Synthesis of novel bio-based amines from vanillin and guaiacol for high performance epoxy thermosets

Abstract

In this work, a novel bio-based diamine was successfully synthesized from lignin-derived vanillin. This marks the first report of the synthesis and characterization of 1,3-bis(aminomethyl)-4,5-dimethoxybenzol (Dimethoxy-MXDA). Furthermore, 2,4-bis(dimethylaminomethyl)-6-methoxyphenol (Methoxy-K54) has been successfully synthesized in one step from bio-based guaiacol, another major lignin derivative. Dimethoxy-MXDA was used as a bio-based amine hardener in the synthesis of an epoxy thermoset polymer and showed similar to superior performance when compared to a petrol-based reference system containing the industrially common m-xylylenediamine (MXDA) hardener. In particular, the post-cured samples of the bio-based system outperformed the petrol-based benchmark material in all properties. Furthermore, the bio-based hardener Methoxy-K54 showed comparable properties to the petrol-based analogue 2,4,6-tris(dimethylaminomethyl)phenol (Ancamine® K54).

---

Green foundation

1. Advancement of green chemistry: our work contributes to the field of green chemistry by introducing novel synthetic pathways for bio-based amine hardeners and accelerators, as well as developing high performance bio-based epoxy thermosets.

2. Specific green chemistry achievement: our synthetic approach offers a sustainable alternative for producing amine hardeners and accelerators from bio-based feedstocks. Moreover, our bio-based epoxy thermosets exhibit comparable or superior mechanical properties to petroleum-based benchmarks.

3. Future directions: future research might explore other synthetic routes from our key intermediate 5-formyl vanillin for other tailored synthetic building blocks. Another important focus should be on scaling up the synthesis processes in accordance with green chemistry principles.

Introduction

Epoxy polymers are thermosetting polymers with excellent physical and chemical properties, such as high mechanical strength, thermal resistance and good processability, and have become indispensable to a wide range of applications and industries. The thermoset polymer is produced by reacting a glycidyl ether-functional epoxy resin with a hardener component, such as phenols, thiols, carboxylic acids and their anhydride derivatives, and most importantly for room temperature curing of primary and secondary amines.¹ During this so-called curing reaction, a three-dimensional cross-linked polymer network is formed, which, together with conformationally stiff aromatic moieties in the repeating units, are responsible for the excellent material properties.² Due to their thermoset nature, conventional epoxy polymers form very stable materials and generally cannot be recycled, unlike typical thermoplastic polymers. Additionally, the global use and prevalence of epoxy polymers is ever growing. Since the vast majority of epoxy thermosets are derived from petrochemical raw materials, renewable alternatives bear immense potential for sustainability improvement.³ Corresponding to the global desire to reduce the carbon footprint, recent interest in the use and study of bio-based raw materials to prepare such materials is increasing and the number of contributions on the subject is growing rapidly. When investigating bio-based raw material groups as synthetic starting points for the hardener and accelerator components of epoxy systems, there are five major bio-sources that have to be considered: plant oils, saccharides, polyphenols and phenols, natural resin and lignin.⁴ The latter recently gained a lot of interest due to an improved biorefinery process⁵ that allows depolymerization of lignin to yield vanillin⁶ and guaiacol⁷ that are highly defined, functional aromatic structures which turned out to be an ideal starting point for the synthesis of novel bio-based high performance epoxy systems.^8,9

Apart from these publications, various reviews^10,11 and patents^12–15 have been published on the use of vanillin and guaiacol as starting materials for the synthesis of novel building blocks for the epoxy resin component of the epoxy thermosets. Presumably, the most straightforward epoxy resin synthesis results from the reduction of vanillin to vanillyl alcohol and subsequent transformation to the diglycidyl ether of vanillin alcohol (DGEVA), as demonstrated by various groups^16–18 and commercialized as a research chemical.¹⁹ Furthermore, the production of analogues to the diaromatic structure of bisphenol-A-digylcidyl ether (BADGE), the most prevalent epoxy resin component, served as a source of inspiration for many research groups. Studies towards bio-based BADGE analogues encompassed the enzymatic coupling of vanillin²⁰ and dimerization using aldol condensation of cyclohexanone or cyclopentanone.²¹ Other diverse synthetic approaches including Williamson ether synthesis using dihalogen alkyls followed by reduction to the dialcohol and subsequent diglycidyl ether formation, pinacol coupling via electrochemical pathways as well as electrophilic aromatic substitution and acetal formation using polyfunctional alcohols were explored.¹⁰

