Study on the Room-Temperature Rapid Curing Behavior and Mechanism of HDI Trimer-Modified Epoxy Resin

Abstract

This study resolves the challenge of balancing curing speed and performance in room-temperature-curing epoxy coatings by developing a novel system grafted with hexamethylene diisocyanate trimer (HDI trimer) and polyethylene glycol 200 (PEG200). Employing DMP-30 as the catalyst, the coating achieves efficient curing at 25 °C, with complete cure within 7.5 h. The cured material exhibits outstanding thermal stability (T₅₀% = 380.83 °C) and mechanical properties. Fracture morphology analysis reveals a uniform ductile structure, confirming its high toughness and durability. Furthermore, kinetic models accurately predict curing behavior across different temperature curves, providing crucial guidance for optimizing industrial coating processes. This research offers a viable strategy for designing high-performance, rapid curing epoxy materials, demonstrating significant application potential in coating systems, composite surfaces, and electronic encapsulation.

1. Introduction

Epoxy resins have become one of the core materials in coatings, composites, electronic packaging, and structural adhesives due to their outstanding bonding properties, high mechanical strength, excellent chemical resistance, and dimensional stability [1,2]. Among these, room-temperature curing epoxy resin systems offer irreplaceable advantages in large-scale construction sites, aerospace rapid repair, and precision electronic assembly due to their lack of external energy input, simplified process operations, and suitability for heat-sensitive components [3,4].

However, traditional room-temperature curing epoxy resin systems commonly suffer from the critical bottleneck of slow curing rates, making it difficult to meet modern manufacturing demands for efficient processes [5]. More critically, a difficult-to-resolve trade-off often exists between curing speed and final material properties. Tertiary amine accelerators (e.g., 2,4,6-tri(dimethylaminomethyl)phenol (DMP-30)), widely used for rapid curing, effectively shorten gel time but tend to cause non-uniform crosslinking networks, thereby increasing brittleness and reducing impact strength [6,7]. Conversely, methods enhancing toughness by introducing flexible units or elastomers often come at the expense of modulus, strength, and thermal stability, creating an inverted “toughness–rigidity” relationship [8,9]. For instance, novel flexible thiourea curing agents based on cashew nut shell liquid (CNSL) significantly enhances epoxy resin toughness while maintaining high reactivity but inevitably cause a marked decrease in glass transition temperature [10].

To overcome these performance bottlenecks, researchers have explored multiple avenues. Among these, rigid nanoparticle filling is regarded as a viable solution [11]. For instance, surface-modified silica nanoparticles have been used as industrial fillers for nearly two decades. Within epoxy matrices, they not only independently contribute toughening effects but also synergize with second phases like elastomers. This combination significantly enhances toughness and fatigue resistance while effectively compensating for modulus and strength losses caused by the elastomer phase [12,13]. However, physically blended nanoparticles still face challenges such as dispersion stability, process complexity, and cost.

Given the current research landscape, a promising approach involves constructing an inherent “hybrid rigid–flexible” structure within the epoxy network through chemical reactions at the molecular design level. This study proposes a synergistic modification system composed of hexamethylene diisocyanate trimer (HDI trimer) and polyethylene glycol (PEG200), achieving rapid room-temperature curing via the dual catalytic function of DMP-30. The core innovation lies in utilizing HDI trimer to react with epoxy groups and form rigid oxazolidinone rings, thereby enhancing crosslink density, modulus, and thermal stability, while simultaneously introducing flexible PEG200 segments to absorb energy and improve toughness. Finally, DMP-30 synergistically catalyzes these reactions to balance their rates, achieving coordinated optimization of curing speed, rigidity, toughness, and thermal stability at the molecular level. Systematic studies confirm that this strategy successfully reduces curing time to 7.5 h while maintaining high crosslink density and outstanding comprehensive mechanical properties. This provides an innovative molecular-level solution to the long-standing “performance trade-off” challenge in room-temperature rapid curing epoxy resins.

