Curing Reaction Kinetics of the EHTPB-Based PBX Binder System and Its Mechanical Properties

: In this research, differential scanning calorimetry (DSC) was employed to compare the curing reaction kinetics of the epoxidized hydroxyl terminated polybutadiene-isophorone diisocyanate (EHTPB-IPDI) and hydroxyl terminated polybutadiene-isophorone diisocyanate (HTPB-IPDI) binder systems. Glass transition temperature ( T g ) and mechanical properties of the EHTPB-IPDI and HTPB-IPDI binder systems were determined using the DSC method and a universal testing machine, respectively. For the EHTPB-IPDI binder system, the change of viscosity during the curing process in the presence of dibutyltin silicate (DBTDL) and tin 2-ethylhexanoate (TECH) catalysts was studied, and the activation energy was estimated. The results show that the activation energies (Ea) of the curing reaction of the EHTPB-IPDI and HTPB-IPDI binder systems are 53.8 and 59.1 kJ · mol − 1 , respectively. While their average initial curing temperatures of the two systems are 178.2 and 189.5 ◦ C, respectively. The EHTPB-IPDI binder system exhibits a higher reactivity. Compared with the HTPB-IPDI binder system, the T g of the EHTPB-IPDI binder system is increased by 5 ◦ C. Its tensile strength and tear strength are increased by 12% and 17%, respectively, while its elongation at break is reduced by 10%. Epoxy groups and isocyanates react to form oxazolidinones, thereby improving the mechanical properties and thermal stability of polyurethane materials. These differences indicate that the EHTPB-IPDI binder system has better thermal stability and mechanical properties. During the EHTPB-IPDI binder system’s curing process, the DBTDL catalyst may ensure a higher viscosity growth rate, indicating a better catalytic effect, consistent with the prediction results obtained using the non-isothermal kinetic analysis method. The performance of a polyurethane elastomer relies not only on the chemical composition and preparation technique but also on the curing conditions. Research on the curing reaction kinetics provides more in-depth insight into the curing process. This study investigated the curing kinetics of the HTPB-IPDI and EHTPB-IPDI binder systems using the DSC method. The glass transition temperatures ( T g ) of the EHTPB-IPDI and HTPB-IPDI polyurethane elastomers were determined. The mechanical properties of the polyurethane elastomers were tested using a universal testing machine. The curing kinetics of the EHTPB-IPDI system under different catalytic conditions were studied, and the corresponding curing kinetic parameters were determined, such as post-processing temperature and curing time in the gelation. Such as post-processing temperature and curing time in the gelation. These results provide theoretical support for the determination of the optimal curing parameters. vulcanized thermoplastic, determined tear strength. Tearing properties of L-shaped specimens were also tested with an EMT-C universal material testing machine at room temperature (25 ◦ C). The gauge length of the specimens was ﬁxed at 25 mm, and the tensile speed was 50 mm / min. Each group of the binder system ﬁlms was tested for at least 5 times to assess the experiment’s reproducibility. order, .


Sample Preparation
Each binder system was prepared according to the following procedures: HTPB-IPDI and EHTPB-IPDI with a weight ratio of R(nNCO/nOH) = 1 were added into a dry beaker. An electrical stirring machine was used to mix the two components for 10 min. The pressure of the mixture was reduced to 100 Pa and kept for 5 min to remove the formed bubbles. Finally, the binder system was obtained and collected.
The films of the binder system were prepared by pouring the as-prepared binder system into polytetrafluoroethylene (PTFE) molds, placing them in a thermostatic oven and curing at 70 °C for 7 days. The thickness of the films is 2 mm.

Calculation of the Curing Reaction Kinetics
Kissinger, Flynn-Wall-Ozawa (F-W-O), Doyle and Crane methods are commonly used to determine the kinetic parameters of a non-isothermal curing process. The kinetics can be expressed by the following equations [22,23]:

Sample Preparation
Each binder system was prepared according to the following procedures: HTPB-IPDI and EHTPB-IPDI with a weight ratio of R(nNCO/nOH) = 1 were added into a dry beaker. An electrical stirring machine was used to mix the two components for 10 min. The pressure of the mixture was reduced to 100 Pa and kept for 5 min to remove the formed bubbles. Finally, the binder system was obtained and collected.
The films of the binder system were prepared by pouring the as-prepared binder system into polytetrafluoroethylene (PTFE) molds, placing them in a thermostatic oven and curing at 70 °C for 7 days. The thickness of the films is 2 mm.