While modification of the bio-based epoxy resin component attracted much attention, the field of vanillin-derived amine hardeners has remained largely untouched. A notable exception presents the conversion of DGEVA with ammonia into a primary diamine with β-hydroxyl groups.^22–24 A possible reason why bio-based amines are underexplored is that amine reactivity desired in amine-epoxy systems is less well known and additionally, the high resulting viscosity is challenging. Exploring bio-based amine components in epoxy thermosets not only closes an existing gap in the literature, but also satisfies a need from industry to use green alternatives in their existing product portfolio. This aspect is signified by a recent patent filing of our research group.²⁵ A primary goal of our work was to provide a novel synthetic pathway to access lignin-derived amine hardeners and accelerators. To this end, we also aimed at producing bio-based epoxy thermosets, investigating their material properties and performance and comparing them to the corresponding petrol-based benchmarks. To achieve our goal of transforming vanillin into a suitable amine hardener, the aldehyde and alcohol functional groups served as a linchpin for the Duff reaction.^26,27 Subsequently, functional group interconversion would finally yield the desired amine functionality. The transformation of guaiacol towards a hardener component is inherently more difficult, because guaiacol does not possess such a versatile aldehyde functional group. Nonetheless, it was envisioned that the Mannich reaction enables the introduction of two tertiary amines.²⁸ The final molecule would then serve as an accelerator for the curing reaction of epoxy resin thermoset polymers. Our methodology also encompassed a commitment to green chemistry principles to afford the most sustainable synthetic pathway.²⁹

The novel amine hardener synthesized starting from vanillin, 4,5-dimethoxy-1,3-benzenedimethanamine (Dimethoxy-MXDA), can be compared to a structurally similar petrol-based analog, m-xylylenediamine (MXDA). Furthermore, the guaiacol-derived accelerator, 2,4-bis(dimethylaminomethyl)-6-methoxyphenol (Methoxy-K54), is structurally similar to 2,4,6-tris(dimethylaminomethyl)phenol, which is commercially available as Ancamine® K54. A comparison between these novel bio-based amine hardener and accelerator components with their petrol-based structural analogs makes sense not only from a scientific point of view, but also from an economic one, since both petrol-based components are very commonly used in contemporary epoxy systems (Fig. 1).

Fig. 1 Comparison of the bio-based hardener molecule dimethoxy-MXDA (3) with its petrol-based analogue MXDA (top) and comparison of the bio-based accelerator methoxy-K54 (4) with its petrol-based analogue Ancamine® K54 (bottom).

Experimental

Preparation of Dimethoxy-MXDA (3)

The synthesis route, illustrated in Scheme 1, comprises 4 reaction steps, starting from lignin-derived vanillin. The Duff reaction introduces a second aldehyde functional group at the ortho-position leading to molecule 1. Then, methylation transforms the phenolic alcohol into a methoxy group, yielding 2. Finally, both aldehydes of molecule 2 are converted to dioximes and reduced to the primary amines to afford the final product Dimethoxy-MXDA (3). After each reaction step isolation and purification were conducted to afford pure intermediates and product.

Scheme 1 Synthesis of dimethoxy-MXDA (3) starting from vanillin by Duff reaction, subsequent methylation reaction, oxime formation and final hydrogenation.

Preparation of Methoxy-K54 (4)

As visualized in Scheme 2, the Mannich reaction allows the introduction of the two tertiary amines starting from guaiacol. This one-step reaction encompasses the transformation of guaiacol to afford the final product Methoxy-K54 (4). Further information on used materials, detailed synthetic procedures and compound characterization can be found in the ESI.†

Scheme 2 Synthesis of methoxy-K54 (4) from guaiacol by Mannich reaction.

Epoxy thermosets

To investigate the material properties of the novel bio-based hardener and accelerator, epoxy thermosets were produced.

