the results, the VAVC monomer exhibits significantly greater thermal stability (321°C) than VAVE (241°C), despite both being positional isomers. Notably, both compounds show higher decomposition temperatures than vanillyl alcohol (188°C), indicating enhanced thermal resistance. The melting point of VAVC (92°C) is also higher than that of VAVE (85°C), which is attributed to more efficient molecular packing in VAVC due to its trans configuration. Interestingly, the enthalpy of fusion for VAVC (42.7 kJ/mol) is lower than that of VAVE (47.2 kJ/mol). The difference in melting points and enthalpies between VAVE and VAVC suggests that molecular packing and stereochemistry play a key role in thermal behavior. VAVC’s trans configuration likely promotes tighter packing and higher melting point, while its lower enthalpy of fusion indicates weaker intermolecular interactions—an observation also noted in vanillin-based networks discussed by. ²¹

OPEN IN VIEWER

Table 2.

DSC-TGA analysis results.

Compound	Decomposition temperature (°C)	Melting point (°C)	Enthalpy (kJ/mol)
VAVE	241	85	47.2
VAVC	321	92	42.7

Characterization of polymers

Five polymers were synthesized as described in Table 1. Polymerization times varied depending on the composition. Post-curing was applied to the PAESO, AESO-co-VAVE, and AESO-co-VAVC samples to ensure near-complete vinyl conversion, as confirmed by the disappearance of double-bond signals in the FTIR spectra and by the polymerization exotherm observed in DSC analysis. All resulting polymers were solid at room temperature.

To confirm the double-bond conversion (DC (%))—HATR-FTIR spectra were collected from the reaction mixtures at the initial stage. Subsequent spectra were recorded hourly throughout the polymerization process to monitor the progression of the double-bond reaction. After normalizing the spectra, the reduction or disappearance of peak areas associated with –C = C–H stretching vibrations indicated the transformation of reactive groups into a crosslinked structure. Characteristic –C = C–H absorption bands for the polymers were observed at: 1600, 985, and 808 cm^-1 for AESO; 989 and 914 cm^-1 for VAVE; and 975 cm^-1 for VAVC.

As an example, Figure 4 illustrates the progression of DB conversion between AESO and VAVE. The spectra were normalized with respect to the signal at 1720 cm^-1, corresponding to the carbonyl (C = O) group of AESO. This signal remains unchanged throughout the polymerization of AESO and during the copolymerization process. Reaction monitoring was performed by tracking the areas under three characteristic peaks associated with the double bonds of both monomers:

• The signal at 990 cm ^-1 represents an overlap of contributions from both monomers.

OPEN IN VIEWER

Figure 4.

Continuous monitoring of the AESO– VAVE copolymerization reaction by FTIR.

The reaction progress, DC (%) was determined by analyzing spectral changes in the 910 cm^-1 band and applying equation (1).^34,35

Considering that, once the curing process begins, the signals at 990 and 809 cm^-1 broaden, thereby increasing their area under the curve, it is important to note that the latter signal does not disappear completely. As previously mentioned, this band represents an overlap of multiple vibrational modes. Therefore, for determining the percentage of conversion, only the area under the 910 cm^-1 signal is considered.

A clear decrease in the 910 cm^-1 band is observed as curing progresses, and it disappears entirely after the post-curing (PC) process. At 7 h (clear blue line), the band is still distinguishable in both overlapping regions. Quantification of the area under the 910 cm^-1 band (Table 3) indicates that the reaction has progressed by 42.6%. After post-curing, the bands reduce to a line. This observation is further supported by DSC analysis, which confirms that the exothermic signal associated with the AESO–VAVE reaction diminishes until only a flat baseline remains.

OPEN IN VIEWER

Table 3.

Determination of the conversion percentage (DC (%))in the AESO–VAVE copolymerization reaction.

