Synergistic Crosslinking of Triglycidyl Ether of Resveratrol With Amino Acids as a Novel Pathway to Design Sustainable High‐Performance Thermosets

Abstract

This study explores the use of amino acids as renewable curing agents for the synthesis of fully biobased epoxy thermosets derived from triglycidylether of resveratrol (TGER). Combining the multifunctionality of amino acids with the aromatic rigidity of resveratrol, high‐performance sustainable thermosets were designed. To increase the green metrics, series of TGER/amino acid formulations were prepared and polymerized under controlled thermal conditions consistent with green chemistry principles. Differential scanning calorimetry and temperature‐assisted FTIR analyses demonstrated complete functional conversion and network formation through amino ether and ester linkages while dynamic mechanical analysis revealed that the prepared thermosets show glass transition values between 115 and 250 °C and high storage moduli up to 3.6 GPa. Among the examined systems, aromatic amino acids such as tyrosine, tryptophan, and histidine produced the most rigid and performant materials. Tensile and hardness tests confirmed that the designed thermosets are tough, moderately brittle, with Shore D values up to 87. Thermogravimetric and moisture absorption analyses indicated good thermo‐oxidative resistance and low water uptake for aromatic amino acid‐based networks. Overall, the study underscores the feasibility of achieving high‐performance, sustainable thermosets using amino acids as curing agents, supporting the transition toward fully biobased polymers.

This study explores the design of sustainable thermosets by crosslinking the triglycidyl ether of resveratrol with series of amino acids to tailor high‐performance thermosets with excellent thermal, mechanical, and environmental properties for advanced industrial applications.

1. Introduction

High‐performance epoxy thermosets have garnered significant attention over recent decades as advanced alternatives to many metal‐based materials [1], finding critical applications in sectors such as aerospace [2, 3], automotive [4], construction [5], and electronics [6]. These materials have become indispensable in modern industry due to their exceptional mechanical properties [7] and versatility [8, 9]. However, the overwhelming dependence of thermoset manufacturing on petroleum‐derived resources [10] presents substantial limitations [11] that threaten both production sustainability and environmental health [12, 13]. Fossil fuel reserves are finite [14], and the escalating global demand exacerbates resource depletion. Moreover, many petrochemical‐derived compounds, such as bisphenol A—the predominant precursor for the production of epoxy thermosets—are recognized for their toxicity, bioaccumulation, and endocrine‐disrupting effects, posing notable risks to human health and ecological systems [15, 16]. Bisphenol A continues to play a pivotal role in the industrial production of epoxies, as reflected by its substantial global market presence. In 2023, the worldwide market for bisphenol A was estimated at 6.4 million tonnes, and it is expected to reach 9.3 million tonnes by 2030, representing a compound annual growth rate (CAGR) of 5.4% between 2023 and 2030 [17]. Recent years have seen annual production volumes ranging from 3.8 million tonnes to approximately 10 million tonnes. Despite its widespread use and ongoing growth, the persistent reliance on bisphenol A highlights the pressing need to develop safer and more sustainable alternatives for the epoxy industry.

Replacing bisphenol A and other petroleum‐based precursors with nontoxic biobased compounds is a promising strategy towards more environmentally friendly and health‐conscious materials [18, 19, 20]. A diverse array of hydroxyl‐bearing molecules—such as resveratrol (typically extracted from botanical sources like Polygonum cuspidatum and grape pomace, or produced via microbial biosynthesis) [21, 22], phloroglucinol [23], ferulic acid [24], and daidzein [25]—can be sourced from plants and lignin, offering rigid, aromatic structures that impart favorable mechanical properties to biobased epoxy networks. These renewable compounds not only mitigate reliance on fossil resources but also introduce opportunities for improved material performance, reduced toxicity, and the design of functional thermosets with tailored properties, including recyclability and flame retardancy [2, 26, 27, 28, 29, 30, 31, 32]. Among aromatic renewable phenols, vanillin [33, 34] has emerged as one of the most extensively investigated platforms for the synthesis of biobased epoxy thermosets owing to its aldehyde functionality and commercial availability. Eugenol [35, 36] has also attracted considerable attention because of its phenolic structure and natural abundance, leading to numerous mono‐, di‐, and multifunctional epoxy precursors. Conversely, isoeugenol‐derived epoxy systems [37, 38] remain less explored, while estragole‐based epoxy monomers [39] have only been occasionally reported. These multifunctional aromatic structures enable the preparation of epoxy thermosets exhibiting high glass transition temperatures and excellent mechanical performance, making them promising candidates for demanding applications such as lightweight aerospace structures, transportation, and advanced composites. Despite the significant advances achieved with these renewable aromatic platforms, the development of novel multifunctional polyphenolic epoxy monomers with improved thermomechanical performances remains an active field of research.

In recent years, research into thermosetting systems based on bio‐derived compounds has yielded impressive results, with many materials showing high potential for industrial applications [40, 41, 42, 43, 44, 45, 46]. Yet, a major limitation has been the reliance on conventional, some of them toxic, curing agents. The most effective aromatic diamines, such as diamino diphenyl methane (DDM) and diamino diphenyl sulfone (DDS), remain tied to petroleum‐based sources. While anhydrides offer partial solutions, their synthesis from renewable precursors is still challenging [47, 48, 49].

In this context, the amino acids represent a breakthrough—standing out as truly renewable, abundant, and inherently nontoxic compounds [50]. Unlike conventional amine hardeners, amino acids are natural building blocks of life and widely available from sustainable resources [50]. Their biocompatibility [51] not only mitigates health concerns but also enables access to functional biobased hardeners. Critically, amino acids possess multiple active hydrogen atoms (often three or more), allowing for the opening of several epoxy rings and thus facilitating exceptionally high crosslinking densities [52, 53]. A such designed network structure would compete or even surpass the performances of networks formed by conventional curing agents in terms of mechanical performance and durability [54, 55].

Recent studies have increasingly explored the use of amino acids as biobased curing agents for diglycidyl ether of bisphenol A (DGEBA) to produce high‐performance thermosets. For instance, Rothenhäusler et al. investigated the curing of DGEBA with aromatic amino acids such as tyrosine and tryptophan, reporting maximum glass transition values T _g ∼ 124 and 126 °C, respectively [56]. It is important to observe that these samples were prepared by weight ratio mixing, which may have led to suboptimal stoichiometric balances between reactive functionalities, ultimately leading to low T _g values. In contrast, another reported study employing molar ratios matched the functional groups; thus, the achieved thermal properties were improved, with the T _g of the tyrosine‐cured thermoset reaching up values ∼190 °C [57]. These results highlight the critical role of stoichiometric control in optimizing the thermosets’ design and so their properties. Furthermore, this work underlines the influence of amino acid structure, particularly the number of reactive functionalities, on network formation. Amino acids with a greater number of reactive centers can generate denser crosslink networks, thereby exhibiting high thermomechanical performances. However, the presence of aliphatic segments may modulate these effects; this was exemplified by L‐arginine (7 functionalities, T _g ∼ 161 °C) and L‐tyrosine (4 functionalities, T _g ∼ 188 °C), where the aliphatic moieties impacted the properties by lowering the T _gs [56, 57].