Calculation of the Curing Reaction Kinetics
Kissinger, Flynn-Wall-Ozawa (F-W-O), Doyle and Crane methods are commonly used to determine the kinetic parameters of a non-isothermal curing process. The kinetics can be expressed by the following equations [22,23]:

Sample Preparation
Each binder system was prepared according to the following procedures: HTPB-IPDI and EHTPB-IPDI with a weight ratio of R(nNCO/nOH) = 1 were added into a dry beaker. An electrical stirring machine was used to mix the two components for 10 min. The pressure of the mixture was reduced to 100 Pa and kept for 5 min to remove the formed bubbles. Finally, the binder system was obtained and collected.
The films of the binder system were prepared by pouring the as-prepared binder system into polytetrafluoroethylene (PTFE) molds, placing them in a thermostatic oven and curing at 70 • C for 7 days. The thickness of the films is 2 mm.

Calculation of the Curing Reaction Kinetics
Kissinger, Flynn-Wall-Ozawa (F-W-O), Doyle and Crane methods are commonly used to determine the kinetic parameters of a non-isothermal curing process. The kinetics can be expressed by the following equations [22,23]: Kissinger equation: (F-W-O) equation: Doyle equation: where, T p and T(K) are the temperatures of the peaks, β (K/min) is the heating rate, R (8.314 J/mol) is the gas constant, E (kJ/mol) is the apparent activation energy and A is the pre-exponential factor, which has the same dimension as the rate constant and depends on the reaction order, n.

DSC Testing
The DSC tests for the analysis of curing reaction kinetics were conducted in a nitrogen atmosphere at a flow rate of 50 mL/min. In a typical experiment, a crucible with 10 mg of sample (propellant slurry) Coatings 2020, 10, 1266 4 of 15 was placed in the heating chamber of the DSC analyzer (HCT-1 thermal analyzer, Beijing Yongjiu Instrument, Beijing, China). The measurements were performed by heating from 25 to 350 • C at four different heating rates (5, 10, 15 and 20 • C/min).
The determination of T g was performed using the described DSC tests, except that the heating rate was 10 • C/min. That for T g the cured samples had been tested by DSC and that an increased heat flow (endothermic behavior) shows T g .

Mechanical Properties of EHTPB-IPDI and HTPB-IPDI Polyurethane Elastomers
The tensile properties of the EHTPB-IPDI and HTPB-IPDI binder system films were tested with an EMT-C universal material testing machine.
According to the Chinese standard: GB/T 528-1998 [24], rubber, vulcanized or thermoplastic, we determined the tensile stress-strain properties. The tensile strength and elongation at break of the specimens were measured by setting the gauge length of the specimens and the tensile speed at 25 mm and 50 mm/min, respectively. Five samples were cut from each type of the binder system film, and further analyzed. Figure 3 represents the dumbbell-shaped specimens. The data recorded in the five tests of the same binder system film were collected. The mean value of the indexes obtained from the five measurements was set as the final test result.

DSC Testing
The DSC tests for the analysis of curing reaction kinetics were conducted in a nitrogen atmosphere at a flow rate of 50 mL/min. In a typical experiment, a crucible with 10 mg of sample (propellant slurry) was placed in the heating chamber of the DSC analyzer (HCT-1 thermal analyzer, Beijing Yongjiu Instrument, Beijing, China). The measurements were performed by heating from 25 to 350 °C at four different heating rates (5, 10, 15 and 20 °C/min).
The determination of Tg was performed using the described DSC tests, except that the heating rate was 10 °C/min. That for Tg the cured samples had been tested by DSC and that an increased heat flow (endothermic behavior) shows Tg.