2. Experimental Section

2.1. Primary Materials

Bisphenol A epoxy resin (E-44, epoxy equivalent weight of 210–230 g/eq) was supplied by Shandong YouSuo Chemical Technology Co., Ltd. (Linyi, China), Diethylenetriamine (DETA), hexamethylene diisocyanate trimer (HDI trimer), and 2,4,6-tris(dimethylaminomethyl)phenol (DMP-30) were obtained from Chengdu Kelong Co., Ltd. (Chengdu, China), Jining Lido Chemical Co., Ltd. (Jining, China), and Jinan Dahui Chemical Technology Co., Ltd. (Jinan, China), respectively. Polyethylene glycol (PEG200) was used as received, purchased from Chengdu Kelong Co., Ltd. (Chengdu, China).

Calculation of Molar Ratios: The molar ratios of the reactive groups were calculated based on the following parameters: the epoxy equivalent weight of E-44 was taken as the median value (227 g/eq) of its specified range; HDI trimer was assigned a molecular weight of 504 g/mol with a functionality of 3; and PEG200 was assigned a molecular weight of 200 g/mol with a functionality of 2.

2.2. Primary Instruments and Equipment

Comprehensive characterization of samples was conducted through integrated analytical techniques to elucidate their chemical structure, thermal behavior, mechanical properties, and morphological features.

Crosslinking Degree Determination: The crosslinking degree of cured epoxy resin is expressed as gel fraction and determined by solvent extraction. Precisely weigh the cured sample (initial mass m₀) and place it in a Soxhlet extractor. Extract using the organic solvent acetone at 80 °C for 8 h. This process effectively removes uncrosslinked polymer chains and residual monomers from the epoxy resin. After extraction, the sample is vacuum-dried at 60 °C for 24 h until constant weight is achieved, yielding the final mass (m₁). The crosslinking degree is calculated as (m₁/m₀) × 100% [14,15].

The chemical structure and functional groups were analyzed using a Fourier Transform Infrared (FT-IR) spectrometer (Vector-22, Bruker, Germany). All spectra were acquired in Attenuated Total Reflectance (ATR) mode over a wavenumber range of 4000 to 400 cm⁻¹ with a resolution of 4 cm⁻¹ and 32 accumulated scans.

Thermal stability was evaluated by Thermogravimetric Analysis (TGA) on an STA449-F3 thermal analyzer (NETZSCH, Germany). Measurements were conducted under dynamic nitrogen atmosphere (50 mL/min flow rate) by heating samples from room temperature to 800 °C at a constant rate of 10 °C/min. Thermal transitions, including the glass transition and melting temperature, were investigated using Differential Scanning Calorimetry (DSC 214 Polyma, NETZSCH, Germany). The DSC tests were performed in sealed aluminum crucibles under a nitrogen purge (50 mL/min), applying a heating program from room temperature to 250 °C at a rate of 10 °C/min.

Static Mechanical Properties: 1. Tensile Strength: Determined using the GTM8010 Universal Material Testing Machine (Fengwang Instruments, China) in accordance with ASTM D638 standard. 2. Flexural Strength: Determined using the GTM8010 Universal Material Testing Machine (Fengwang Instruments, China) in accordance with ASTM D790 standard employing a three-point bending test.

Finally, the microstructure and fracture surface morphology of the samples were examined using a TH-F120 scanning electron microscope (FEI, USA). To ensure sufficient conductivity, the samples were sputter-coated with a thin layer of gold before observation, and micrographs were captured at an accelerating voltage of 5 kV.

2.3. Characterization and Analysis

(1)

Preparation Process

Modification Process: The modification process begins by adding hexamethylene diisocyanate trimer (HDI trimer) and polyethylene glycol 200 (PEG200) in a stoichiometric ratio to a glass three-neck flask to form an isocyanate-capped prepolymer. The mixture is heated to 60 °C in a constant-temperature water bath and stirred continuously for 3 h. The main component, bisphenol A epoxy resin (E-44), is then added to the reactor. After raising the temperature to 85 °C, the reaction continues for 2 h under mechanical stirring, allowing the epoxy groups to react with residual isocyanate functional groups and graft flexible units onto the epoxy backbone. Upon reaction completion, volatile components and residual solvents were removed under reduced pressure using a rotary evaporator, yielding a viscous modified epoxy prepolymer, specific addition amounts are detailed in

Table 1. Formulations for each sample.