Walter et al. expanded on these findings by examining amino acids as curing agents for both petroleum‐based and partially biobased epoxy resins through kinetic studies. Their results demonstrated that epoxy networks cured using amino acids exhibited enhanced thermomechanical properties alongside improved sustainability, attributable to the nontoxic nature of amino acid hardeners [58].

Additionally, Shibata et al. combined biobased epoxy resins derived from sorbitol polyglycidylether (SPE) with amino acids to form matrix reinforced with chitosan nanofibers [58]. While these materials displayed promising properties linked to their structure–property relationships, they currently fall short of the mechanical performance required for high‐demand industrial applications, as the study primarily focused on elucidating curing mechanisms rather than application readiness.

Collectively, these studies underscore the growing potential of amino acids as sustainable, effective curing agents capable of producing thermosets with competitive thermal and mechanical performance, positioning them as key components in the advancement of biobased epoxy materials.

The aim of this study is the use of amino acids as curing agents for triglycidylether of resveratrol (TGER). As mentioned earlier, the amino acids offer remarkable advantages including nontoxicity, renewable sourcing, and diverse chemical functionalities, positioning them as ideal candidates for next‐generation biobased thermoset development. This approach not only promotes sustainable material design but also enables the production of high‐performance epoxy networks suitable for demanding industrial applications. The core objective is to systematically crosslink TGER with a variety of amino acids and thoroughly evaluate the resulting structure–property relationships. Special emphasis will be placed on how the presence of aliphatic versus aromatic moieties in the amino acid, alongside the number of active hydrogens available for crosslinking, impact the thermomechanical, mechanical, and stability characteristics of the cured materials. By these correlations, this study aims to advance the understanding of potential application of amino acid‐cured biobased epoxy thermosets, unlocking new possibilities for safe, high‐value, and sustainable polymeric materials.

A simple preparation protocol was employed, aligning with green chemistry principles by minimizing the number of synthesis steps. Slightly heated monomers yielded viscous, homogeneous mixtures without additional dispersants. Stoichiometric combinations of TGER and selected amino acids (L‐cystine, L‐glutamic acid, L‐lysine, L‐tyrosine, L‐tryptophan, L‐histidine, L‐phenylalanine, and L‐methionine) were prepared with an imidazole accelerator and then cured using a controlled temperature program to ensure complete crosslinking.

2. Materials and Methods

2.1. Materials

The compounds used for the epoxy monomer synthesis were resveratrol (from BLD Pharma), epichlorohydrin (ECH), and sodium hydroxide (from Merck Sigma‐Aldrich). Amino acids such as L‐cystine (Cys), L‐glutamic acid (Glu), L‐lysine (Lys), L‐tyrosine (Tyr), L‐tryptophane (Tryp), L‐histidine (His), L‐phenylalanine (Phe), and L‐methionine (Met) were employed as hardeners in the thermosetting formulation, and 1‐methylimidazole (1MIM) was used to initiate the copolymerization reactions. All these compounds were purchased from Merck Sigma‐Aldrich and utilized as received. The chemical structures of the compounds are depicted in Table 1.

TABLE 1.

The name, acronyms and structures of the TGER/amino acid formulation compounds and their physicochemical parameters.

Name	Chemical structure	Mass, g/mol	Melting temperature, °C	Functionality
Triglycidylether of resveratrol (TGER)		396 (EEW = 134)	70–74	2.96
L‐Cystine (Cys)		240	>240	6
L‐Glutamic acid (Glu)		147	205	4
L‐Lysine (Lys)		215	215	5
L‐Tyrosine (Tyr)		181	>343	4
L‐Tryptophane (Tryp)		204	—	4
L‐Histidine (Hys)		242	>282	4
L‐Phenylalanine (Phe)		165	>270	3
L‐Methionine (Met)		149	>284	3
1‐Methylimidazole		82	Liquid	—

2.2. Synthesis and Preparation

2.2.1. TGER Synthesis

In a three‐neck round‐bottom flask with a condenser, 15 eq. of ECH was added together with 1 eq. of resveratrol. ECH, which is widely used for the glycidylation of renewable phenolic compounds, can also be produced from biobased glycerol rather than fossil‐derived propylene. The use of renewable ECH therefore offers an important route for increasing the biobased content of aromatic epoxy monomers, although the renewable origin of the reagent must be verified for each specific commercial source. The mixture was heated, when reached 95 °C, a 12 wt.% of ethanolic solution with 3.2 eq. of NaOH was added dropwise to the reaction during 1.5 h. Then, the reaction was left for another 5 h at 95 °C. Thereafter, the mixture was filtered, washed, and dried with MgSO₄. The final product is a pink wax obtained at ∼92% yield.

¹H‐NMR (CDCl₃, δ in ppm): 7.43–7.41 d (2H), 7.03–6.99 s (1H), 6.92–6.90 d (2H), 6.89–6.95 d (1H), 6.68 s (2H), 6.42 s (1H), 4.25–4.22 dd (3H), 3.98–3.95 m (3H), 3.37–3.34 m (1H), 2.92–2.90 t (3H), 2.76–2.75 dd (3H) (Figure S1).

FTIR (cm⁻¹): 3462 (O—H), 3063 (C—H ar), 2998–2872 (C—H al), 1587 (C=C Ar), 1446 (C—H methylene), 1243 (C—O ether), 1172, 1126 (CH—O‐Ar ether), 907 (oxirane), 858, 803 (C—H Ar bending) (Figure S2).

2.2.2. Manufacturing of TGER/Amino Acid Thermosets

Stoichiometric mixtures of the two monomers—TGER epoxide and the respective amino acid—were prepared based on their active functional group equivalents (Table 1). The epoxy equivalent weight (EEW) of the synthesized epoxy monomer was determined by quantitative ¹H NMR spectroscopy in ${\text{CDCl}}_3$ using 1,1,2,2‐tetracloroethane as an internal standard from the integration of the epoxy proton signals relative to the standard proton signals. The experimentally determined EEW was used for all stoichiometric calculations. To accelerate crosslinking, 5 wt.% of 1MIM was added to the mixture. This relatively high 1MIM loading (5 wt.%) was chosen to compensate the lower intrinsic reactivity of amino acids compared with conventional aliphatic amines. Furthermore, because amino acids contain both amino and carboxylic acid functionalities, the curing process involves concurrent epoxy–amine and epoxy–carboxylic acid reactions. Under these conditions, 1MIM efficiently accelerates both reaction pathways, enabling complete network formation within the selected curing schedule.

The formulations were prepared without the use of organic processing solvents. The resulting solid mixtures were gently heated to 50–60 °C to obtain homogeneous, viscous blends before being transferred into silicone molds. Thermoset networks were then cured following a multistep thermal schedule: 1 h at 100 °C, 1 h at 120 °C, 2 h at 160 °C, and a final postcuring stage of 3.5 h at 180 °C.

The formulation processability was assessed by temperature‐dependent rheological measurements, demonstrating the existence of a suitable processing window prior to gelation. The rheological behavior of three representative systems (TGER/Tyr, TGER/Tryp, and TGER/Hys) was investigated, and the corresponding viscosity data are provided in the Supporting Information (Table S1 and Figure S12). At room temperature, TGER as well as the Tyr and Hys curing agents are solid, resulting in high initial viscosities. Upon heating above the melting temperature of TGER (≈60 °C), the viscosity decreases sharply as the formulations become fully molten and homogeneous. Within the processing temperature range of 70–90 °C, the formulations reach viscosities compatible with conventional solvent‐free thermoset processing, with values of approximately 35–40 Pa s for TGER/Tyr and 55–56 Pa s for TGER/Hys. These relatively low viscosities provide a practical processing window before the onset of gelation, facilitating operations such as casting, mold filling, and composite impregnation without the need for organic solvents.