Mechanical Properties of EHTPB-IPDI and HTPB-IPDI Polyurethane Elastomers
The tensile properties of the EHTPB-IPDI and HTPB-IPDI binder system films were tested with an EMT-C universal material testing machine.
According to the Chinese standard: GB/T 528-1998 [24], rubber, vulcanized or thermoplastic, we determined the tensile stress-strain properties. The tensile strength and elongation at break of the specimens were measured by setting the gauge length of the specimens and the tensile speed at 25 mm and 50 mm/min, respectively. Five samples were cut from each type of the binder system film, and further analyzed. Figure 3 represents the dumbbell-shaped specimens. The data recorded in the five tests of the same binder system film were collected. The mean value of the indexes obtained from the five measurements was set as the final test result. According to the Chinese standard: GB/T 529-2008 [25], Rubber, vulcanized or thermoplastic, we determined tear strength. Tearing properties of L-shaped specimens ( Figure 4) were also tested with an EMT-C universal material testing machine at room temperature (25 °C). The gauge length of the specimens was fixed at 25 mm, and the tensile speed was 50 mm/min. Each group of the binder system films was tested for at least 5 times to assess the experiment's reproducibility.  According to the Chinese standard: GB/T 529-2008 [25], Rubber, vulcanized or thermoplastic, we determined tear strength. Tearing properties of L-shaped specimens ( Figure 4) were also tested with an EMT-C universal material testing machine at room temperature (25 • C). The gauge length of the specimens was fixed at 25 mm, and the tensile speed was 50 mm/min. Each group of the binder system films was tested for at least 5 times to assess the experiment's reproducibility.

DSC Testing
The DSC tests for the analysis of curing reaction kinetics were conducted in a nitrogen atmosphere at a flow rate of 50 mL/min. In a typical experiment, a crucible with 10 mg of sample (propellant slurry) was placed in the heating chamber of the DSC analyzer (HCT-1 thermal analyzer, Beijing Yongjiu Instrument, Beijing, China). The measurements were performed by heating from 25 to 350 °C at four different heating rates (5, 10, 15 and 20 °C/min).
The determination of Tg was performed using the described DSC tests, except that the heating rate was 10 °C/min. That for Tg the cured samples had been tested by DSC and that an increased heat flow (endothermic behavior) shows Tg.

Mechanical Properties of EHTPB-IPDI and HTPB-IPDI Polyurethane Elastomers
The tensile properties of the EHTPB-IPDI and HTPB-IPDI binder system films were tested with an EMT-C universal material testing machine.
According to the Chinese standard: GB/T 528-1998 [24], rubber, vulcanized or thermoplastic, we determined the tensile stress-strain properties. The tensile strength and elongation at break of the specimens were measured by setting the gauge length of the specimens and the tensile speed at 25 mm and 50 mm/min, respectively. Five samples were cut from each type of the binder system film, and further analyzed. Figure 3 represents the dumbbell-shaped specimens. The data recorded in the five tests of the same binder system film were collected. The mean value of the indexes obtained from the five measurements was set as the final test result. According to the Chinese standard: GB/T 529-2008 [25], Rubber, vulcanized or thermoplastic, we determined tear strength. Tearing properties of L-shaped specimens ( Figure 4) were also tested with an EMT-C universal material testing machine at room temperature (25 °C). The gauge length of the specimens was fixed at 25 mm, and the tensile speed was 50 mm/min. Each group of the binder system films was tested for at least 5 times to assess the experiment's reproducibility.

Viscosity Monitoring Experiment
An SNB-1A digital viscometer was used to monitor the EHTPB-IPDI binder system's viscosity changes in the presence of DBTDL and TECH catalysts during the curing process. The binder system was stirred by a 29# size rotor, with a rotating speed of 0.5 rpm at 25 • C.

Determination of the Activation Energy of EHTPB-IPDI and HTPB-IBDI Binder Systems
A non-isothermal DSC method with multiple scanning rates was selected to determine the curing kinetics of the systems. Figure 5 shows the non-isothermal curves of the HTPB-IPDI system recorded at different heating rates. As the heating rate increased, the exothermic peak temperature shifted to the right, while the shape of the exothermic peak gradually widened. With the increase of heating rate, not all the reactants have a curing reaction, and the crosslinking degree of curing reaction was not high. Thermal decomposition occurs between the reactant and the cured crosslink, and the exothermic peak of decomposition overlaps with the exothermic peak of the curing reaction. Therefore, there will be a baseline shift in the DSC curve. The total heat output increased in the same manner, while the corresponding curing time decreased with the heating rate increase. These changes indicate that the curing reaction of the HTPB-IPDI system was not only thermodynamically but also a kinetically controlled process.
An SNB-1A digital viscometer was used to monitor the EHTPB-IPDI binder system's viscosity changes in the presence of DBTDL and TECH catalysts during the curing process. The binder system was stirred by a 29# size rotor, with a rotating speed of 0.5 rpm at 25 °C.

