Effect of ultrasonic pretreatment on thermo-mechanical properties of epoxy adhesive

TGA (TGA55 TA Instruments, USA) was conducted at a heating rate of 10 °C· min⁻¹ under a nitrogen stream. The two components were analyzed in the temperature range of 30 °C–1000 °C, and the cured adhesive was in the range of 30 °C–600 °C. The thermal degradation onset temperature and the thermal degradation weight loss of the specimens were investigated.

DSC (214 Polyma NETSCH Group, Germany) was used to measure the heat flow change during the curing process. The test mass was 5–7 mg. An empty aluminum pan with a cover was used as the reference. The DSC experiments were carried out under a nitrogen flow (40 ml· min⁻¹).

The specimens were fabricated into rectangular bars in a PTFE mold with dimensions of 30 mm × 6 mm × 2 mm. DMA analysis was performed using a DMA 8000 (Perkin Elmer). A single cantilever bending mode at a frequency of 1 Hz was used for the analysis. The heating rate was 2 °C· min⁻¹ and temperature range was 0 °C–150 °C.

Figure 4 shows the TGA results of the adhesive. It can be seen that the component A has high thermal stability, and its mass decreased obviously after approximately 200 °C. High temperature (above 150 °C) led to component B mass loss of more than 2.3%. The thermal degradation behavior was significant. The thermal degradation at HT caused a decrease in tensile strength and fracture toughness. The degradation phenomenon was related to the decrease in crosslinking density and the formation of linear molecules in the network [31].

Zoom In Zoom Out Reset image size

The temperatures for different weight-loss percentages were obtained from the TGA thermograms, as shown in table 3. No significant difference was observed in the initial degradation temperatures. The initial decomposition temperature of the HTC was slightly lower than that of the other specimens. The 50% weight loss for RTC occurred at approximately 383 °C, but higher temperatures of 397 °C and 401 °C were observed for MTC and USP-MTC, respectively. However, from approximately 240 °C, the weight loss rate of the HTC increased, and its thermostability worsened. The weight loss trends in the other three groups were consistent.

To study the curing reaction behavior of the adhesive, it was subjected to non-isothermal DSC measurements at different heating rates ranging from 5 °C min⁻¹ to 20 °C min⁻¹. The results are shown in figure 5(a). An abrupt drop in heat power at the initial stage was observed because of thermal equilibrium. The curing process is an exothermic reaction because of the crosslinking reaction of the adhesive. The heating range was 20 °C–180 °C. The characteristic parameters (onset temperature ${T}_{i}$, peak temperature ${T}_{p}$, terminal temperature ${T}_{f}$ and total exothermic reaction heat ${\rm{\Delta }}H$) can be obtained from the non-isothermal DSC curves. The data are summarized in table 4. Only one exothermic peak is observed during the entire curing process. As the heating rate increased, the peak temperature shifted to a higher value as well as the enthalpy. This behavior can be attributed to the increased thermal effect [32]. The T-β extrapolation was utilized to determine the theoretical curing temperature by fitting the characteristic temperatures at different heating rates β. The theoretical values of ${T}_{i}$, ${T}_{p}$ and ${T}_{f}$ for the adhesive were 28.775 °C, 69.625 °C and 118.07 °C, as shown in figure 5(b).

Zoom In Zoom Out Reset image size

Table 4. Characteristic temperatures of the adhesive at different heating rate.

Heating rate (o C min−1)	${T}_{i}$ ( oC)	${T}_{p}$ ( oC)	${T}_{f}$ ( oC)	${\rm{\Delta }}T$ (oC)	${\rm{\Delta }}H$ (J· g−1)
5	36.08	77.91	128.2	92.12	78.6
10	41.18	88.87	130.1	88.92	106.9
15	48.23	94.23	134.9	86.67	161.1
20	55.79	105.44	150.4	94.57	206.0

The α-T curves were derived from non-isothermal DSC curves, as shown in figure 5(c). α·is the curing degree, and T is the temperature. The 'S'-shaped curve indicates an autocatalytic reaction. The increased heating rate led to a right shift in the temperature of the maximum curing reaction rate. During the curing reaction, the curing rate first increased and then decreased. The adhesive changed into a gel state as the curing reaction proceeded. It can be observed that the curing rate showed a large increase between 60 °C and 120 °C.