2.3. Experimental Techniques

2.3.1. Differential Scanning Calorimetry (DSC)

DSC measurements were performed using a DSC 3 Mettler Toledo equipment operated via STAR software. Freshly prepared mixtures of unreacted formulations, with sample mass of 5–10 mg, were placed into 40 µL standard aluminum crucibles fitted with perforated lids under a nitrogen atmosphere. A heating profile was applied from 25 to 250 °C at a heating rate of 10 K/min.

2.3.2. Dynamic Mechanical Analysis (DMA)

DMA measurements were done using a Mettler Toledo DMA 1 apparatus, equipped with a three‐point bending clamp or dual cantilever clamp. During this analysis, the mechanical properties such as storage moduli (E′) and damping factor (tan δ = E″/E′) were evaluated. Rectangular specimens with dimensions of 50 × 8 × 2.5 mm³ were subjected to dynamic heating at 3 K/min from −150 to 300 °C and to an oscillating stress with a frequency of 1.0 Hz and an amplitude of 20 µm. The maximum temperature peak of the damping factor was attributed to the glass transition values (α transitions).

To further analyze the thermoset materials, crosslinking densities (υ) and the average mass of segments between crosslinks (Mc) were calculated using the equations derived from Flory (1953) [59] and Tobolsky (1960) [60] υ = E′/3RT and M _c = 3ρ RT /E′ where E′ represents the storage modulus of the thermoset in the rubbery plateau region at T _g + 50 °C, R is the universal gas constant, T is the absolute temperature, and ρ is the apparent density of the resin (g/cm³).

2.3.3. Hardness Tests

The hardness of the cured thermoset specimens was measured using a Zwick Roell 3116 device, in accordance with ASTM D2240 and ISO 48‐4. Rectangular samples with dimensions of 20 × 20 × 6 mm³ were tested. A standardized load of 50 N was applied. For each formulation, six independent indentations were performed at different positions on the sample surface to ensure repeatability, and the arithmetic mean and standard deviation were reported.

2.3.4. Apparent Density

Rectangular specimens measuring ∼50 × 8 × 4 mm³ were used from each system to evaluate their apparent density. Initially, the volume of each specimen was calculated, while the mass of each sample was measured with a Mettler Toledo ML3002T precision balance. The apparent density of the TGER/amino acid thermosets was calculated by dividing the mass of each sample to its corresponding volume. The final density values were obtained by averaging the densities of five samples from each formulation to minimize the errors.

2.3.5. Tensile Testing

The tensile tests were done using an Instron 34SC‐5 testing machine with a load capacity of ±5 kN. For the experiment monitoring, controling, and data evaluation, a BlueHill software developed by Instron (Norwood, MA, USA) was used. Dogbone Type V specimens were fabricated by casting the reactive liquid formulations into silicone molds. After curing, the plates were machined into Type V dog‐bone specimens. Prior to tensile testing, the specimen surfaces and edges were carefully finished by wet sanding with fine‐grit abrasive paper to remove machining marks and surface defects that could act as stress concentrators. The final specimen dimensions were 75 × 5 × 3 mm³, and all specimens were visually inspected before testing. Testing was conducted at 21 ± 3 °C in accordance with ASTM D638‐08 (Type V geometry) using wedge‐action grips at a constant crosshead speed of 5 mm/min. At least six specimens were tested for each formulation, and average values along with standard deviations were recorded.

2.3.6. Thermogravimetric Analysis (TGA)

The evaluation of thermal stability of the TGER/amino acid thermosets was done using a Mettler Toledo TGA 2 apparatus. Around 8–12 mg of sample was placed into 70 µl alumina crucibles. The temperature range used for the analysis was from 25 to 1000 °C. Thermogravimetric analyses were carried out under a nitrogen atmosphere (50 mL/min), followed by oxidation under synthetic air (50 mL/min) at a heating rate of 10 K/min. To estimate thermal resistance, the statistical heat resistance Index (T _s) was calculated according to Doyle [61]

$$T_{\mathrm{s}}=0.49\times [T_{5\%}+0.6\times (T_{30\%}‐T_{5\%}\left)\right]$$

where T _30% or T _5% represents the temperature at which the material loses 30% or 5% of its mass.

The LOI was estimated from the char yield at 800–850 °C (Cy₈₅₀) under an inert nitrogen atmosphere using the Van Krevelen empirical correlation [62]

$$\mathrm{LOI}\hspace{0.17em}=\hspace{0.17em}17.5\hspace{0.17em}+\hspace{0.17em}0.4\hspace{0.17em}\times \hspace{0.17em}{C}_{\mathrm{y}850}$$

2.3.7. Water Absorption (WA%)

The behavior of the obtained materials in an aqueous medium was evaluated in accordance with the ASTM D570‐98 standard. Rectangular specimens were first dried in an oven at 50 °C for 24 h under vacuum, cooled to room temperature in a desiccator to avoid moisture uptake from the atmosphere, and immediately weighed to obtain the initial dry mass. Then, the samples were weighed at room temperature to obtain the initial mass (W ₀) and placed in 20 mL vials filled with distilled water and maintained at a temperature of 25 ± 1 °C for 24 h. After this period, the samples were removed from the water, softly wiped with filter paper to remove excess surface water, and immediately reweighed (W _d). After the samples were weighed, they were immersed back into the distilled water. This weighing procedure was repeated every 24 h until the weight of the samples reached a constant value, indicating that it reached saturation with water. The WA% was calculated using the following equation

$$\mathrm{WA},\hspace{0.17em}\mathrm{\%}=\frac{W_{\mathrm{d}}-W_0}{W_0}\times 100$$

where W _d is the mass of the samples in their wet state while W ₀ the initial mass.

2.3.8. Gel Content (GC%)

The GC% of the thermosets was measured by solvent extraction method. Toluene was selected as the extraction solvent because it efficiently dissolves uncrosslinked epoxy monomers and low‐molecular‐weight oligomers while leaving the crosslinked epoxy network insoluble. Samples with dimensions of 10 × 10 × 2 mm³ were dried in oven for 24 h at 50 °C under reduced pressure (vacuum) until a constant mass was achieved, and then their mass (W ₀) was weighted. These dried samples were then immersed in toluene at room temperature. After a 72 h immersion period, the samples were removed from the solvent, dried again in an oven at 50 °C for 24 h, and then reweighted (W _f). The experiment was repeated three times to minimize the errors. The GC% of the thermoset materials was calculated using the following equation

$$\mathbf{GC}\hspace{0.17em}\mathbf{\%}=\hspace{0.17em}\frac{W_{\mathrm{f}}}{W_0}\hspace{0.17em}\times 100$$

where W ₀ is the initial mass of each dried sample and W _f the final mass after immersion in the organic solvent.