Determination of the Activation Energy of EHTPB-IPDI and HTPB-IBDI Binder Systems
A non-isothermal DSC method with multiple scanning rates was selected to determine the curing kinetics of the systems. Figure 5 shows the non-isothermal curves of the HTPB-IPDI system recorded at different heating rates. As the heating rate increased, the exothermic peak temperature shifted to the right, while the shape of the exothermic peak gradually widened. With the increase of heating rate, not all the reactants have a curing reaction, and the crosslinking degree of curing reaction was not high. Thermal decomposition occurs between the reactant and the cured crosslink, and the exothermic peak of decomposition overlaps with the exothermic peak of the curing reaction. Therefore, there will be a baseline shift in the DSC curve. The total heat output increased in the same manner, while the corresponding curing time decreased with the heating rate increase. These changes indicate that the curing reaction of the HTPB-IPDI system was not only thermodynamically but also a kinetically controlled process. Figure 5. Non-isothermal differential scanning calorimetry (DSC) curves of the HTPB-IPDI system at different heating rates. Figure 6 shows the non-isothermal curves of HTPB-IPDI and EHTPB-IPDI systems at a heating rate of 5 °C/min. It can be seen from Figure 6 that the initial temperature and peak temperature of exothermic peak of the curing reaction of the EHTPB-IPDI system were lower than that of the HTPB-IPDI system. Figure 5. Non-isothermal differential scanning calorimetry (DSC) curves of the HTPB-IPDI system at different heating rates. Figure 6 shows the non-isothermal curves of HTPB-IPDI and EHTPB-IPDI systems at a heating rate of 5 • C/min. It can be seen from Figure 6 that the initial temperature and peak temperature of exothermic peak of the curing reaction of the EHTPB-IPDI system were lower than that of the HTPB-IPDI system.
The corresponding peak temperatures, T p , of the binder systems at different heating rates, listed in Table 1, were linearly fitted. ln(β/T 2 P ) vs. 1/T p (Figure 7) was plotted according to the Kissinger method Equation (2), and the pre-exponential factor, A, and activation energy, E, were calculated from the slope and intercept of the fitting linear curves. The values of the determined pre-exponential factor, A, and activation energy, E, of the two binder systems are listed in Table 2. The corresponding peak temperatures, Tp, of the binder systems at different heating rates, listed in Table 1, were linearly fitted. ln(β/ 2 p T ) vs. 1/Tp ( Figure 7) was plotted according to the Kissinger method Equation (2), and the pre-exponential factor, A, and activation energy, E, were calculated from the slope and intercept of the fitting linear curves. The values of the determined pre-exponential factor, A, and activation energy, E, of the two binder systems are listed in Table 2.   Figure 6. Non-isothermal DSC curves of the HTPB-IPDI system and HTPB-IPDI system a scanning rate of 5 • C min −1 .    The activation energy of the EHTPB-IPDI binder system was lower than that of the HTPB-IPDI binder system, indicating the curing reaction of the EHTPB-IPDI binder system was more likely to occur.  The activation energy of the EHTPB-IPDI binder system was lower than that of the HTPB-IPDI binder system, indicating the curing reaction of the EHTPB-IPDI binder system was more likely to occur.