Formulation

Detailed information on the formulation can be found in the ESI.† For the production of dog-bone specimens, an epoxy model resin (EP-bulk) was used (Table S1†). For the thin film plates, an epoxy model coating (EP-coat) was used (Table S2†). For the comparison of accelerator properties, an accelerated epoxy model coating (EP-coat-accelerated) was used (Table S3†). The formulation of epoxy thermosets was based on a 1 : 1 stoichiometry of the reactive groups, i.e. the ratio of EEW and AEW represented the molar ratio.

The epoxy equivalent weight (EEW) was confirmed by titration. All materials were prepared by first calculating the functional group equivalents. The synthesized amine hardener (3) contained two primary amine functionalities. Thus, each of the active amine hydrogens can react with an epoxide. Consequently, the number of active hydrogens per primary amine is equal to two. To calculate the necessary amount of hardener for resin formulation of 100 g, the following eqn (1) and (2) were used:

Dimethoxy-MXDA (3) with an AEW of 54.04 g val⁻¹ determined by titration (0.1 N HClO₄ in acetic acid against crystal violet) was used as the hardener component in the bio-based epoxy thermoset. As a benchmark for the bio-based amine hardener, the petrol-based amine hardener MXDA with an AEW of 34.00 g val⁻¹ was used. The guaiacol-based accelerator (Methoxy-K54 (4)) was used as a catalyst (1.2 wt%) with the amine hardener N-benzyl-1,2-ethandiamine (B-EDA) with an AEW of 50.10 g val⁻¹. As a benchmark for the bio-based accelerator, Ancamine® K54 was used instead of molecule 4.

The reported AEWs were confirmed by titration. Hardeners and resins were homogenized using a bi-axial centrifugal mixer (speedmixer™ DAC 150, Flack Tek Inc.). Subsequently, their mixing viscosities were determined.

Production of specimens for mechanical tests

To test the mechanical properties, dog-bone-shaped specimens (type 5A, thickness 2 mm) were prepared from the EP-bulk formulation as described (Table S1†). For that purpose, the reaction mixture was poured into a silicone mold. After 24 h curing at room temperature (RT), all specimens were released from the mold. A set of specimens was then stored at RT for a total of 7 days (referred to as RT). Another set of specimens was post-cured at 120 °C for 24 h after release from the mold and then stored at room temperature for 24 h (referred to as PC). Finally, a set of specimens was also post-cured at 120 °C for 24 h after being released from the mold, then submerged in water for 5 days at RT, and then taken out of the water and stored at RT for another 24 h (referred to as H₂O).

Preparation of thin film plates

Thin film plates (thickness: 0.5 mm) were prepared from the EP-coat formulation as described (Table S2†) and cured at varying temperatures and humidities (23 °C at 50% r.h. and 8 °C at 80% r.h.) to determine the performance of the epoxy thermoset polymer regarding blushing, yellowing and the development of pendulum-hardness according to König during the curing process.

Accelerator properties

Thin film plates (thickness: 0.5 mm) have been prepared from the EP-coat-accelerated formulation as described (Table 3) and cured once at 23 °C and 50% r.h. for 14 days and once for 7 days at 8 °C at 80% r.h. followed by 7 days at 23 °C and 50% r.h. Over the course pendulum-hardness according to König was measured to determine the performance of the accelerator.

Results and discussion

Synthesis of Dimethoxy-MXDA (3)

Herein, two synthetic pathways have been successfully identified for the synthesis of novel bio-based aromatic amines starting from major products of lignin depolymerization, namely vanillin and guaiacol. Another goal of this work was to investigate the properties of novel bio-based amines for use as hardeners and accelerators in reactive epoxy resins. Finally, substituting commonly used petrol-based components in epoxy thermosets with bio-based alternatives can give significant sustainability impact and is a central part of political initiatives aimed at a more sustainable and circular economy.³⁰