2.3.9. Biobased Organic Carbon Content (BOC) and Biobased Carbon Content (BCC)

The BCC of the designed materials was calculated using the method described by Pan et al.

$$\mathrm{BCC},\hspace{0.17em}\mathrm{\%}\hspace{0.17em}=\frac{\mathit{\sum }\mathrm{W}100\mathrm{\%}\cdot \mathrm{TC}100\mathrm{\%}}{\mathit{\sum }\mathrm{W}100\mathrm{\%}\cdot \mathrm{TC}100\mathrm{\%}+\mathit{\sum }\mathrm{W}0\mathrm{\%}\cdot \mathrm{TC}0\mathrm{\%}}\ast 100$$

where W100% is the mass of the 100% biobased constituent, TC100% is the total carbon content of the 100% biobased constituent, W0% is the mass of the 0% biobased constituent, and TC0% is the total carbon content of the 0% biobased constituent.

The BOC was calculated using the Norton and Devlin's method, applying the following equation

$$\mathrm{BOC},\mathrm{\%}\hspace{0.17em}=\frac{\mathit{\sum }\text{biobased carbon}}{\mathit{\sum }\text{biobased carbon}+\mathit{\sum }\text{petrobased carbon}}\ast 100$$

2.3.10. Temperature‐Assisted Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR‐FTIR)

The structural evolution during crosslinking reactions of freshly prepared formulations was monitored by in situ FTIR spectroscopy. These analyses were performed using a Nicolet iS50 FTIR spectrometer in ATR mode with a PIKE GladiATR single diamond system. The FTIR spectra were collected over the 4000–500 cm⁻¹ wavelength using 32 scans and a resolution of 4 cm⁻¹, with the air spectrum being recorded as the background before each sample analysis. To study the structural evolution of the reactive systems during crosslinking, fresh mixtures were tested under dynamic heating from 30 to 180 °C at a heating rate of 10 K/min, with the spectra being collected every 10 °C.

2.3.11. Green Metrics

The environmental implications of the employed reactions and processes were assessed through various green chemistry indicators. For this evaluation, both mass and molar parameters of the initial reactants, products, and auxiliary materials—such as catalysts, accelerators, solvents, and additives—were considered. The analyzed metrics included biobased content (BC), atom economy (AE), reaction yield, carbon efficiency (CE), reaction mass efficiency (RME), stoichiometry factor (SF), molar efficiency (Mol‐E), environmental factor (E‐Factor), process mass intensity (PMI), and mass productivity (MP) [63, 64]. The calculation methods and the formulas are detailed in Tables S3 and S4, Supporting information.

2.3.12. Rheometry

The rheological behavior of the epoxy formulations was investigated using an HR20 rheometer (TA Instruments, Waters)TA Instruments rotational rheometer equipped with a parallel‐plate geometry to evaluate the viscosity evolution as a function of temperature prior to curing. Approximately 1 g of the uncured formulation was placed between the plates (diameter: 25 mm, gap: 1 mm). Dynamic viscosity was measured at an oscillatory frequency of 1 Hz and strain of 1% while heating from 25 to 120 °C at a rate of 10 K/min. The viscosity–temperature profiles were used to assess the processability of the formulations and to identify the temperature window providing sufficiently low viscosity before the onset of gelation.

3. Results and Discussion

One of the aims of this study was to propose a simple method to prepare the TGER/amino acid thermosets by applying a reduced number of steps and without using any dangerous solvents during synthesis (except ECH) or preparation of the formulations. The mixtures for curing were done by mixing the slightly heated formulation components at 60 °C until the formation of a viscous paste. The mixture was homogeneous without any additional compounds to help dispersion. The simplicity of the applied methodology aims to align with green chemistry principles (Figure 1).

FIGURE 1.

The schematic representation of methodology used to prepare the thermosetting formulations and their crosslinking.

3.1. Reactivity Studies of the Biobased TGER/Amino Acid Thermosetting Resins

A series of eight TGER/amino acid formulations comprising the imidazole initiator were prepared. Their curing program was evaluated by following the reactivity behavior using DSC measurements. Four systems out of the eight prepared were selected for further investigation of their reactivity using temperature‐assisted FTIR analysis.

Figure 2 shows the DSC thermograms of the studied crosslinking systems. All thermograms exhibit narrow, intense peaks, except that of the TGER/Cis system, which presents a broad exotherm with two convoluted peaks. The interval of temperature and enthalpy of thermos‐curing are presented in Table 2.

FIGURE 2.

DSC thermograms of TGER/amino acids reactive mixtures during heating at 10 K/min.

TABLE 2.

DSC data obtained during crosslinking the TGER/amino acids formulations.

Sample	T onset, °C	T peak, °C	T endset, °C	ΔH, J/g
TGER/Cys	93	121	156	284
TGER/Glu	119	128	135	310
TGER/Lys	110	120	127	242
TGER/Tyr	117	125	130	300
TGER/Tryp	112	119	124	256
TGER/Hys	110	117	122	277
TGER/Phe	117	126	132	259
TGER/Met	116	126	134	350

Examining the structures of the selected amino acids (Table 1), all share the amino acid backbone, with distinct additional functional groups that influence their reactivity [65, 66]. As only TGER/Cys system presents two convoluted peaks, a correlation was searched with the cystine structure. Because Cys is the single diamino acid studied here containing a disulfide bond, the second peak could be due to some disulfide exchange reaction [67, 68]. Or, another hypothesis is the dual reactivity of the two amino acid functions of the Cys or a delayed reactivity due to the stability of the Cys molecule.

All the other TGER/amino acid systems have a unique and narrow polymerization peak, which has a low onset temperature (Table 2), ranking between 90 and 120 °C. The enthalpy of all studied systems has values between 240 and 350 J/g, suggesting a good TGER/amino acid reactivity toward crosslinking of epoxy with amine or acid functions from the amino acids. The curing temperature was comparable to that reported for other aromatic epoxy systems cured with amino acids. For example, the curing interval of DGEBA/amino acids was reported to be in the interval 100–160 °C with T _peak at ∼125 °C [56, 57].

Figure 3 presents the temperature‐assisted FTIR spectra of in situ polymerization of TGER/Lys formulation. A key observation is the decrease of the band at 3055 cm⁻¹ assigned to the N—H bond from the Lys amino groups and the modification of the band at 1425 cm⁻¹, which is attributed to the hydroxyl (—OH) group from the carboxylic acid. The decreasing of the amino band suggests the reaction of the amino groups with the oxirane rings, forming secondary and then tertiary amines and hydroxyl groups, as it is observed in the fingerprint region at 1040 cm⁻¹. As the temperature increases, the carboxylic band at 1425 cm⁻¹ diminishes, reflecting the reaction between the carboxylic group and oxirane cycle. The same behavior is observed in the other studied formulations (TGER/Tyr, TGER/Tryp, and TGER/Hys) (Figures S7–S9). Alongside the decrease of the OH band, a new low‐intensity band appears at 1730 cm⁻¹, characteristic of ester formation, proving the epoxy‐carboxyl reaction (Figures 3 and S7–S9).

FIGURE 3.

The thermal assisted FTIR spectra of the TGER/Lys during in situ polymerization.