Analysis of the Glass Transition Temperature, T g
The glass transition temperature (T g ) of polyurethane elastomer reflects its thermal stability and thermal degradation properties. The T g value is related to the cross-linking density and chemical structure of the polyurethane elastomer. Figure 8 indicates that due to epoxy groups' introduction, T g of the EHTPB-IPDI polyurethane elastomer increased by 5 • C, and the thermal stability also improved to a certain extent. These effects originated from the reaction of the introduced epoxy groups that react with isocyanate and generate oxazolidinone (Figure 9) during the polyurethane elastomers' curing process. There are three positions of C=C in the molecular structure of HTPB ( Figure 1). In the epoxidation reaction, the double bond may be opened to form an epoxy group. With the curing reaction going on, a large number of oxazolidinones can be produced by the reaction of epoxy group and isocyanate radical. The existence of oxazolidinone in the binder system may improve the performance of polyurethane elastomers [26]. The presence of oxazolidinone in polyurethane can increase the glass transition temperature of polyurethane, which is consistent with the experimental results of Yu et al. [27], polyurethane material was prepared by the reaction of diglycidyl terephthalate and toluene diisocyanate. The glass transition temperature was 110 • C, and it had good thermal stability. Besides, the reaction between isocyanate and epoxy groups will branch the EHTPB-IPDI binder system and increase its cross-linking density, improving the elastomer's thermal stability.   Table 3 shows that the tensile strength of the EHTPB-IPDI polyurethane elastomer was 12% higher than that of the HTPB-IPDI elastomer. Besides, compared with EHTPB-IPDI, the tear strength of HTPB-IPDI increased by 17%, while its elongation at break was reduced by 10%. These changes   Table 3 shows that the tensile strength of the EHTPB-IPDI polyurethane elastomer was 12% higher than that of the HTPB-IPDI elastomer. Besides, compared with EHTPB-IPDI, the tear strength of HTPB-IPDI increased by 17%, while its elongation at break was reduced by 10%. These changes can be attributed to the fact that besides the carbamate formation ( Figure 10) via the reaction with   Table 3 shows that the tensile strength of the EHTPB-IPDI polyurethane elastomer was 12% higher than that of the HTPB-IPDI elastomer. Besides, compared with EHTPB-IPDI, the tear strength of HTPB-IPDI increased by 17%, while its elongation at break was reduced by 10%. These changes can be attributed to the fact that besides the carbamate formation ( Figure 10) via the reaction with hydroxyl groups, isocyanate in the EHTPB-IPDI elastomer can also react with epoxy groups and form oxazolidinone (Figure 9). Oxazolidinone can improve the mechanical properties of polyurethane [28]. As a result, more branched chains ( Figure 11) were generated in the elastomer, yielding the increased cross-linking density of the polyurethane elastomer and enhancing its tensile and tearing strength, which is consistent with the experimental results of Yang et al. [29], EHTPB-modified epoxy resin exhibited excellent mechanical properties than that of neat epoxy resin. Moreover, the increase of the cross-linking density may block the sliding and movement of the molecular segments, resulting in a lower fracture growth rate, which is consistent with results from a study by Shen et al. [30].   Compared with the HTPB-IPDI binder system, the EHTPB-IPDI binder system had lower Ea and better thermal stability and mechanical properties. Therefore, the EHTPB-IPDI binder system was selected for further research in this paper.

Kissinger Method
The peak temperature, Tp, of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH systems at   Compared with the HTPB-IPDI binder system, the EHTPB-IPDI binder system had lower Ea and better thermal stability and mechanical properties. Therefore, the EHTPB-IPDI binder system was selected for further research in this paper.

Kissinger Method
The peak temperature, Tp, of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH systems at 2 Figure 11. The curing reaction of EHTPB-IPDI binder system forms a network structure.
Compared with the HTPB-IPDI binder system, the EHTPB-IPDI binder system had lower E a and better thermal stability and mechanical properties. Therefore, the EHTPB-IPDI binder system was selected for further research in this paper.

Kissinger Method
The peak temperature, T p , of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH systems at different heating rates (Table 4) was plotted as ln (β/T 2 P ) vs. 1/T p according to the Kissinger method (as shown in Figure 12) and linearly fitted.  As Figure 12 shows, the linear correlation coefficients,
As Figure 13 shows, the linear correlation coefficients of the two fitting lines,   Table 4. DSC data of the TECH-catalyzed and DBTDL-catalyzed EHTPB-IPDI curing reaction. β: heating rate; T i : initial temperature; T P : peak temperature; T f : final temperature; T 1 and T 2 : curing reaction temperatures of the TECH-catalyzed and DBTDL-catalyzed EHTPB-IPDI curing reaction.
As Figure 12 shows, the linear correlation coefficients, R 2 1 and R 2 2 , of the two fitting lines were 0.9808 and 0.9994, respectively, indicating a good correlation of the linear fitting curves. The activation energies of the two systems, E 1 and E 2 , were 45.4 and 40.3 kJ/mol, respectively. The pre-exponential factors of two systems, A 1 and A 2 , were 1.32 × 10 5 and 5.42 × 10 3 s −1 , respectively.

F-W-O Method
The peak temperature, T p , of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH systems obtained at different heating rates (Table 4) was plotted as lnβ vs. 1/T p (Figure 12) then linearly fitted.
As Figure 13 shows, the linear correlation coefficients of the two fitting lines, R 2 1 and R 2 2 , were 0.9865 and 0.9995, respectively, indicating a good correlation of the two linear fitting curves. The activation energy obtained by the FWO method was slightly higher than that calculated using the Kissinger method, the reason for the slightly higher activation energy is that there is a certain difference in the formula fitting between the FWO method and the Kissinger method, but this does not affect the accuracy of the result, which is consistent with results from other studies, e.g., the studies by Lee et al. [31] and Liu et al. [32]. The difference in the activation energy could originate from the limitations of the used equations. Still, it is not related to the characteristics of the studied systems. Therefore, the kinetics of the curing reaction of the EHTPB-IPDI-TECH and EHTPB-IPDI-DBTDL binder systems can be fitted with the nth-order reaction model.