The synthesis of a novel, bio-based aromatic amine starting from vanillin comprises 4 steps and is depicted in Scheme 1. Vanillin displays single aldehyde, alcohol and methoxy functionalities on a single aromatic ring structure. Introduction of an additional aldehyde functional group through the Duff reaction provided a versatile linchpin for the rest of the synthesis. Through activation and direction from the electron-donating phenol group, the free ortho-position is selectively substituted. Hexamethylene-tetramine (HMTA) served as the source of the carbonyl group. The Duff reaction conditions were initially based on a publication by Yue.³¹ Optimization of the reaction conditions and adaption to vanillin as a substrate resulted in a yield of 96%. A downside of the Duff reaction is the use of trifluoroacetic acid (TFA). Although it allows for the highest yields achieved, it is toxic, has a possibly negative impact on the environment and is likely to be more tightly regulated under a future legislation banning per- and polyfluoroalkyl substances (PFAS) in Europe. To improve this methodology, the conditions were adjusted according to green chemistry principles. We were successful in designing a “greener” version of the Duff reaction by halving the amount of TFA and substituting it with glacial acetic acid (AcOH), while obtaining acceptable yields (66%). Tenfold reduction of TFA and replacement with AcOH still yielded 44% of the desired aldehyde. Full replacement of TFA resulted in 32% yield, showing that the reaction can still proceed, albeit with considerably less yield. Distillation, including azeotropic distillation, and reuse of TFA were attempted but remained unsuccessful. This marks a significant improvement of the Duff reaction towards a greener type of chemistry.

The resulting 5-formyl vanillin (1) was then subjected to methylation of its phenolic group yielding 4,5-dimethoxy-1,3-benzenedicarboxyaldehyde (2). Methylation was performed to decrease molecular interactions and thus the viscosity in the final hardener molecule. The reaction was carried out under the conditions used by Saiz-Poseu,³² by using dimethyl sulfate (DMS) as an alkylating agent. DMS served as a very potent alkylating agent, allowing the reaction to proceed in a clean manner and in very high yield (88%). Unfortunately, DMS is highly toxic. To improve this synthesis, dimethyl carbonate (DMC) could act as a greener alternative to DMS in future experiments.

To afford our final amine hardener, the conversion of the aldehyde functionalities was the next step. The aldehyde functional group acts as a linchpin for a transformation into the corresponding oxime. In a similar fashion to a procedure in the literature,³³ the oxime forming reaction step corresponds to the reaction of the aldehyde functions with hydroxylamine in the presence of sodium acetate (yield: 80%).

As a next and final step, the formed oximes needed to be reduced to the corresponding free amines. Reduction via heterogeneous catalytic hydrogenation afforded the final amine hardener product 1,3-bis(aminomethyl)-4,5-dimethoxybenzol, herein called Dimethoxy-MXDA (3). The reaction was performed for 6 h at 80 °C under 80 bars of H₂ pressure in a pathway adapted from the literature³⁴ using RANEY® nickel as the catalyst (yield: 92%).

To our knowledge, this work marks the first proven synthesis and characterization of 1,3-bis(aminomethyl)-4,5-dimethoxybenzol. A concise summary of the characterization of Dimethoxy-MXDA (3) is provided in Fig. 2. The structures of the intermediates were confirmed by FT-IR, ¹H-NMR and ¹³C-NMR spectroscopy and the purity checked using GC spectra (Fig. S1–S12†). For the novel bio-based amine hardener 3, the characterization was performed similarly (Fig. 2a–d and S13–S16†). Additionally, the attribution of the NMR signals was also confirmed by HSQC NMR spectroscopy (Fig. 2e–f). The ¹H NMR spectrum (Fig. 2a) shows the characteristic broad signal belonging to the amine protons at around 1.81 ppm, and signals at 3.66 and 3.67 ppm correspond to the two methylene groups next to the amines. The signals at 3.70 and 3.79 ppm stem from the two methoxy groups, while the signals at 6.90 and 6.92 ppm represent the protons on the aromatic ring. For the ¹³C NMR spectrum, both the chemical shifts and the number of signals are in agreement with the chemical structure of the product (Fig. 2b). In the FTIR spectrum the characteristic N–H stretch appears at 3367 cm⁻¹ together with other characteristic IR vibrations (Fig. 2c). The GC spectrum shows that the target molecule was afforded in pure form (Fig. 2d). The two-dimensional ¹H–¹³C HSQC NMR spectrum unequivocally confirms the structure by correctly providing the correlations of the aromatic proton–carbon (Fig. 2f, blue box) couplings and the aliphatic proton couplings, respectively (Fig. 2f, green box).