Time (h)	Area (910 cm-1)	DB conversion (%)
0	33.7	0
1	30.4	9.8
2	26.7	20.8
4	26.1	22.6
5	24.0	28.8
6	21.3	36.8
7	19.0	43.6
Postcuring	--	100

With the exception of the VAVE-co-VAVC system whose monitoring proved challenging due to signal broadening, which hindered accurate quantification of double-bond-associated peaks, the degree of double bond conversion was plotted in Figure 5 to illustrate the reaction progress for each formulation (excluding postcuring data). A summary of the reaction performance and the thermal properties is provided in Table 4. The values in Table 4 are condensed from the corresponding DSC and TGA thermograms shown in Figures 6 and 7, respectively.

OPEN IN VIEWER

Figure 5.

Progress of the crosslinking reaction for PAESO, PVAVE, AESO-co-VAVE, and AESO-co-VAVC.

OPEN IN VIEWER

Table 4.

Double bond conversion, decomposition temperatures (T₁₀ and T₅₀), and T_g for po-lymers.

Sample	Curing time (hr)	Postcuring (140°C for 1hr)	DB conversion (%)	T10 (°C)	T50 (°C)	Tg (°C)
PAESO	9	✓	95	347.0	398.0	16
PVAVE	4	X	100	331.6	425.0	55
AESO-CO-VAVE	7	✓	100	337.0	--	60
AESO-CO-VAVC	8	✓	75.2	322.0	393.1	20
VAVE-CO-VAVC	6	X	100	317.0	417.4	70

OPEN IN VIEWER

Figure 6.

Comparative T_g of the polymers obtained by DSC.

OPEN IN VIEWER

Figure 7.

TGA curves of the polymers showing T₁₀ and T₅₀.

The variation in curing times observed among the different samples can be attributed to differences in polymerization kinetics and network formation, which are strongly influenced by the chemical structure of the monomers involved. PAESO formation required the longest curing time (9 h), consistent with its relatively low reactivity and flexible aliphatic backbone, which slows down crosslinking. In contrast, the homopolymer of the terminal vinyl monomer (PVAVE) cured much faster (4 h, Figure 5), reflecting the high accessibility and reactivity of terminal double bonds toward radical attack. ³⁶

When AESO was copolymerized with VAVE (AESO-co-VAVE), the curing time increased to 7 h, indicating that although terminal vinyl groups accelerate polymerization, the presence of AESO introduces steric hindrance and reduces mobility, delaying gelation. ³⁷ Similarly, AESO-co-VAVC required 8 h, which can be explained by the lower reactivity of internal double bonds in VAVC; their conjugated nature and steric environment limit radical addition, slowing network formation. The copolymer of both vanillin-derived monomers (VAVE-co-VAVC) showed an intermediate curing time (6 h), suggesting a balance between the rapid reactivity of terminal vinyl groups and the stabilizing, but slower, contribution of internal double bonds.

Overall, the curing times reflect a combination of kinetic limitations, steric effects, and vitrification phenomena. Terminal double bonds promote faster polymerization, while internal double bonds and AESO’s flexible aliphatic segments introduce delays due to reduced reactivity and increased steric hindrance. These results highlight the critical role of double bond positioning and molecular architecture in determining the curing behavior of bio-based copolymers.

The visual characteristics of the solid crosslinked polymer films are shown in Figure 8. The homopolymer PAESO exhibits a yellow coloration and a flexible consistency. In contrast, the copolymers AESO-co-VAVE and AESO-co-VAVC display deeper hues and increased rigidity. Among them, AESO-co-VAVC is the most rigid but also the most brittle, whereas AESO- co-VAVE presents an orange tint and moderate flexibility compared to PAESO. PVAVE is the most brittle polymer in the series and displays a lighter yellow coloration. All polymers are transparent, indicating uniform crosslinking and good optical clarity.

OPEN IN VIEWER

Figure 8.

Cured polymers: (1) PAESO, (2) AESO-co-VAVE, (3) AESO-co-VAVC, (4) PVAVE and (5) VAVE-co-VAVC.