Doyle Method
The DSC data listed in Table 3 were analyzed using the Doyle method. lnβ vs. 1/Tp was plotted ( Figure 13) and linearly fitted. The linear correlation coefficients of the two fitted lines, 2 1 R and 2 2 R , were 0.9808 and 0.9994, respectively, which indicates a good linear correlation. The activation energies E1 and E2 of the two systems, calculated from the fitted lines' slopes, were 44.9 and 40.8 kJ/mol, respectively. Table 5 shows the average activation energy calculated using the Kissinger, F-W-O and Doyle methods. The average activation energy values, ΔE1 and ΔE2, of the two systems were 47.1 and 42.3 kJ/mol, respectively. The results show that the average activation energy of the EHTPB-IPDI-DBTDL binder system was lower, and the initial reaction temperature was lower and the reaction activity was higher. It has reference value for the design of curing formula at room temperature.  The activation energy obtained by the FWO method was slightly higher than that calculated using the Kissinger method, the reason for the slightly higher activation energy is that there is a certain difference in the formula fitting between the FWO method and the Kissinger method, but this does not affect the accuracy of the result, which is consistent with results from other studies, e.g., the studies by Lee et al. [31] and Liu et al. [32]. The difference in the activation energy could originate from the limitations of the used equations. Still, it is not related to the characteristics of the studied systems. Therefore, the kinetics of the curing reaction of the EHTPB-IPDI-TECH and EHTPB-IPDI-DBTDL binder systems can be fitted with the nth-order reaction model.

Doyle Method
The DSC data listed in Table 3 were analyzed using the Doyle method. lnβ vs. 1/T p was plotted ( Figure 13) and linearly fitted. The linear correlation coefficients of the two fitted lines, R 2 1 and R 2 2 , were 0.9808 and 0.9994, respectively, which indicates a good linear correlation. The activation energies E 1 and E 2 of the two systems, calculated from the fitted lines' slopes, were 44.9 and 40.8 kJ/mol, respectively. Table 5 shows the average activation energy calculated using the Kissinger, F-W-O and Doyle methods. The average activation energy values, ∆E 1 and ∆E 2 , of the two systems were 47.1 and 42.3 kJ/mol, respectively. The results show that the average activation energy of the EHTPB-IPDI-DBTDL binder system was lower, and the initial reaction temperature was lower and the reaction activity was higher. It has reference value for the design of curing formula at room temperature.

Monitoring of Viscosity
The viscosity, or precisely the viscosity growth rate of the EHTPB binder system, increases with the progress of the curing reaction. It grows rapidly around the gelation point of the system and approaches infinity when the system is completely cured. In this research, it is not completely cured, which is the early stage of curing reaction. The temperature of exothermic peak of curing reaction can be obtained from the DSC curve, full chemical curing can not be seen from finalization of the DSC peak. It can be seen from the FTIR spectrum, the disappearance of reaction groups and the formation of new groups. Figure 14 shows the viscosity increase of the EHTPB-IPDI binder system in the presence of two catalysts, DBTDL and TECH, at the curing temperature of 25 • C. The pot life of the binder system for solid rocket propellants and PBXs is usually defined as the consumed time when the system viscosity increases to 20,000-25,000 mPa·s [33,34]. From the results of viscosity build-up measurement under the same curing temperature, it is possible to conclude that the pot-life of EHTPB-IPDI-TECH system was longer than that of EHTPB-IPDI-DBTDL system. The viscosity growth rate (first derivation of the viscosity curve in Figure 14) of the EHTPB-IPDI binder system was higher for DBTDL than for TECH, indicating a higher catalytic activity of DBTDL, which is consistent with the results obtained from DSC tests. EHTPB-IPDI-DBTDL binder system had higher reaction activity and faster viscosity growth. For the curing reaction of EHTPB-IPDI binder system, the catalytic effect of DBTDL was better than that of TECH, which is consistent with the experimental results of Ma et al. [15].