Fig. 2 (a) ¹H NMR spectrum, (b) ¹³C NMR spectrum, (c) FT-IR spectrum, (d) GC spectrum, (e) molecular structures with labelled carbon and hydrogen atoms and (f) ¹H–¹³C HSQC spectrum of dimethoxy-MXDA (3).

Moreover, the successful optimization of the multi-step synthesis and improvement of the reaction conditions according to green chemistry principles displays a significant gain in chemical efficiency. More about the intended use of Dimethoxy-MXDA (3) as a bio-based amine hardener can be found in later sections of this paper.

Synthesis of Methoxy-K54 (4)

The synthesis of the second target molecule starts from guaiacol, another lignin derivative (Scheme 2). In contrast to vanillin, guaiacol does not possess an aldehyde functionality. Despite this absence, we were able to introduce the desired diamine functionality by employing a Mannich reaction. By using dimethylamine and paraformaldehyde as reagents, we afforded Methoxy-K54 (4) in a simple and efficient one-step synthesis.

Tertiary amines or phenols can act as accelerators. Therefore, the phenolic functionality is beneficial for activation and was deliberately retained in the structure. It is well known that phenolic hydroxyl groups can form hydrogen bonds with the oxygen of the epoxy group which is consequently more susceptible to be attacked by the nucleophilic amine. The phenolic alcohol acts as a very efficient hydrogen bond donor that can (similar to the tertiary amine) accelerate the hardening reaction of epoxy compounds and amines. Since it will be used in comparatively smaller amounts than the amine hardener itself, a possible increase in viscosity due to the influence of the phenolic group becomes negligible. The minor side product that formed in the Mannich reaction is inevitable, as it is the product with just one added amine group. Since it also acts as an accelerator and is present in small quantities, it was seen as negligible too, and it was used without further purification.

Like the amine hardener molecule Dimethoxy-MXDA (3), the amine accelerator molecule Methoxy-K54 (4) was also characterized using ¹H NMR and ¹³C NMR spectroscopy (Fig. S17 and S18†). The chemical shifts and integrals for the ¹H NMR spectrum and the number of signals and chemical shifts in the ¹³C NMR spectrum are all in agreement with the target chemical structure.

Synthesis of thermosets

Another goal of this work was to produce innovative epoxy thermoset polymers using the novel bio-based amine hardener 3 and the bio-based accelerator 4. To compare the performances of the novel molecules, a comparison with purely petrol-based reference systems was set up. The epoxy component in both systems was BADGE. MXDA was chosen as a reference hardener, because it is similar in structure to 3.

First, we compared the properties of the two hardeners (Table S4†). The bio-based hardener has a higher molecular weight due to the additional methoxy groups compared to MXDA. This also translates into a larger AEW. Interestingly, the bio-based hardener 3 is almost 20 times more viscous than MXDA, thus confirming our earlier decision to mask the phenolic functionality with a methyl ether group to decrease the viscosity. When comparing the mixing viscosity, i.e. the viscosity just after combination of the hardener and epoxy resin, the difference in viscosity is only two-fold, allowing for good processability in both cases. Surprisingly, the gelation time in the bio-based system is significantly shorter (1–2 h) compared to that in the petrol-based system (>4 h). Not only did the bio-based system harden faster, which hints at faster curing kinetics for the bio-based hardener cured system, but it also showed significantly less skin formation. This contributed to the uniform surface appearance for the bio-based system (Fig. 3, right). In contrast, the petrol-based system showed evidence of surface blushing, as is typical of less sterically hindered amines such as MXDA (Fig. 3, left). The synthesized epoxy thermosets are shown in Fig. 3.

Fig. 3 Thermosets from the petrol-based hardener (left) and bio-based hardener 3 (right).