Serial Number		E-44 (wt%)	DETA (wt%)	HDI Trimer (wt%)	PEG200 (wt%)	DMP-30 (wt%)
1#	Pure epoxy without accelerators	10	3
2#	Modified epoxy without accelerator	10	3	0.049	0.025
3#	Modified epoxy with accelerator	10	3	0.049	0.025	0.039

Note: For Samples #2 and #3, the molar ratio of epoxy groups/NCO groups/hydroxyl groups is 1:0.0068:0.0045.

.

Curing Process: The curing agent diethylenetriamine (DETA) was added to the prepolymer in a stoichiometric ratio to the epoxy groups. For the catalytic system, a predetermined amount of DMP-30 accelerator was subsequently introduced. The mixture was mechanically stirred for 10 min to achieve homogeneity, then degassed in a vacuum oven for 20 min to eliminate entrapped bubbles. The resulting sample was coated with a controlled thickness of 100 ± 25 microns and cured at 25 °C to form the final crosslinked network, the reaction mechanism is illustrated in Figure 1.

(2)

Synthesis Pathway

Figure 1. Reaction mechanism diagram of trimethylolpropane trimer-modified bisphenol a epoxy resin.

Figure 1. Reaction mechanism diagram of trimethylolpropane trimer-modified bisphenol a epoxy resin.

3. Analysis of Results

3.1. Analysis of Curing Behavior and Crosslink Density

As shown in

Table 2. Curing times and crosslinking degrees for Samples 1, 2, and 3.

	Surface Cure Time	Crosslinking Degree
1#	Over 14 h	87.00%
2#	Over 14 h	98.13%
3#	7 h and 30 min	97.86%

, the curing time was significantly reduced from over 14 h to 7.5 h after adding DMP-30, nearly doubling the reaction rate.

This stems from DMP-30’s highly efficient catalytic action, which markedly accelerates the kinetics of the amine–epoxy reaction. DMP-30 substantially accelerates the reaction process through its unique synergistic catalytic mechanism. Its tertiary amine group first forms a molecular complex with the primary amine group of the amine curing agent. This complex activates the amine group via a proton transfer mechanism, significantly enhancing its nucleophilicity. This enables more efficient attack on the epoxy ring, representing the primary pathway accelerating the amine/epoxy curing reaction. Simultaneously, the tertiary amine group of DMP-30 can directly nucleophilically attack the isocyanate group (-NCO), forming an amphoteric ion intermediate. This intermediate polarizes and activates the epoxy group via an ion-pair mechanism, providing a low-energy barrier, rapid reaction pathway for the -NCO to react with the epoxy group to form an oxazolidinone ring. These two catalytic mechanisms [16,17] collectively reduce the overall activation energy of the curing system, enabling the simultaneous achievement of high crosslink density and high-performance networks within an extremely short timeframe.

3.2. Molecular Structure Analysis

To verify the successful synthesis of the prepolymer and the formation of the oxazolidinone ring during curing, Fourier Transform Infrared spectroscopy analysis was conducted. The comparative spectra are shown in Figure 2, with the following detailed interpretation:

To further analyze the molecular structural changes in the epoxy resin network, detailed Fourier Transform Infrared (FT-IR) spectroscopy was conducted, systematically comparing the spectra of modified samples against the baseline (unmodified) sample. The FT-IR spectrum of the unmodified, uncured epoxy resin (Figure 2a) serves as the baseline, exhibiting characteristic absorption peaks: a broad band around 3500 cm⁻¹ (-OH stretching vibration) and a distinct peak at ~915 cm⁻¹ (epoxy ring vibration).