Another notable modification in the FTIR spectra is the diminution of the oxirane band at 907 cm⁻¹ simultaneously with the band at 762 cm⁻¹, corresponding to the bending vibration of the 1,2‐disubstituted C—H group in the oxirane ring. To confirm that the 762 cm⁻¹ band belongs to the oxirane ring, the FTIR spectrum of TGER was compared with that of resveratrol (Figure S2), and this band was observed only in TGER. The decrease and eventual disappearance of these bands validate the consumption of oxirane rings and the formation of new amino‐ether linkages within the polymer. The development of amino—ether bonds is further supported by the disappearance of the peak at 1172 cm⁻¹ and the emergence of two broad bands in the region between 1150–1000 cm⁻¹, where the fingerprints of C—N bonds from tertiary amines and C—O bonds from ether groups are detected (Figures 4 and S3–S5). The modification of the previously described vibrational bands confirms the interaction of the amino and carboxylic groups and then additional reactive functional groups (OH or NH) with the oxirane cycles from TGER monomer.

FIGURE 4.

The TGER/Tyr polymerization mechanism considering the chainwise or step growth reactions, together with the secondary reactions.

A schematic representation of the reaction mechanism between amino acid and TGER is proposed in Figure 4. The TGER/amino acid mechanism tend to be a complex one, potentially involving two types of polymerisations. Tyrosine was chosen as example due to the three different functional groups on its structure: amino, acid, and hydroxyl. The chain‐wise anionic polymerization involves three stages (Figure 4a): initiation, propagation, and proton transfer with reformation of initiator [69, 70, 71]. The step growth polyaddition reaction can involve the three different reactions: the addition of carboxylic, phenolic, or amino group to the oxirane ring. Due to the formation of the hydroxyl groups during polyaddition, some side reactions of etherification can take place (Figure 4c). These side reactions may give additional crosslinking but also can consume a part of the oxirane rings.

Considering all reactions from the studied system, the final polymer will have a complex crosslinked structure, involving ester, amino, and ether bonds. The final structure of the crosslinked thermoset is represented in Figure 4d, and the physical appearance of the obtained thermosets after curing is presented in Figure S6. It should be noted that the photographs shown in the Supporting Information correspond to DMA specimens after surface abrasion. The opaque appearance observed for the cured resins should not be considered, by itself, an indication of inhomogeneity as a similar opaque appearance was also observed for the neat TGER homopolymer. Nevertheless, some samples exhibited local variations in color and visible air bubbles. These features may be related to the manual mixing of the paste‐like formulations, which can promote air entrapment and local variations in the distribution of the solid amino acid hardeners. Such effects may contribute to the observed optical variations in their aspects.

3.2. Thermosets Thermomechanical Investigation by DMA

DMA was employed using two different methods: the three‐point bending and the dual cantilever. This approach was adopted to obtain accurate values for both glass transition value and storage modulus of the samples. The maximum tan δ values were assigned to the α‐relaxation (T _α ) phenomenon related to cooperative chain motions and associated with macroscopic T _g. At temperatures below T _α , the mechanical behavior from the subglassy state to the glassy region is compared. As shown in Figure 4 and reported in Table 3, the TGER/amino acid thermosets are characterized by 2 secondary transitions in the glassy state: γ and β relaxations. In the glassy state, the physical behavior depends on cohesion and local molecular mobility. These relaxations appear to be due to residual segmental mobility at temperatures below T _g. The localized motions related to these secondary transitions involve rotation or crankshaft motions of short‐chain segments.

TABLE 3.

Thermo‐mechanical and network properties of the designed TGER/aminoacid thermosets.

System	E′ glassy, GPa, 25 °C	E′ rubbery, MPa	T α , T α′, °C	υ, mmol/cm3	Mc, g/mol	SD	Density, g/cm3
TGER/Cys	2.5	44.3	127, 187	3.344	281.38	81 ± 1.32	0.94
TGER/Glu	2.0	—	110, 165	—	—	79 ± 2.7	1.038
TGER/Lys	3.8	66.8	192	4.904	216.84	77 ± 3.3	1.063
TGER/Tyr	3.2	16	245	1.172	1028.07	85 ± 1.16	1.205
TGER/Tryp	2.3	31.6	222	2.390	464.19	80 ± 1.0	1.109
TGER/Hys	3.3	—	247	—	—	87 ± 1.04	1.217
TGER/Phe	2.1	59	110, 214, 245	4.315	253.82	76 ± 3.0	1.095
TGER/Met	1.2	15	115, 181	1.327	728.18	76 ± 2.13	0.967

All samples displayed a first peak around −60 °C, which is attributed to the γ‐relaxation [72, 73, 74]. The glass transition values of the developed thermosets were consistently high, ranging from 115 to 250 °C. Notably, the highest tan δ values were observed in the samples containing tryptophan (Tryp, 222 °C), tyrosine (Tyr, 245 °C), and histidine (Hys, 247 °C). These samples also displayed a β relaxation peak at ∼80 °C, associated with the hydroxy ether and phenyl propane moieties. The respective amino acids possess aromatic rings in their backbone, each featuring an active hydrogen atom, which enables the aromatic structures to participate directly in polymer crosslinking and establishes strong chemical bonds within the polymer network.

Comparing the TGER/Tryp and TGER/Phe systems, the last one presents four distinct peaks: at −60, 116, 214, and 245 °C (Figure 5 and Table 3). The first peak is assigned to the γ ‐relaxation, the second to the subglass β‐relaxation free movement of the branching diester segment within the polymer matrix, while the higher temperature peaks correspond to two glass transitions into the rubbery state. The transition at 214 °C arises from molecular hindrance induced by the phenyl group, which impacts network rigidity. Samples containing (Glu), (Lys), and (Met) exhibited broad tan δ transitions, typically with a peak or shoulder at around 110 °C. This could be related to their aliphatic moieties located between functional groups. TGER/Met shows a similar pattern to TGER/Phe, which can be attributed to the methyl thioether group lacking active hydrogens. We can notice that the TGER/Met, TGER/Phe, and TGER/Cys exhibit relatively broad α‐relaxation accompanied by a shoulder centered around 120 °C. This behavior suggests a broader distribution of relaxation times, which may originate from a less homogeneous crosslinked network or regions with different local crosslink densities. Consequently, the viscoelastic response of these thermosets cannot be fully described by a single T _g value. The progressive decrease in storage modulus observed at lower temperatures further supports the existence of additional molecular mobility prior to the main glass transition.

FIGURE 5.

Graphical representation of DMA results. (a) Damping factor (tan δ) versus temperature curves for the TGER/amino acid thermosets obtained in dual cantilever mode and (b) storage moduli (E′) in function of temperature curves obtained in 3 point bending mode.

These T _α values exceed previously reported amino acid‐cured epoxy thermosets (typically 100–170 °C) and approach the performance of high‐temperature aromatic epoxy networks cured with conventional aromatic diamines such as DDS. As example, reported studies combining DGEBA with tyrosine or tryptophane showed that the maximum glass transition values were around 124 and 126 °C [56]. In another study, the authors reported T _g not higher than 188 °C [57]. Compared with vanillin‐based epoxy thermosets cured with isophorone diamine (IPDA), which exhibit glass transition temperatures of approximately 120 °C [75] and 140, 177, and 200 °C for di‐, tri‐, and tetrafunctional epoxy vanillin monomers, respectively [76], the TGER/amino acid thermosets developed in this work generally display comparable or higher glass transition temperatures. This demonstrates that the highly aromatic and trifunctional TGER structure, combined with amino acid curing agents, enables the formation of rigid, highly crosslinked networks, exhibiting excellent thermomechanical stability. These results further highlight the potential of TGER as a renewable epoxy platform for the development of high‐performance, fully biobased thermosets.