Isothermic tests at 60 °C and 150 °C were carried out, as shown in figure 6(a). In the tests, the heat flow was monitored until the gradient was zero. The adhesive was virtually cured after this time. The curing times at 60 °C and 150 °C were 55.9 and 17.6 min, respectively, but the time of the ultrasonic pretreated adhesive was only 28.4 min Table 5 shows the curing cycles of the different curing schemes from isothermic DSC test results. Compared with the traditional thermal curing, ultrasonic-assisted thermal curing can shorten the curing cycle by 50%.

Zoom In Zoom Out Reset image size

Tg value can be used to characterize the thermal stability of the cured specimens. To determine Tg, the heat flow of the cured adhesive specimens was monitored by DSC analysis, as shown in figure 6(b). It is observed that·the Tg values of RTC, MTC, and USP-MTC were 66.5 °C, 72.93 °C and 76.74 °C, respectively. The increase in Tg·indicates that the crosslinking and ordering of the chains increased. The·Tg·value of HTC was 69.58 °C because of chain scission in the crosslinking network at the high temperature.

Figure 7(a) shows a plot of the storage modulus and loss modulus against temperature for the cured specimens. In the temperature scanning in the DMA test, three states of the adhesive were observed: energy elastic state at low temperatures (0 °C–30 °C), glass transition state at medium temperature (31 °C–80 °C) and entropy elastic state at high temperature (greater than 80 °C). When the state changes from the glass state to the rubber state, the oscillation is mainly converted into internal friction and non-elastic deformation, and the loss modulus also reaches its maximum. The complex modulus E* is the ratio of the stress amplitude to the deformation amplitude, and is calculated by

$$\begin{eqnarray}&&E* =\displaystyle \frac{\sigma * }{\varepsilon * }=E^{\prime} +iE^{\prime\prime} \end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&E^{\prime} \left(w\right)=\left|E* \right|\cos \,\delta \end{eqnarray} \tag{ 3 }$$

$$\begin{eqnarray}&&E^{\prime\prime} \left(w\right)=\left|E* \right|\sin \,\delta \end{eqnarray} \tag{ 4 }$$

where E' is the storage modulus, and E'' is the loss modulus. They represent the dynamic elastic properties of the cured adhesive, depending on the frequency and history of the adhesive specimens [33].

Zoom In Zoom Out Reset image size

Compared with RTC, MTC exhibited a higher storage modulus. The increase in storage modulus at MT can be attributed to the increase in crosslinking. The storage modulus of HTC decreased. This drop in storage modulus indicates an increase in the flexibility of the epoxy system at HT. The movement of molecules and chains has become easier [23, 34].

The damping property of the epoxy adhesive is defined by the ratio of energy dissipation (loss modulus) to energy storage (storage modulus). The loss factor is calculated by

$$\begin{eqnarray}&&\tan \,\delta =E^{\prime\prime} \left(w\right)/E^{\prime} \left(w\right)\end{eqnarray} \tag{ 5 }$$

Figure 7(b) shows the loss factor versus temperature. The temperature at the maximum $\tan \,{\delta }\,$value is the Tg of the adhesive. The temperature range of $\tan \,\delta \gt 0.3$ is used as a criterion to evaluate the damping property of the adhesive [2]. The temperature range for USP-MTC was 54.2° C–72.5 °C, and corresponding peak of $\tan \,{\delta }\,$ was 0.79. For RTC, MTC and HTC, the range were 38.3 °C–57.6 °C, 57.1°C –79.4 °C and 45.7 °C–65.1 °C, with corresponding peaks of 0.58, 0.78, and 0.73. The right shift of the temperature range indicates that the adhesive cured at MT has a better damping property.