Crucially, the spectrum of the modified uncured sample (prepolymer, red line in Figure 2a) exhibits a distinct new peak at 2270 cm⁻¹. This corresponds to the characteristic stretching vibration peak of the isocyanate group (-NCO) in the grafted HDI trimer. This peak, together with the persistent epoxy absorption peak, confirms the successful synthesis of a prepolymer retaining terminal -NCO groups for subsequent reactions. After curing, the comparative analysis of cured Sample 1 (control) and Sample 3 (modified) in Figure 2b reveals significant chemical changes. Key differences include the complete disappearance of the -NCO peak at 2270 cm⁻¹ in Sample 3, indicating complete consumption of all isocyanate groups. Simultaneously, a characteristic absorption peak at 1740–1750 cm⁻¹ appears in Sample 3, absent in the control sample. This peak corresponds to the characteristic carbonyl (C=O) stretching vibration of the five-membered oxazolidinone ring. The intensity of the epoxy peak (~915 cm⁻¹) in Sample 3 is significantly reduced compared to the control group, confirming the consumption of epoxy groups during the cyclization reaction. These spectral differences directly indicate the disappearance of the NCO peak and the appearance of the oxazolidinone C=O peak, proving that DMP-30 catalyzed the cyclization reaction between the epoxy group and isocyanate, forming rigid oxazolidinone rings in the network [18]. This rigid structure simultaneously enhances crosslink density and network rigidity, enabling Sample 3 to achieve a crosslinking degree of 97.86% within 7.5 h.

3.3. Molecular Weight Analysis

The molecular weight distribution of the epoxy systems was characterized by GPC, with the results presented in Figure 3 and the corresponding statistical data summarized in

Table 3. Molecular weight statistics results.

	Mn	Mw	Mp	PDI
Pure epoxy	2602	3089	1389	1.18
HDI trimer–PEG-modified Epoxy	8690	13842	7944	1.59

.

As shown in Table 2, the number-average molecular weight (Mn) and weight-average molecular weight (Mw) of the HDI trimer–PEG-modified epoxy increased by approximately 3.3-fold and 4.5-fold, respectively, compared to those of the pure epoxy resin. This remarkable increase confirms the successful grafting of HDI trimer and PEG200 onto the epoxy backbone, leading to the formation of a macromolecular prepolymer. Furthermore, the polydispersity index (PDI) broadened significantly from 1.18 to 1.59, indicating that the modified product is a mixture with a more heterogeneous molecular weight distribution. This broadening suggests that while many epoxy chains were grafted with multiple HDI trimer–PEG segments, resulting in high-molecular weight species, a portion of the epoxy molecules either remained unreacted or exhibited a low grafting degree, retaining lower-molecular weights. This specific structural modification—introducing flexible segments while preserving reactive sites—is anticipated to facilitate the efficient formation of a denser three-dimensional network during the subsequent DMP-30-catalyzed curing process.

3.4. Study on Mechanical Properties

Tensile and flexural test results indicate that different epoxy resin systems exhibit distinct mechanical properties due to structural variations. As shown in Figure 4,

Table 4. Tensile mechanical properties of Samples 1–3.

	Elongation at Break (%)	Young’s Modulus (GPa)
1#	8.08	1.52
2#	4.00	1.65
3#	3.50	1.84

and

Table 5. Flexural mechanical properties of Samples 1–3.

	Deflection at a Strain of 0.0005 (mm)	Deflection at a Strain of 0.0025 (mm)	Elasticity Flexural Modulus (GPa)
1#	0.08	0.42	2.06
2#	0.09	0.43	3.11
3#	0.09	0.43	3.56

, Sample 3 (modified with DMP-30) demonstrates the highest modulus and appropriate toughness, making it the most superior in overall performance.

Sample 3 (modified with DMP-30) demonstrates superior mechanical properties, exhibiting the highest tensile modulus (1.84 GPa) and flexural modulus (3.56 GPa) among all samples. This enhancement is attributed to the synergistic effect of the rigid oxazolidinone rings and the flexible PEG segments within the crosslinked network. The DMP-30 catalyst promotes the formation of oxazolidinone structures, leading to a high crosslink density and increased stiffness. Simultaneously, the PEG segments impart toughness, as evidenced by Sample 3 maintaining a respectable elongation at break (3.50%) despite its significant increase in rigidity. This combination results in a densely crosslinked yet tough three-dimensional network.