The measured storage modulus (E′) values were notably high, exceeding 1 GPa, which is characteristic of amino‐epoxy thermosets. The highest values were recorded for samples with elevated glass transition, specifically those incorporating lysine (E′ ∼ 3.8 GPa), tryptophan (E′ ∼ 2.3 GPa), tyrosine (E′ ∼ 3.2 GPa), and histidine (E′ ∼ 3.0 GPa). This enhanced E′ is strongly correlated with the molecular structure of the compounds, particularly the presence of active hydrogen atoms on the amino acid backbones, which promotes increased crosslink densities and, as a result, greater capacity for mechanical energy storage. Compared with DGEBA/amino acid thermosets reported by Rothenhäusler et al. [56, 57] (DGEBA/Tryp, E′ ≈ 2.5 GPa; DGEBA/Tyr, E′ ≈ 3.2 GPa; DGEBA/Tryp, E′ ≈ 2.6 GPa; and DGEBA/Tyr, E′ ≈ 3.3 GPa), the TGER/amino acid thermosets developed in this work exhibit comparable storage moduli, indicating that replacing the conventional petroleum‐derived epoxy precursor with TGER does not compromise the stiffness of the resulting networks. In comparison with recently reported fully biobased vanillin‐derived thermosets based on di‐, tri‐, and tetrafunctional epoxy monomers cured with isophorone diamine (DiGEDVA/IPDA, E′ ≈ 1.9 GPa; TriGEDVA/IPDA, E′ ≈ 2.4 GPa; TetraGEDVA/IPDA, E′ ≈ 2.1 GPa) [76], the TGER/amino acid systems consistently exhibit higher storage moduli. These results demonstrate that the proposed TGER/amino acid platform provides stiffer crosslinked networks while maintaining a fully biobased composition, highlighting its potential for high‐performance structural applications requiring high rigidity and excellent thermomechanical performance.

Conversely, samples containing glutamic acid, phenylalanine, and methionine exhibited the lowest storage modulus values, reflecting differences in their amino acid backbone structures. While glutamic acid possesses an active hydrogen, its extended aliphatic chain reduces its thermomechanical performance. Although both Phe and Met contain inactive backbones, a significant difference was found in their storage moduli; the superior properties of Phe are attributed to its aromatic ring, which facilitates physical interactions like π–π stacking [77], thereby enhancing modulus, whereas Met lacks such interactions due to its methyl thioether side chain, resulting in lower mechanical properties.

The storage modulus in rubbery region was hard to determine because of the high transition temperatures of materials, which are close to their degradation temperatures. For several samples, it was examined and used to calculate the crosslink density. The values of crosslink density are ranking between 1 and 5 mmol/cm³ (Table 3). The higher crosslink density values were found for Cys (υ = 3.344 mmol/cm³), Lys (υ = 4.904 mmol/cm³), and surprisingly for Phe (υ = 4.315 mmol/cm³). Cys and Lys present the highest number of active hydrogens, explaining the high crosslinking density. In the case of Phe with the phenylic core, probably this increased density is given by the π–π interactions of the aromatic rings. Tryp (υ = 2.390 mmol/cm³) and Tyr (υ = 1.172 mmol/cm³) presented a lower crosslink density and consequently a higher molar mass of the segments between crosslinks (Table 3).

The stiffness of the systems was evaluated through Shore D hardness testing, and the obtained values show similar behavior as that in the DMA results. The results obtained for the designed thermosets are attributed to medium‐hard to hard materials, obtaining values between 75 and 87 SD (Table 3). The hard samples are considered TGER/Tyr (85 SD) and TGER/Hys (87 SD) as they presented the highest values, being in concordance with DMA results. The other samples are medium‐hard materials with values between 75 and 81 SD. The obtained results are similar to those of the commercial petroleum‐based resins (Abatron AboCast 8110‐4/AboCure 8110‐3 = 15–94 SA; Kohesi Bond KB 1048 FL = 60–70 SD; Master Bond EP51FL‐1ND‐2 = 50–60 SD). Besides, the designed thermosets presented low apparent density, ranking between 0.93 and 1.23 g/cm³ (Table 3), which categorize these materials useful for lightweight applications.

3.3. Tensile Tests

Four samples exhibiting superior thermomechanical performance in DMA analyses were subsequently selected for tensile testing. Table 4 summarizes the measured Young's modulus (E), tensile strength (σ), and fracture strain (ε) for the amino acid‐cured TGER systems. Young's modulus values ranged from 2.0 to 3.0 GPa, demonstrating substantial stiffness across the series. Notably, these values differ from the corresponding DMA storage moduli reported in Table 3. For example, TGER/Tryp and TGER/Tyr yielded similar Young's modulus values in tensile mode, whereas their storage modulus values differed by nearly 0.9 GPa under DMA conditions. Such discrepancies may be inherent to sample defects formed during curing. Specifically, the carboxylic groups on amino acids can participate in condensation reactions with hydroxyl functionalities formed postamino‐oxirane addition, leading to some water elimination. At elevated temperatures, volatilized water may create micro voids and disrupt the polymer network, negatively impacting macroscopic mechanical properties. Compared with conventional DGEBA/amino acid thermosets reported by Rothenhäusler et al. [56, 57] which exhibit Young's moduli of approximately 2.9 GPa for DGEBA/Tryp and 3.5–3.8 GPa for DGEBA/Tyr, the TGER/amino acid thermosets developed in this work display slightly lower stiffness. Nevertheless, their Young's moduli remain significantly higher than those reported for fully biobased di‐, tri‐, and tetrafunctional vanillin‐derived epoxy thermosets cured with IPDA (1.45–1.90 GPa) [76]. These results demonstrate that the proposed TGER/amino acid platform successfully combines a high renewable content with mechanical properties approaching those of conventional petroleum‐based epoxy systems while clearly outperforming several state‐of‐the‐art fully biobased epoxy networks.

TABLE 4.

Tensile properties of developed TGER/amino acids thermosets.

System	Tensile strain ε, %	Tensile stress σ, MPa	Young modulus E, MPa	Tensile stress at maximum force, MPa
TGER/Lys	1.18 ± 0.38	17 ± 3.17	1986 ± 227	17 ± 3.17
TGER/Tyr	1.27 ± 0.32	29 ± 4.87	2910 ± 387	29 ± 4.87
TGER/Tryp	0.9 ± 0.36	20.4 ± 5.04	2968 ± 795	20.4 ± 5.04
TGER/Hys	1.09 ± 0.39	22.7 ± 8.27	2641 ± 166	22.7 ± 8.27

Fracture strain values ranged from 0.9% to 1.3% and tensile strengths from 17 to 30 MPa (Table 4 and Figure 6), signifying relative brittleness throughout the sample set—particularly for TGER/Tryp (ε = 0.9%, σ = 20.4 MPa) and TGER/Hys (ε = 1.09%, σ = 22.7 MPa). When benchmarked against amino acid‐cured DGEBA epoxy systems reported in the literature, TGER‐based samples exhibit lower elasticity and strength; this outcome is closely tied to the intrinsic structure of Tri glycidyl ether of resveratrol. TGER incorporates a rigid double bond between two aromatic rings and possesses three epoxy functionalities, which increase the density of crosslinking sites and restrict molecular mobility.

FIGURE 6.

Representative stress–strain curves for the TGER/amino acids thermosets.