Characterisation of epoxy thermosets

Next, the mechanical properties of the cured epoxy thermoset polymers were investigated. A direct comparison of the tensile properties of the bio-based hardener cured system showed that the performance is at least similar if not better when compared to the petrol-based reference system (Fig. 4a–c). Over the course of the different curing and storage conditions (RT, PC, and H₂O), both the bio-based and petrol-based systems followed the same trends, namely, that prolonged exposure to higher temperature led to post-curing (PC), i.e. an increase in crosslink density for both systems and that both systems retained their performance after intensive water exposure (H₂O).

Fig. 4 (a) Tensile strength, (b) elongation at break, (c) Young's Modulus and (d) glass transition temperature of epoxy thermosets under different curing and storage conditions. (e) Dynamic mechanical analysis and (f) stress–strain curves of epoxy thermosets after curing.

This trend manifests when looking at the elongation at break (Fig. 4b), where a significant increase of 4–6% can be observed when changing from RT to PC or H₂O conditions. The Young's Modulus behaves similarly for both systems with the highest value of 3320 MPa for the bio-based system cured at room temperature (Fig. 4c). The tensile strength at RT is slightly lower for the bio-based system, but again reaches the highest value of 72.9 MPa after post-curing (Fig. 4a). Finally, the glass transition temperature (T_g) is found to be slightly higher in the bio-based system, ranging from 62 °C (RT) to 88 °C when post-cured and 82 °C when stored in water (Fig. 4d). A collection of all mechanical data can be found in Table 1.

Table 1 Mechanical data of epoxy thermoset polymers cured with bio-based and petrol-based hardeners

Sample and storage conditions		Petrol-based	Bio-based
Tensile strength (MPa)	RT	58.1 ± 2.8	46.7 ± 2.7
	PC	68.0 ± 3.7	72.9 ± 4.1
	H2O	66.3 ± 4.7	66.0 ± 7.0
Elongation at break (%)	RT	2.3 ± 0.1	1.6 ± 0.2
	PC	6.8 ± 0.9	7.7 ± 1.5
	H2O	6.1 ± 1.1	6.2 ± 2.0
Young's modulus (MPa)	RT	3160 ± 120	3320 ± 190
	PC	2995 ± 180	3120 ± 110
	H2O	3070 ± 85	2980 ± 260
T g (°C)	RT	58	62
	PC	84	88
	H2O	82	82

---

The pendulum hardness measured on thin film plates showed that in the bio-based system a faster hardening reaction occurs (Table S5†). Furthermore, it was shown that the curing of the bio-based hardener cured films is more robust, as it is much less impaired by surface effects such as blushing that would inhibit the curing reaction (Table S6†).

Epoxy thermosets after post-curing

The bio-based hardener cured system outperformed the petrol-based system regarding mechanical properties, especially after post-curing. A closer look at the thermo-mechanical behaviour by DMTA indicates that the storage modulus G′ at 30 °C of the bio-based sample is 121.7 MPa, much higher than that of the petrol-based sample (91.5 MPa), indicating higher stiffness (Fig. 4e). Moreover, the T_g (read at the maximum of tan δ) is also higher for the bio-based sample. Generally, the structural rigidity and cross-link density of the polymer networks are the main factors that determine the glass transition temperature T_g. In particular, the aromatic structure restricts molecular chain movement. Using the rubber elasticity model, the cross-link density (v_e) can be calculated using eqn (3):where G′, R and T represent the storage modulus at T_g + 30 °C, the gas constant and the absolute temperature. The results displayed in Table 2 clearly indicate that the bio-based sample is characterised by a higher cross-link density displayed by the higher value for v_e, corresponding to the superior mechanical performance and higher T_g of the bio-based sample after curing. Also, the bio-based sample shows superior strength and ductility over the petrol-based sample after curing, as shown in the stress–strain curve (Fig. 4f).

Table 2 Rubber elasticity model of epoxy thermoset polymers cured with bio-based and petrol-based hardeners after curing (PC)

Epoxy sample	G′ at 30 °C	G′ at Tg + 30 °C	v e (mmol m^−3)
Bio-based	121.7 MPa	4.03 MPa	16.4
Petrol-based	91.5 MPa	3.07 MPa	12.3

---

Properties of the accelerator

Finally, the accelerating properties of Methoxy-K54 (4) were investigated by comparing a system containing no accelerator and one containing the same amount of the structurally most similar petrol-based benchmark, namely Ancamine® K54.