3.5. Microstructural Analysis

Figure 5 shows the surface microstructure of the tensile fracture surfaces for (a) Sample 1, (b) Sample 2, and (c) Sample 3.

The pure sample’s SEM image reveals a fracture surface exhibiting sharp edges and minimal striations, characteristic features of brittle fracture behavior [19]. Following the introduction of flexible segments, the fracture surfaces in (b) and (c) become relatively rough with increased microcrack formation, reflecting enhanced segment entanglement and improved toughness and energy absorption capacity. The modified epoxy resin undergoes plastic deformation under external loading. Following the addition of accelerator DMP-30, the number of microcracks in (c) decreases compared to (b), while fine elevations appear. The protrusions on the SEM fracture surface in (c) represent shear-yielding zones induced by localized plastic deformation due to stress concentration. This phenomenon indicates a shift in fracture mechanism from microcrack-dominated failure to more efficient shear yielding, reflecting the synergistic interaction between the rigid oxazolidinone network and flexible segments.

3.6. Study on Heat Resistance Properties

The thermal stability of cured epoxy systems was evaluated via Thermogravimetric Analysis (TGA), with representative weight loss curves and detailed parameters shown in Figure 6 and

Table 6. Thermal weight loss parameters.

	T5% (°C)	T50% (°C)	Tmax (°C)	Residual Sample Mass at 500 °C (%)
1#	337.75	379.00	375.11	13.95
2#	325.01	374.07	376.50	9.74
3#	315.81	380.83	381.44	10.49

.

Thermogravimetric Analysis data indicate that the unmodified epoxy resin (Sample 1) exhibits the highest initial thermal stability (T_5% = 337.75 °C), but its primary decomposition temperature stages (T_50% and T_max) are relatively low. The system modified with hexamethylene diisocyanate trimer and polyethylene glycol without accelerator (Sample 2) exhibited reduced initial decomposition temperature and residual carbon content at 500 °C due to the introduction of less thermally stable aliphatic segments. The modified system incorporating DMP-30 accelerator (Sample 3) demonstrated superior performance. While its initial decomposition temperature decreased due to the decomposition of small-molecule accelerators and flexible segments, accelerator DMP-30 significantly promotes the formation of a more complete and dense crosslinked network. During the main decomposition stage, this sample exhibited thermal stability, with residual content at 500 °C higher than the control system without added accelerator. This indicates that DMP-30 effectively promotes the formation of a more complete and dense crosslinked network, thereby enhancing the material’s thermal stability under high-temperature conditions to a certain extent.

3.7. Non-Isothermal Curing Kinetic Analysis

A systematic analysis was conducted on an epoxy resin system modified with HDI trimer and PEG200 and cured using DMP-30 as a curing agent. Non-isothermal DSC testing revealed a distinct double-peak exothermic characteristic in this system, which elucidates the mechanism by which DMP-30 promotes epoxy curing [20,21], Figure 7 shows the DSC curves of Sample 3 at different heating rates.

The kinetic analysis of non-isothermal DSC experiments is based on the assumption that the curing reaction rate is proportional to the measured heat flow value, as expressed by the curing kinetic equation (Equation (1)):

$${\displaystyle \frac{d\alpha }{dt}}=k\left(T\right)f\left(\alpha \right)h\left(P\right)\approx k\left(T\right)f\left(\alpha \right)=Ae({\displaystyle \frac{-E}{RT}})f\left(\alpha \right)$$

(1)

where α represents the reaction conversion rate and dα/dt denotes the reaction rate. The reaction rate is primarily determined by three functions: k(T), the reaction rate constant at temperature T governed by the Arrhenius Equation (2) and defined by A (the pre-exponential factor) and E_a (apparent activation energy), where R is the universal gas constant and T is the absolute temperature; f(α), the reaction mechanism function, whose specific form determines the reaction’s kinetic model; and h(P), the pressure function, where R is the universal gas constant (8.314 J/mol·K).