The result is enhanced rigidity and reduced elongation at break, manifesting as increased brittleness relative to systems based on DGEBA [56, 57, 78]. Moreover, imidazole accelerators serve not only to catalyze curing but also function as effective curing agents, thereby increasing the crosslink density of the thermoset [79]. As a result, the material acquires greater rigidity but exhibits increased brittleness, with reduced capacity to absorb or redistribute stress concentrations originating from structural defects.

Furthermore, the chemical architecture of each amino acid (Table 1). exerts a pronounced influence on network structure and properties. For example, the additional aromatic ring in Tryp can introduce steric hindrance, further limit the chain mobility, and impact mechanical performance. Conversely, the expectation that the aliphatic nature of Lys would increase fracture strain was not confirmed, likely due to the high crosslink density of TGER/Lys thermoset. Tyr yielded the most favorable results (ε = 1.27%; σ = 29 MPa; Table 4 and Figure 6), with its para‐positioned phenolic hydroxyl group providing additional crosslinking capability without introducing significant steric effects. The aromatic moiety of Tyr may also contribute additional intermolecular interactions, such as π–π stacking, further reinforcing the network. This comprehensive analysis underscores the critical interplay between resin structure, amino acid selection, and crosslinking chemistry in defining the mechanical landscape of biobased epoxy thermosets.

3.4. Thermal Stability of the Designed Bio‐Resins

The TGA curves reveal a primary degradation event of THER/amino acid resins, in an oxidative atmosphere, occurring above 200 °C (between 240–290 °C), accounting for approximately 5%–10% of the total sample mass (Figures 7 and S7). Compared with fully biobased di‐, tri‐, and tetrafunctional vanillin‐derived epoxy thermosets cured with IPDA, which exhibit T _5% values of 273, 292, and 275 °C, respectively [76], the TGER/amino acid thermosets display comparable or slightly higher thermal stability. These results demonstrate that replacing vanillin‐derived epoxy monomers with the highly aromatic TGER platform does not compromise the thermal resistance of the resulting networks. On the contrary, the excellent thermal stability achieved highlights the suitability of TGER/amino acid thermosets for applications requiring both a high renewable content and robust thermomechanical performance. This initial decomposition step is highly dependent on the structure of the amino acid curing agent as the epoxy resin matrix remains constant across all samples. The common component, resveratrol‐based epoxy (TGER), contains a double bond bridging two aromatic rings. This double bond is more reactive than others, and it increase the temperature sensitivity. Previous studies on epoxy resveratrol systems cured with diamines similarly reported initial mass loss (2%–3%) at relatively low temperatures. On the other hand, for the systems with DGEBA, thermal stability was found to increase with higher DGEBA content [80].

FIGURE 7.

Thermogravimetric analysis and derivative thermogravimetric curves of the TGER/amino acids thermosets under oxidative atmosphere during heating at 10 K/min.

Materials evaluated under nitrogen atmosphere exhibit much enhanced thermal stability, corroborating prior findings and highlighting the vulnerability of TGER systems to oxidative conditions [81, 82, 83].

The reduced thermal stability of these materials is influenced by the inherent lability of amino acid moieties. According to Ren‐Guo et al., the thermal resistance of amino acids is structure‐dependent, with aliphatic chains conferring lower stability than aromatic counterparts [84]. Consistent with this observation, samples incorporating aromatic amino acids such as tyrosine, tryptophan, and histidine (except for phenylalanine) display the highest onset temperatures for degradation (Figures 7 and S2 and Table 5). In contrast, cystine, phenylalanine, and methionine exhibit ∼15% mass loss around 245 °C, attributed to the cleavage of S—S bonds in cystine, S—C bonds in methionine, and C—C (Ar) bonds in phenylalanine. The S—S linkage in cystine is structurally dynamic and susceptible to thermal cleavage, while phenyl and thiomethyl side groups are weakly integrated as pendant substituents. For glutamic acid, increased mass loss is observed, likely due to its short aliphatic side chain, while lysine—bearing a longer aliphatic segment—exhibits greater stability. When compared to the nitrogen atmosphere TGA results (Figures 7 and S7), similar degradation profiles are observed, albeit at higher temperatures, especially for Lys, Phe, and Met. A second primary degradation stage occurs for all samples under both oxidative and inert (nitrogen) environments, corresponding to the major molecular breakdown and release of volatile fragments. Under oxidative conditions, the TGA curves show a third weight loss event, corresponding to the combustion of residual carbon following pyrolysis. This final combustion step is not present under nitrogen as the resulting polyaromatic char remains stable and resists further degradation. Quantitative analysis of N₂‐TGA data shows residual carbon (Cy₈₅₀) ranges from 26% to 42%, while the calculated limiting oxygen index (LOI) values are between 28% and 35% (Table S2).

TABLE 5.

The thermal degradation parameters of the samples under inert and oxidative atmospheres.

System	T 5%, °C		T 30%, °C		T s, °C		C 850, %	LOI
	Air	N2	Air	N2	Air	N2	N2	N2
TGER/Cys	240	320	377	406	157.88	182.08	37	32.38
TGER/Glu	290	307	388	405	170.91	179.24	37	32.34
TGER/Lys	252	255	395	403	165.52	168.46	28	28.74
TGER/Tyr	282	306	393	407	170.81	179.63	33	30.98
TGER/Tryp	263	278	397	404	168.27	173.26	29	29.3
TGER/Hys	281	300	407	421	174.73	182.57	42	34.38
TGER/Phe	239	257	367	380	154.74	162.09	26	28.1
TGER/Met	240	254	373	383	156.70	162.39	28	28.82

The statistical heat resistance index (T _s) reflects the physical thermal tolerance limit of a material and is calculated according to Ma et al. [85]. The calculated T _s values for the studied thermosets (Tables 5 and Figure S2) range from 183–188 °C in inert atmospheres and 154–175 °C in oxidative environments. Based on industrial classification criteria, these materials fall within the second (out of four) category for heat‐resistant polymers, which includes systems with T _s values between 100 and 200 °C. These results confirm that the designed biobased thermosets display appropriate thermal stability for moderately high‐temperature applications [86] The T _s values of the TGER/amino acid thermosets are comparable to, or even exceed, those reported for recently developed fully biobased di‐, tri‐, and tetrafunctional vanillin‐derived epoxy thermosets cured with IPDA, which exhibit T _s values of 153, 156, and 147 °C, respectively [76]. These results demonstrate that the TGER/amino acid platform provides excellent resistance to thermal degradation while maintaining a fully biobased composition. The high T _s values are attributed to the highly aromatic and multifunctional structure of TGER, which promotes the formation of thermally robust crosslinked networks suitable for demanding structural applications.

Overall, these results showed the pronounced effect of amino acid structure on thermal response and char development in TGER‐based thermosets, assisting in the rational design of advanced, bio epoxy materials.