Monitoring of Viscosity
The viscosity, or precisely the viscosity growth rate of the EHTPB binder system, increases with the progress of the curing reaction. It grows rapidly around the gelation point of the system and approaches infinity when the system is completely cured. In this research, it is not completely cured, which is the early stage of curing reaction. The temperature of exothermic peak of curing reaction can be obtained from the DSC curve, full chemical curing can not be seen from finalization of the DSC peak. It can be seen from the FTIR spectrum, the disappearance of reaction groups and the formation of new groups. Figure 14 shows the viscosity increase of the EHTPB-IPDI binder system in the presence of two catalysts, DBTDL and TECH, at the curing temperature of 25 °C. The pot life of the binder system for solid rocket propellants and PBXs is usually defined as the consumed time when the system viscosity increases to 20,000-25,000 mPa·s [33,34]. From the results of viscosity build-up measurement under the same curing temperature, it is possible to conclude that the pot-life of EHTPB-IPDI-TECH system was longer than that of EHTPB-IPDI-DBTDL system. The viscosity growth rate (first derivation of the viscosity curve in Figure 14) of the EHTPB-IPDI binder system was higher for DBTDL than for TECH, indicating a higher catalytic activity of DBTDL, which is consistent with the results obtained from DSC tests. EHTPB-IPDI-DBTDL binder system had higher reaction activity and faster viscosity growth. For the curing reaction of EHTPB-IPDI binder system, the catalytic effect of DBTDL was better than that of TECH, which is consistent with the experimental results of Ma et al. [15].

Curing Reaction Kinetic Equations of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH Binder Systems
After determining the ΔE of the systems, the curing reaction's order can be obtained using the Crane equation (Equation (5)). lnβ vs. 1/Tp was plotted and linearly fitted. The equations of the two fitted lines obtained using the F-W-O method, y1 = −6.457x + 15.303 and y2 = −5.797x + 14.447, were substituted into Equation (5). The curing reaction orders of the two systems (n1 and n2) were calculated to be 0.95 and 0.95, respectively. Since the reaction orders for the two systems are both non-integers, the reaction rate of curing reaction is not only related to the concentration of reactants, but also depends on other factors, so it is a complex reaction. The equations of the curing reaction kinetics of the binder systems can be expressed as follows:

Curing Reaction Kinetic Equations of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH Binder Systems
After determining the ∆E of the systems, the curing reaction's order can be obtained using the Crane equation (Equation (5)). lnβ vs. 1/T p was plotted and linearly fitted. The equations of the two fitted lines obtained using the F-W-O method, y1 = −6.457x + 15.303 and y2 = −5.797x + 14.447, were substituted into Equation (5). The curing reaction orders of the two systems (n 1 and n 2 ) were calculated to be 0.95 and 0.95, respectively. Since the reaction orders for the two systems are both non-integers, the reaction rate of curing reaction is not only related to the concentration of reactants, but also depends on other factors, so it is a complex reaction. The equations of the curing reaction kinetics of the binder systems can be expressed as follows: EHTPB-IPDI-DBTDL : d∂ dt = 1.32 × 10 5 e −5.7×10 3 /T (1 − ∂) 0.95 (6) EHTPB-IPDI-TECH : d∂ dt = 5.42 × 10 3 e −5.1×10 3 /T (1 − ∂) 0.95 (7)

Parameters Optimization in the Curing Process
DSC curves show that when the system was cured by a dynamic heating method, the curing temperature varied with the heating rate. Besides, the peak temperature also changed, so the determination of the actual curing temperature was challenging. To obtain a reliable curing temperature, the characteristic curing temperature at a zero heating rate was obtained by the extrapolation method. Specifically, the initial (T i ), peak (T p ) and final (T f ) temperatures of the curing process at the-zero heating were obtained by extrapolating the heating rate to zero ( Table 6). As shown in Figures 15 and 16, when the heating rate of the EHTPB-IPDI-TECH binder system was extrapolated to zero, T i of the curing process was 153 • C, T p of the exothermic peak was 185 • C and T f of the curing process was 198 • C; when the heating rate of the EHTPB-IPDI-DBTDL binder system was extrapolated to zero, T i of the curing process was 158 • C, T p of the exothermic peak was 164 • C and T f of the curing process was 184 • C.