$$k\left(T\right)=Ae(-{\displaystyle \frac{E_a}{\mathit{RT}}})$$

(2)

Simultaneously, the E_a value can be calculated using the Kissinger model, specifically through Equation (3):

$${\displaystyle \frac{d[\mathit{ln}\left(\frac{\beta }{T_P^2}\right)]}{d[(\frac{1}{T_P})]}}=-({\displaystyle \frac{E_a}{\mathit{RT}}})$$

(3)

β is the heating rate, and T_p is the peak temperature of the exothermic peak. For more precise calculation, the Kissinger–Akahira–Sunose (KAS) model is employed:

$${\displaystyle \frac{d[\mathit{ln}\left(\frac{{\beta }_i}{T_{\alpha ,i}^{1.92}}\right)]}{d[(\frac{1}{T_{\alpha }})]}}=-1.0008({\displaystyle \frac{E_{\alpha }}{R}})$$

(4)

The value of E_a can be obtained from the slope of the curve ln(β = T_P^1.92) versus (1/T_P) in the left graph below. Figure 8 and Figure 9 below show the 1/Tα-ln(β/Tα) relationship (a) and the Eα-(α) curve for Peak 1 and Peak 2, respectively.

Peak 1 primarily corresponds to the ring-opening addition reaction between amine curing agents and epoxy groups. Under the alkaline catalysis of DMP-30, the nucleophilicity of amine groups is significantly enhanced, enabling them to react with epoxy groups first and form an initial three-dimensional crosslinked network.

Peak 2 corresponds to the reaction between the terminal -NCO groups of the prepolymer and epoxy groups. As the three-dimensional crosslinked network forms, system viscosity increases and molecular segment motion becomes restricted. Under DMP-30 catalysis, residual -NCO groups react with epoxy groups via cyclization to form rigid oxazolidinone structures.

To further establish suitable curing kinetic equations, the Malek method was employed to calculate y(α) and z(α) for model identification. The specific equations are as follows:

$$y\left(\alpha \right)={\displaystyle \frac{d\alpha }{dt}}exp(\chi )$$

(5)

$$z\left(\alpha \right)=\pi (\chi )({\displaystyle \frac{d\alpha }{dt}}){\displaystyle \frac{T}{\beta }}$$

(6)

Here, χ denotes E_a/RT, and π(χ) is computed using the fourth-order rational approximation formula proposed by Senum and Yang [22].

$$\pi \left(\chi \right)\approx {\displaystyle \frac{{\chi }^3+18{\chi }^2+88\chi +96}{{\chi }^4+20{\chi }^3+120{\chi }^2+240\chi +120}}$$

(7)

Normalizing the data yields the following figure, Figure 10 displays the y(α)-α and z(α)-α curves for Peak 1, with

Table 7. Characteristic parameters of the Peak 1 Malek method.

β (K/min)	αM (y(α) Peak Value)	αp (DSC Peak Value)	αp∞ (z(α) Peak Value)
5	0.30	0.25	0.30
10	0.40	0.40	0.40
15	0.50	0.45	0.50
20	0.50	0.40	0.50

presenting its characteristic parameters; Figure 11 shows the y(α)-α and z(α)-α curves for Peak 2, with

Table 8. Characteristic parameters of the Peak 2 Malek method.

β (K/min)	αM (y(α) Peak Value)	αp (DSC Peak Value)	αp∞ (z(α) Peak Value)
5	0.50	0.25	0.30
10	0.50	0.25	0.30
15	0.50	0.25	0.30
20	0.50	0.25	0.30

listing its characteristic parameters:

Based on the data from the above chart, α_p^∞ does not depend on the heating rate, and since α_M ∈ (0, 1) and α_p^∞ ≠ 0.633, the Malek method indicates that the DMP-30-catalyzed epoxidation system follows an autocatalytic model, specifically the Sestak–Berggren model:

$${\displaystyle \frac{d\alpha }{\mathit{dt}}}=Ae\left({\displaystyle \frac{-E}{\mathit{RT}}}\right){\alpha }^m{\left(1-\alpha \right)}^n$$

(8)

The curing kinetics parameters of the autocatalytic model obtained through nonlinear fitting are as follows,

Table 9. Peak 1 curing kinetic parameters.