Analysis of the development of pendulum hardness according to König over curing time showed that under different curing conditions same trends were followed (Table 3): the system with no accelerator resulted in the lowest values for hardness under both conditions, while the petrol-based benchmark shows already an improvement and the Methoxy-K54 containing system consistently delivers the highest values for hardness according to König. Values for viscosity and tack free time are similar in all systems.

Table 3 Performance data of the bio-based epoxy accelerator and petrol-based accelerator

Sample and storage conditions			No accelerator	Ancamine® K54	Methoxy-K54
a First 7 days at 8 °C/80% r.h., then 7 days at 23 °C/50% r.h.
Viscosity (Pa·s)			0.13	0.15	0.16
Tack free time (h)			5	5	5
Hardness, König (s)	23 °C/50% r.h.	1 d	129	139	165
		2 d	133	174	199
		7 d	188	195	216
		14 d	193	204	230
	Cold^a conditions	7 d	42	48	63
		14 d	84	169	182

---

Conclusions

In this work, novel bio-based diamines were successfully synthesized from lignin-derived vanillin and guaiacol. This effort marks the first report of the synthesis and characterization of 1,3-bis(aminomethyl)-4,5-dimethoxybenzol, herein called Dimethoxy-MXDA. Additionally, the 4-step synthesis was improved according to green chemistry principles. Furthermore, 2,4-bis(dimethylaminomethyl)-6-methoxyphenol, herein called Methoxy-K54, was successfully synthesized in one reaction step from guaiacol. Dimethoxy-MXDA was used as a bio-based amine hardener in the synthesis of an epoxy thermoset polymer network and showed similar to superior performance when compared to a petrol-based reference system containing the industrially relevant MXDA amine hardener. In particular, the post-cured samples of the bio-based system outperformed the petrol-based benchmark in all properties and using the rubber-elasticity model also a significantly higher cross-link density v_e was calculated for the bio-based epoxy sample. Finally, the Methoxy-K54 accelerated system consistently achieved the highest values for pendulum hardness according to König compared to both the accelerator-free and petrol-based systems, while viscosity and tack-free time remained similar across all systems.

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

The authors gratefully acknowledge Sika Technology AG for financial support. In addition, we thank Steffen Kelch for fruitful discussions and his constructive feedback, as well as Edis Kasemi, Ursula Stadelmann and Urs Burckhardt for their support. Finally, we would like to express our gratitude towards Tim Mamie for his continuous support in the laboratory.