Parameters Optimization in the Curing Process
DSC curves show that when the system was cured by a dynamic heating method, the curing temperature varied with the heating rate. Besides, the peak temperature also changed, so the determination of the actual curing temperature was challenging. To obtain a reliable curing temperature, the characteristic curing temperature at a zero heating rate was obtained by the extrapolation method. Specifically, the initial (Ti), peak (Tp) and final (Tf) temperatures of the curing process at the-zero heating were obtained by extrapolating the heating rate to zero ( Table 6). As shown in Figures 15 and 16, when the heating rate of the EHTPB-IPDI-TECH binder system was extrapolated to zero, Ti of the curing process was 153 °C, Tp of the exothermic peak was 185 °C and Tf of the curing process was 198 °C; when the heating rate of the EHTPB-IPDI-DBTDL binder system was extrapolated to zero, Ti of the curing process was 158 °C, Tp of the exothermic peak was 164 °C and Tf of the curing process was 184 °C.   The curing time can be expressed by the curing degree (∂) relative to the curing speed. According to the calculation results, assuming ∂ was 99%, the curing times of the two systems were 4 and 19 h. The curing temperature and the curing time of EHTPB-IPDI-DBTDL were lower than those of EHTPB-IPDI-TECH. The cured thermosetting materials of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH binder systems were obtained under the curing conditions listed in Table 7.

Conclusions
In this paper, the curing kinetics of EHTPB-IPDI and HTPB-IPDI binders were studied by the DSC method, and the related mechanical properties were measured. The initial temperature, Ti, and activation energy, Ea, of the EHTPB-IPDI curing reaction were lower than those of the HTPB-IPDI system. Compared with the HTPB-IPDI binder system, the Tg of the EHTPB-IPDI binder system was increased by 5 °C, indicating better thermal stability of the EHTPB-IPDI binder system. The tensile strength and breaking strength of the EHTPB-IPDI system were increased by 12% and 17%. In comparison, its elongation at break was decreased by 10%, indicating better mechanical properties of the EHTPB-IPDI binder system. When DBTDL and TECH were used as catalysts in the EHTPB-IPDI system, the Ea of the EHTPB-IPDI-DBTDL binder system was lower than that of the EHTPB-IPDI-TECH binder system. This indicates that the EHTPB-IPDI-DBTDL system had a higher The curing time can be expressed by the curing degree (∂) relative to the curing speed. According to the calculation results, assuming ∂ was 99%, the curing times of the two systems were 4 and 19 h. The curing temperature and the curing time of EHTPB-IPDI-DBTDL were lower than those of EHTPB-IPDI-TECH. The cured thermosetting materials of the EHTPB-IPDI-DBTDL and EHTPB-IPDI-TECH binder systems were obtained under the curing conditions listed in Table 7.

Conclusions
In this paper, the curing kinetics of EHTPB-IPDI and HTPB-IPDI binders were studied by the DSC method, and the related mechanical properties were measured. The initial temperature, T i , and activation energy, E a , of the EHTPB-IPDI curing reaction were lower than those of the HTPB-IPDI system. Compared with the HTPB-IPDI binder system, the T g of the EHTPB-IPDI binder system was increased by 5 • C, indicating better thermal stability of the EHTPB-IPDI binder system. The tensile strength and breaking strength of the EHTPB-IPDI system were increased by 12% and 17%. In comparison, its elongation at break was decreased by 10%, indicating better mechanical properties of the EHTPB-IPDI binder system. When DBTDL and TECH were used as catalysts in the EHTPB-IPDI system, the Ea of the EHTPB-IPDI-DBTDL binder system was lower than that of the EHTPB-IPDI-TECH binder system. This indicates that the EHTPB-IPDI-DBTDL system had a higher reactivity, and its curing process occurred at lower temperatures. In summary, the EHTPB-IPDI binder system had a lower reaction initiation temperature, higher reactivity and was more prone to curing reaction at room temperature. This is of reference value for the design of the room temperature curing formula of polyurethane materials. Due to the introduction of epoxy groups, the cured film of EHTPB-IPDI had good mechanical properties and thermal stability. It can be used for composite solid propellant and solid rocket motor insulation and lining materials, and has a wide range of application prospects.
Author Contributions: Y.L., Z.M., and T.C. conceived and designed the experiments; X.Z. and K.J. performed the experiments; X.Z. analyzed the data; X.Z. and K.J. contributed reagents/materials/analysis tools; X.Z. wrote the paper. All authors have read and agreed to the published version of the manuscript.