β (K/min)	lnA (s−1)	Average	n	Average	m	Average	R2
5	36.76	36.69	1.51	1.09	−0.05	0.01	0.93
10	36.58		1.02		−0.02		0.96
15	36.76		0.91		0.08		0.98
20	36.64		0.92		0.04		0.98

presents the curing kinetic parameters for Peak 1, while

Table 10. Peak 2 curing kinetic parameters.

β (K/min)	lnA (s−1)	Average	n	Average	m	Average	R2
5	28.09	28.12	1.12	1.12	−0.03	−0.06	0.99
10	28.55		1.31		0.06		0.99
15	28.05		1.09		−0.09		0.99
20	27.78		0.97		−0.06		0.99

shows those for Peak 2:

Peak 1 yields the curing kinetics Equation (9) through derivation:

$${\displaystyle \frac{d\alpha }{\mathit{dt}}}=e^{36.68}{e}^{-{\displaystyle \frac{13839.86}{T}}}{\alpha }^{0.01}{\left(1-\alpha \right)}^{1.09}$$

(9)

Peak 2 obtained the curing kinetic parameters based on the fitting:

Peak 2 yields the curing kinetics Equation (10) through derivation:

$${\displaystyle \frac{d\alpha }{dt}}=e^{28.12}{e}^{-{\displaystyle \frac{12231.86}{T}}}{\alpha }^{-0.06}{\left(1-\alpha \right)}^{1.12}$$

(10)

Peaks 1 and 2 were fitted using the Sestak-Berggren model, yielding the curve shown in Figure 12.

As shown in the figure, the experimental values closely match the fitted values, indicating that the Sestak–Berggren model accurately simulates the curing reaction of this system. The accelerator significantly reduces the activation energy through a catalytic mechanism, shifting the curing peak temperature to lower temperatures and substantially increasing the reaction rate, with particularly pronounced effects in the low-temperature range.

The formation of the amine–epoxy three-dimensional crosslinking network ensures the material’s early mechanical properties, while the subsequent DMP-30-catalyzed formation of oxazolidinone rings as rigid crosslinking nodes significantly enhances the final product’s crosslink density and thermal stability. TGA indicates that Sample 3 exhibits a T_50% of 380.83 °C and a T_max of 381.44 °C, both significantly outperforming the unmodified system.

4. Conclusions

Through graft modification of epoxy resin with hexamethylene diisocyanate trimer and polyethylene glycol, coupled with the catalytic action of DMP-30 accelerator, an epoxy resin system exhibiting rapid room-temperature curing properties and excellent comprehensive performance was successfully prepared. The study demonstrates that this system rapidly completes the amine–epoxy reaction at room temperature to form a basic network. Subsequently, a rigid reinforcement node is constructed via the oxazolidinone cyclization reaction, enabling complete curing within 7.5 h and exhibiting optimal comprehensive properties: its high-temperature thermal stability (T_50% = 380.83 °C, T_max = 381.44 °C) and mechanical strength (maximum tensile/flexural strength) significantly outperformed Samples 1 (unmodified) and 2 (modified but uncured). Collectively, these data demonstrate that DMP-30 achieves synergistic enhancement in curing efficiency, thermal stability, and mechanical properties through the synergistic action of catalyzed amine–epoxy curing and oxazolidinone ring formation.

The concurrently established Sestak–Berggren kinetic model precisely describes the curing process. Through accurate fitting of kinetic triplet parameters (Ea, m, n), this model not only quantitatively reveals the coexisting autocatalytic characteristics and diffusion-controlled mechanisms in the stepwise curing reaction catalyzed by DMP-30 but also successfully constructs a mathematical model capable of accurately predicting material curing behavior under any temperature profile. This model provides a theoretical tool for optimizing curing process parameters and guiding formulation design. It significantly reduces costs while offering new design concepts and technical pathways for developing high-performance epoxy insulating materials.