3.5. Moisture Behavior, Gel Content, BCC, and BOC of the TGER/Amino Acid Thermosets

As shown in Table 6, after 24 h of immersion, the most hydrophilic samples—Cys, Lys, and Met—exceeded 2.5% water uptake, whereas Tyr and Tryp remained below 0.65%. After 7 days, water absorption increased further, reaching approximately 10% for Lys. The lower moisture content in Tyr and Tryp samples (under 1.4%) can be attributed to the involvement of aromatic structures in the crosslinked network, which not only reduces available hydrophilic sites but also facilitates hydrophobic π–π interactions, enhancing network stability and minimizing water access. Despite these initial trends, extended exposure for 7 days (Table 6) resulted in substantially higher water uptake across all samples, though Tyr and Tryp continued to demonstrate superior resistance compared to other systems. It should be noted that water content measurements at 7 days may have been affected by ongoing sample degradation, particularly in more hydrophilic systems. Nevertheless, GC % analysis indicated values approaching 100%, suggesting minimal presence of unreacted small molecules in the thermoset matrix. This level of water uptake likely corresponds not to chemical degradation but rather to physical defects among crosslinked polymer chains.

TABLE 6.

Biobased carbon content and biobased organic carbon content, water absorption, and gel content of the TGER/amino acid cured thermosets.

System	BCC, %	BOC, %	WA, %		GC (25 °C), %
			24 h	7 days
TGER/Cys	95.39	98.96	2.51	6.88	99.96
TGER/Glu	95.63	98.97	1.76	4.74	100.11
TGER/Lys	95.68	98.95	3.21	10.14	99.61
TGER/Tyr	95.83	98.98	0.65	1.39	99.93
TGER/Tryp	95.90	98.99	0.50	1.31	100.00
TGER/Hys	95.14	98.88	0.43	1.58	99.93
TGER/Phe	95.90	98.95	0.91	3.26	99.96
TGER/Met	95.48	98.91	5.53	11.63	99.89

Ultimately, due to the intrinsic hydrophilicity of amino acids, some degree of moisture uptake in these thermosets is inevitable. While the formation of physical bonds such as hydrogen bonding may enhance initial toughness, these interactions remain susceptible to disruption by water, highlighting the fundamental role that curing agent structure plays in governing the long‐term durability and moisture resistance of biobased epoxy systems.

The BOC (%) and BCC (%) of the synthesized materials were determined using the method described in Section 2.4.9 [47]. The analysis yielded exceptionally high values for both BOC and BCC, ranging from 95% to 99%, as summarized in Table 7. Such elevated BCC is a strong indicator of the environmental friendliness of these materials, reflecting their potential to significantly reduce reliance on petroleum‐derived resources. Notably, the substantial contribution of amino acids—which are naturally occurring, fully renewable compounds—ensures that all constituent atoms are of biogenic origin. This high level of biobased carbon underscores the sustainable character of the developed thermosets and positions them as promising candidates for sustainable advanced material applications.

TABLE 7.

Green parameters for monomers compared with others declared in the literature.

Monomer	Triglycidyl ether of resveratrol (TGER)	Diglycidyl ether of bisphenol A (DGEBA)	Monoglycidyl ether of vanillin (MGEV)
Yield, %	92.17	70	67.2
AE, %	78.29	82.36	85.1
CE, %	92.25	32.67	36.91
RME, %	63.29	19.32	22.17
PMI	8.72	5.17	18.27
MP, %	11.47	19.32	5.47
SF	7.10	1	4.05
Mol E, %	12.82	5.38	0.71
E‐factor	11.47	3.21	15.93
E‐factor (solvent recover)	4.18	—	—
BC, %	100.00	0	72.69
Reference	This work	[87]	[88, 89]

3.6. Green Metrics‐Based Sustainability Assessment

To assess the environmental impact of the synthesized TGER monomer, the green chemistry metrics were calculated for it and compared to those of the other two monomers, di‐glycidyl ether of bisphenol A (DGEBA) and monoglycidyl ether of vanillin (MGEV). The compared data are presented in Table 7, and the calculation methods and formulas are given in the Supporting Information, Tables S3 and S4.

The obtained values for the green metrics reveal both advantages and disadvantages of the employed synthesis. The yield, AE, CE, and RME showed high values—particularly yield, CE, and RME—compared with the other two syntheses (DGEBA and MGEV) (Table 7). These values suggest a high conversion of raw materials into the desired product with minimal losses.

The PMI and SF reflect the excess of additional components used in the reaction, such as solvents and catalysts. Ideally, these values should be as close to 1 as possible; however, when large amounts of auxiliary materials are required, both parameters increase. In our case, the PMI and SF were relatively high due to the significant amount of solvent used during the process. Nevertheless, compared with MGEV, the PMI is still lower, considering that the TGER product incorporates three oxirane rings in a single molecule.

The MP is directly influenced by PMI and, consequently, by the amount of auxiliary components involved. Its value remains higher than that of the MGEV synthesis but lower than that of DGEBA. A similar trend is observed for the E factor, another indicator of process greenness. The high solvent usage increased its value considerably, but since the solvents were recovered and not treated as waste, the corrected E factor decreased to 4.18, approaching that of DGEBA.

The Mol E is an additional favorable indicator of a greener synthesis, representing the molar ratio of the desired product to the total moles of reactants used. A higher value indicates a more sustainable process, which in our case exceeded 12%, surpassing the examples presented in Table 7. Moreover, unlike the other two compounds, the monomer used in our synthesis is entirely biobased (100%), further enhancing the overall green profile of the process.

For a better view on the TGER monomer green parameters compared to that of DGEBA and MGEV, Figure 8 shows a radar graph comparing all three monomers.

FIGURE 8.

The graphical representation of the green metrics for TGER, MGEV, and DGEBA synthesis.

4. Conclusion

This study demonstrates the successful development of high‐performance, fully biobased thermosets by curing TGER with various amino acids as sustainable hardeners. The innovative approach leverages in the benign design of thermosetting formulations in a single step process. Then, the natural abundance, nontoxicity, and multifunctionality of amino acids result in polymer networks with excellent environmental credentials and advanced material properties.

The designed TGER/amino acid thermosets exhibit impressive thermomechanical behavior, notably high glass transition values (T _g up to 247 °C), exceptional storage moduli (up to 3.6 GPa), and Shore D hardness values (up to 87 SD), positioning them as strong candidates for demanding applications that require both mechanical robustness and lightweight design. The comprehensive analysis highlights the critical impact of amino acids molecular structure on crosslink density, network formation, and macroscopic properties. Aromatic amino acids such as tyrosine, tryptophan, and histidine delivered good thermal stability and mechanical performance, with tyrosine showing the best balance between stiffness, strength (σ = 29 MPa), and toughness (ε = 1.27%).

In addition, all studied thermosets demonstrated high BCC (95%–99%), confirming their potential to reduce reliance on petroleum‐derived resources and thus significantly improve environmental sustainability. Water uptake tests revealed that samples cured with aromatic amino acids possess notably enhanced hydrophobicity and durability compared to those with aliphatic or sulfur‐containing structures.

Although the developed thermosets exhibited excellent thermomechanical performance, additional investigations of fire behavior (e.g., UL94, LOI, or cone calorimetry) will be required before considering their implementation in applications subject to stringent fire safety regulations.

Overall, this work provides substantial evidence that amino acid‐cured resveratrol epoxies can compete with traditional petroleum‐based systems in key performance metrics while enabling safer, greener, and more versatile thermoset materials. The findings not only validate the rational design of biobased epoxy networks tailored through amino acid selection but also chart a path forward for next‐generation sustainable composites in advanced engineering and industrial domains.

Funding

This study was supported by the Air Force Office of Scientific Research (FA8655‐23‐1‐7024) and Agence Nationale de la Recherche (ANR‐15‐IDEX‐01).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.