The thermal degradation of all polymers was characterized by thermogravimetric analysis (TGA) under a nitrogen atmosphere, as previously described (Figure 7). Weight loss was recorded as a function of temperature, and the results are summarized in Table 4. For all polymers, thermal decomposition occurred in a single stage, within the temperature range of 315–420°C. PAESO shows high thermal stability, with a T₁₀ of 347°C. This is attributed to its long aliphatic chains and dense crosslinking after post-curing, which delay thermal. ³⁸ However, once the decomposition process starts, its degradation process is continuous as reflected by its T₅₀ value. PVAVE and VAVE-co-VAVC showed the highest T₅₀ values (425°C and 417.4°C, respectively), indicating superior thermal stability. This is attributed to the presence of vanillin-derived aromatic structures and efficient DB conversion via terminal vinyl groups.

In contrast, AESO-co-VAVC, which incorporates a central double bond and experiences steric hindrance, exhibited a lower T₁₀ (322°C) and T₅₀ (393.1°C), reflecting its reduced thermal resistance. The aromatic group contributes to char formation and thermal shielding, while the terminal vinyl group ensures efficient crosslinking.^21–23,39 AESO-co-VAVE shows intermediate thermal stability, with T₁₀ at 337°C. The combination of AESO’s flexible chains and VAVE’s aromatic rigidity results in a balanced network. Interestingly, AESO-co-VAVE did not yield a clear T₅₀ value. This absence may be explained by incomplete degradation, even at 650°C, 52.5% of the sample weight remains (Figure 7). The polymer may retain a significant portion of its mass beyond the temperature range analyzed, possibly due to char formation or highly stable residues. This behavior is common in aromatic thermosets and bio-based polymers with high carbon content. ³⁸ These results confirm that aromatic content and vinyl group positioning significantly influence thermal degradation. Terminal vinyl groups promote efficient crosslinking, while central vinyl groups (as in VAVC) reduce reactivity and thermal stability. ⁴⁰

Regarding the glass transition temperature (T_g) values, this behavior is associated with the presence of aromatic rings, as polymer properties are closely linked to their chemical structure. PAESO, for instance, is a modified triglyceride containing long hydrocarbon chains, which leads to a low crosslinking density and a high degree of flexibility in the polymer structure—resulting in a low T_g.^22,23,41 Conversely, when the polymer is composed of aromatic (as in this case) or heterocyclic rings, its thermal resistance is higher, since the rigid structure restricts polymer chain mobility.^13,21,41 This same behavior is observed in PVAVE, which has a T_g of 55°C, and in the copolymer VAVE-co-VAVC, which exhibits a T_g of 70°C. In both cases, the presence of aromatic rings and π–π interactions increases the crosslinking density, resulting in higher glass transition temperature due aromatic stacking between vanillin-derived rings promotes additional non-covalent interactions that restrict segmental motion. These stacking arrangements create localized domains of enhanced rigidity, which complement the covalent network and effectively raise the energy required for chain mobility. As a result, the co-polymer exhibits a relatively higher T_g compared to other systems, demonstrating the synergistic contribution of both chemical crosslinking and physical π–π associations. ⁴²

For homopolymers, PAESO exhibits the lowest T_g (16°C), consistent with its flexible aliphatic structure and low crosslinking density. This makes it suitable for soft, elastomeric applications but limits its thermal performance. Whereas PVAVE shows a higher T_g (55°C), due to its rigid aromatic backbone and efficient crosslinking via terminal vinyl groups. Aromatic monomers restrict segmental motion, elevating T_g. With copolymerization, AESO-co-VAVE achieves a T_g of 60°C, higher than either homopolymer, indicating synergistic enhancement. The combination of AESO’s flexibility and VAVE’s rigidity creates a balanced network with improved thermal properties. AESO-co-VAVC has a T_g of 20°C, only slightly higher than PAESO. The central double bond in VAVC limits crosslinking efficiency, and steric hindrance from both monomers reduces network density. VAVE-co-VAVC reaches the highest T_g (70°C), reflecting the dominance of aromatic content and efficient copolymerization. Despite VAVC’s internal double bond, the presence of VAVE ensures strong network formation. These T_g values align with the principle that higher crosslinking density and aromatic rigidity raise T_g, while long aliphatic chains and steric hindrance lower it.