References

1. H. Q. Pham and M. J. Marks, Ullmann's Encycl. Ind. Chem., 2005, 13, 155–244.
2. R. Auvergne, S. Caillol, G. David, B. Boutevin and J.-P. Pascault, Chem. Rev., 2014, 114, 1082–1115.
3. R. Chinthapalli, P. Skoczinski, M. Carus, W. Baltus, D. de Guzman, H. Käb, A. Raschka and J. Ravenstijn, Full Report - Bio-based Building Blocks and Polymers – Global Capacities, Production and Trends 2018–2023, Nova Institute for Ecology and Innovation, 2019.
4. E. A. Baroncini, Y. S. Kumar, G. R. Palmese and J. F. Stanzione, J. Appl. Polym. Sci., 2016, 133, 44103.
5. A. J. Ragauskas, G. T. Beckham, M. J. Biddy, R. Chandra, F. Chen, M. F. Davis, B. H. Davison, R. A. Dixon, P. Gilna, M. Keller, P. Langan, A. K. Naskar, J. N. Saddler, T. J. Tschaplinski, G. A. Tuskan and C. E. Wyman, Science, 2014, 344, 1246843.
6. S. Caillol, M. Fache, E. Darroman, V. Besse, R. Auvergne and B. Boutevin, Green Chem., 2014, 16, 1987–1998.
7. X. Du, A. W. Tricker, W. Yang, R. Katahira, W. Liu, T. T. Kwok, P. Gogoi and Y. Deng, ACS Sustainable Chem. Eng., 2021, 9, 7719–7727.
8. M. Fache, R. Auvergne, B. Boutevin and S. Caillol, Eur. Polym. J., 2015, 67, 527–538.
9. M. Fache, B. Boutevin and S. Caillol, Green Chem., 2015, 18, 712–725.
10. D. Schmiedl, R. Kunkel and V. M. Schmidt, Nachr. Chem., 2021, 69, 35–39.
11. J. Parameswaranpillai, S. M. Rangappa, S. Siengchin and S. Jose, Bio-based Epoxy Polymers, Blends and Composites: Synthesis, Properties, Characterization and Applications, Wiley-VCH, 2020.
12. D. C. Webster and A. Rohly, NDSU Research Foundation, Epoxy Resin derived from Vanillin and Thermosets therefrom, WO2016172353A1, 2016.
13. S. Zheng, G. A. Vedage, M. Rufo, D. I. Totev and M. Cook, Evonik Degussa GmbH, Benzylated Mannich Base Curing Agents, Compositions, and Methods, WO2017147138A1, 2017.
14. W. R. E. Raymond and G. A. Vedage, Evonik Degussa GmbH, Benzylated Polyamine Curing Agents, WO2013043241A1, 2013.
15. S. Grelier, H. Cramail, A. Llevot, S. Carlotti and E. Grau, Université de Bordeaux, New Process for Preparing Biphenyl Compounds, WO2016050988A1, 2016.
16. E. D. Hernandez, A. W. Bassett, J. M. Sadler, J. J. La Scala and J. F. Stanzione, ACS Sustainable Chem. Eng., 2016, 4, 4328–4339.
17. Z. Wang, P. Gnanasekar, S. Sudhakaran Nair, R. Farnood, S. Yi and N. Yan, ACS Sustainable Chem. Eng., 2020, 8, 11215–11223.
18. Y. Wang, H.-L. Gan, C.-X. Liang, Z.-Y. Zhang, M. Xie, J.-Y. Xing, Y.-H. Xue and H. Liu, Giant, 2021, 7, 100063.
19. A. Genua, S. Montes, I. Azcune, A. Rekondo, S. Malburet, B. Daydé-Cazals and A. Graillot, Polymers, 2020, 12, 2645.
20. E. Savonnet, E. Grau, S. Grelier, B. Defoort and H. Cramail, ACS Sustainable Chem. Eng., 2018, 6, 11008–11017.
21. M. Shibata and T. Ohkita, Eur. Polym. J., 2017, 92, 165–173.
22. E. Savonnet, C. C. Le, E. Grau, S. Grelier, H. Cramail and B. Defoort, Front. Chem., 2019, 7, 606.
23. A.-S. Mora, R. Tayouo, B. Boutevin, G. David and S. Caillol, Green Chem., 2018, 20, 4075–4084.
24. A.-S. Mora, R. Tayouo, B. Boutevin, G. David and S. Caillol, Eur. Polym. J., 2020, 123, 109460.
25. M. Huber, E. Kasemi and F. Häfliger, Sika Technology AG, Bio-Based Amine Curing Agent for Curable Compositions, WO2023072786A1, 2021.
26. J. C. Duff and E. J. Bills, J. Chem. Soc., 1932, 1987–1988.
27. J. C. Duff, J. Chem. Soc., 1941, 547–550.
28. C. Mannich and W. Krösche, Arch. Pharm., 1912, 250, 647–667.
29. P. T. Anastas and J. C. Warner, Green Chemistry: Theory and Practice, Oxford University Press, 2000.
30. European-Comission, EU policy framework on biobased, biodegradable and compostable plastics, 2022.
31. Y. Yue, F. Huo, C. Yin, J. Chao, Z. Yongbin and X. Wei, RSC Adv., 2015, 5, 77670–77672.
32. J. Saiz-Poseu, I. Alcón, R. Alibés, F. Busqué, J. Faraudo and D. Ruiz-Molina, CrystEngComm, 2012, 14, 264–271.
33. A. Nicolae, A. Draghici and A. David, Rev. Chim., 2003, 54, 388–392.
34. D. M. Musatov, E. V. Starodubtseva, O. V. Turova, D. V. Kurilov, M. G. Vinogradov, A. K. Rakishev and M. I. Struchkova, Russ. J. Org. Chem., 2010, 46, 1021–1028.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5gc00446b

---