The Tg value measured by the DMA test showed the same trend as that measured by the DSC test. Compared with the Tg values measured by DSC, the values measured by DMA were lower. This difference is due to the excitation frequency used in the DMA test. The maximal Tg value (69.3 °C) was found for MTC. The RTC showed a lower value of 49 °C. The increase in temperature intensified the intermolecular movement and increased the collision between functional groups, thus increasing the crosslinking degree. HTC showed a lower value of 51.7 °C, because high temperature caused scission of the chain in the crosslinking network, resulting in adhesive aging [1, 34]. USP-MTC showed a value of 65 °C, indicating performance similar to that of MTC.

In the curing reaction of the bisphenol A epoxy resin-amine system, the epoxy group and primary amine are gradually consumed to form secondary amines, tertiary amines, and hydroxyl groups. The curing reaction includes three steps. In the first step, an epoxide ring reacts with a primary amine to produce , which includes a secondary amine and a hydroxyl group. Second, the secondary amine in the formed product can also react with the epoxy group to produce , and thus a tertiary amine and a hydroxyl group are formed. The third possible step is etherification between the hydroxyl group and epoxy group, and the reaction product is . A tertiary amine connects two straight chains with the N to form a branched structure, so that the tertiary amine content mainly determines the crosslinking degree [26, 35, 36].

Figure 8 summarizes the partial FTIR spectra during the curing process at 60 °C and the ultrasonic pretreatment process. Compared with the spectrum of the uncured adhesive, a decrease in the epoxy group (-CH(O)CH-) and the amine group (N-H), and an increase in the hydroxyl group (–OH) are observed in the cured adhesive. The peak at 914 cm⁻¹ is attributed to the stretching vibration of the epoxy group [37]. The peaks at 3305 cm⁻¹ and 3361⁻¹ are due to the stretching vibration of N-H. Because of the hydroxyl group (–OH) formation, a characteristic peak at 3423 cm⁻¹ is observed. The peak at 1607 cm⁻¹ was caused by stretching vibration of the aromatic ring, which did not participate in the curing reaction. Compared with those of the thermally cured specimen, the positions of the characteristic absorption peaks of the ultrasonically pretreated specimen did not change. This means that ultrasonic treatment did not change the chemical structure of the adhesive.

Zoom In Zoom Out Reset image size

According to the above reaction introduced in section 3.4.1, the ring-opening addition reaction between the epoxy group and amine group produces a hydroxyl group. The decrease in the epoxy group and the increase in the number of hydroxyl groups determine the conversion of the epoxy group into the corresponding polymer, as well as the crosslinking process [27]. OMNIC software (version 9.2) was used to analyze the FTIR results. The peak at 1607 cm⁻¹ (stretching vibration of the aromatic ring) can be used as a benchmark for calculating the absorbency of the epoxy groups according to ISO 20368:

$$\begin{eqnarray}&&Cr=\left[1-\displaystyle \frac{\left({A}_{914,t}\right)\left({A}_{1607,0}\right)}{\left({A}_{914,0}\right)\left({A}_{1607,t}\right)}\right]\times 100 \% \end{eqnarray} \tag{ 6 }$$

According to equation (6), the epoxy conversion versus time curves at different temperatures was obtained, as shown in figure 9. The increase in temperature not only accelerated the crosslinking reaction rate, but also increased the crosslinking degree. Curing degrees of 92.66%, 94.62% and 96.33% were found at 60 °C, 80 °C, and 100 °C, respectively. As the temperature increased, the slope of the curve increased, indicating an increase in the crosslinking rate. The results show that the curing reaction was completed within 60 min at a curing temperature of 60 °C. This result was consistent with that of the isothermal DSC test. The crosslinking rate was accelerated under the ultrasonic vibration. The epoxy conversion reached 47.5% within 1 min (figure 9).

Zoom In Zoom Out Reset image size

During the thermal curing process, the consumption rate of the epoxy group was much lower. Ultrasonic vibration promotes mass transfer between the epoxy resin and the curing agent and increases the collision probability between the reactants, thus increasing the reaction rate [25, 26]. During the ultrasonic pretreatment process, the non-thermal effect of the ultrasonic treatment greatly increases the epoxy-amine reaction rate.