These results demonstrate that the incorporation of vanillin-based monomers significantly enhances the thermal stability of the resulting polymers. Similar findings have been reported in other studies, where both photopolymerization and thermal initiation led to increased rigidity and brittleness. The behavior observed in AESO-co-VAVC and PVAVE samples reflects comparable structure–property relationships. Moreover, the findings of this study reinforce the potential of vanillin-derived monomers as functional comonomers for bio-based thermosets, offering tunable thermal properties and structural versatility. Future research may explore alternative curing methods or blend ratios to further optimize performance.

Once the polymers were obtained, swelling experiments were performed in three different solvents. THF, toluene, and DMF were selected to represent different polarity ranges and solvation capabilities, which provide complementary insights into the network architecture of the synthesized polymers. Toluene, a non-polar aromatic solvent, is commonly used to probe the affinity of polymer networks with hydrophobic environments, and it is particularly relevant for vanillin-derived monomers due to their aromatic character; in this medium, the monomers were soluble, highlighting the affinity between aromatic domains and hydrophobic solvents. THF, a moderately polar aprotic solvent, offers strong solvation power for a wide range of polymers and is frequently employed to evaluate the accessibility of crosslinked networks; its small molecular size and high diffusivity allow it to reveal differences in network compactness. Here, the monomers were also soluble, confirming its strong solvation power and suitability for evaluating differences in network density. DMF, a highly polar aprotic solvent, interacts strongly with polar functional groups such as carbonyls and hydroxyls, making it valuable for assessing how polarity and hydrogen-bonding capacity influence swelling resistance in copolymers containing AESO segments.

By combining these three solvents, it is possible to establish a comparative framework: polymers that resist swelling in all three media can be considered highly crosslinked and structurally compact, whereas those that show selective swelling or mass loss reveal weaker networks or the presence of soluble fractions. This solvent triad therefore enables a more comprehensive evaluation of crosslink density and structure–property correlations than would be possible with a single solvent system.

The weight ratios before and after the swelling test were related through the following equation (2), and the swelling percentage are reported in Table 5.

% s w e l l i n g = [ m f − m i m i ] x 100

(2)

where m_i is the initial mass and m_f is the final mass after solvent immersion.

OPEN IN VIEWER

Table 5.

Swelling behavior of synthesized polymers in THF, toluene, and DMF.

Polymer	(Δ mass %)			Interpretation of crosslink density
	THF	Toluene	DMF
PAESO	−33	−27	−19	Significant mass loss: low crosslink density, soluble fractions leached out, weak network
PVAVE	0	0	0	No change: Very high crosslink density, rigid and impermeable network
AESO-CO-VAVE	0	+9	−20	Mixed behavior: Intermediate density; terminal vinyl groups promote reactivity, AESO segments reduce compactness
AESO-VAVC	0	−38	0	Heterogeneous network: Internal vinyl groups slow polymerization, producing less uniform crosslinking
VAVE-CO-VAVC	0	0	0	Stable: Balanced network; combination of terminal and internal vinyl groups yields moderate but consistent crosslink density

The swelling data supports the qualitative discussion of crosslink. The swelling behavior in THF, toluene, and DMF highlights clear differences in network architecture among the polymers. PAESO shows substantial mass loss in all three solvents, indicating a weakly crosslinked network with soluble fractions that are easily extracted. In contrast, PVAVE remains completely stable, reflecting a highly crosslinked and compact structure formed by the rapid polymerization of terminal vinyl groups. ³⁶ The copolymers AESO-co-VAVE and AESO-VAVC exhibit intermediate or heterogeneous behaviors reflecting the interplay between AESO’s flexible aliphatic segments and the reinforcing effect of vanillin-derived aromatic. AESO-co-VAVE shows slight swelling in toluene but mass loss in DMF, suggesting a balance between reactive terminal vinyl groups and flexible AESO segments ³⁹ ; AESO-co-VAVC shows strong mass loss in toluene, consistent with slower polymerization and less uniform crosslinking due to internal vinyl groups. Notably, the VAVE-co-VAVC copolymer remains stable across all three solvents, indicating a balanced network where terminal double bonds drive reactivity and internal double bonds contribute to efficient packing and rigidity.

These results confirm that double bond positioning (terminal vs internal) directly influences polymerization kinetics and network architecture, and the swelling ratios provide quantitative evidence to support the structure–property correlations discussed in the manuscript.