3.1 Infrared spectrum analysis
The infrared spectra of naringenin and NER are shown in Fig. 3. The characteristic peaks at 1650 cm− 1 and 1251 cm− 1 correspond to C-O-Ph and C-O-C, respectively. The characteristic peaks at 1750 cm− 1 and 1605 cm− 1 correspond to C = O and C = C, respectively. The characteristic peak at 836 cm− 1 belong to -CH of benzene ring. Both naringenin and NER contains the characteristic peaks mentioned above. In addition, the characteristic peak of NER at 910 cm− 1 belong to epoxy group. The hydroxyl absorption peaks of NER at 3100–3500 cm− 1 were obviously weakened, and the peaks at 2920 cm− 1 and 2880 cm− 1 belongs to -CH2- stretching vibration absorption. These indicate that the phenolic hydroxyl group of the naringenin molecule has been modified to an epoxy group. Therefore, it can be proved that NER was successfully synthesized. The epoxy value of NER was determined by hydrochloric acid-acetone titration. The measured epoxy value of naringenin epoxy resin was 0.62 (theoretical value is 0.68. The weak hydroxyl peak near 3480 cm− 1 indicating that the phenolic hydroxyl group was not reacted completely).
The infrared spectra of UiO-66 as shown in Fig. 4. The intense and broad characteristic peak at 3369 cm− 1 due to the intercrystalline water and the physisorbed water condensed inside the cavities. The characteristic peaks at 1581 and 1398 cm− 1 can be assigned to asymmetric and symmetric tensile vibration of O-C-O, respectively. The characteristic peak at 1506 cm− 1 belong to the C = C in the benzene ring. In addition, the presence of the characteristic peak at 1660 cm− 1 indicates the presence of DMF in the framework of UiO-66. The characteristic peak at 1100 cm− 1 belong to the stretching vibration peak of the skeleton Zr-O single bond of UiO-66. The peaks at 823, 745, and 666 cm− 1 are due to OH and C-H vibration in the terephthalic acid ligand. At lower frequencies, the characteristic peaks of Zr-O overlap with OH and C-H flexural vibration peaks (the main bands are 745, 666, 552, and 487 cm− 1, respectively) [21].
3.2 X-ray diffraction analysis
The X-ray diffraction spectrum of UiO-66 and its simulated are shown in Fig. 5.The X-ray diffraction spectrum of UiO-66 shows diffraction peaks at 2θ = 7.3°, 8.5°, 12.0°, 17.1°, 22.2°, 25.7°, which corresponding to crystal planes of (111), (002), (022), (044), (115) and (006), respectively.The peak shape, peak intensity and position of UiO-66 are agreement with the results reported in reference 20. These evidences indicate that the UiO-66 with regular structure was successfully synthesized.
3.3 Dynamic mechanical properties analysis
The internal friction factor (Tan δ) and storage modulus (E´) of each sample are shown in Fig. 6. The maximum value of internal friction peak (Tan δmax), crosslink density (Ve) and glass transition temperature (Tg) of each sample are shown in Table 2.
The Ve (sum of chemical crosslink and physical crosslink) could be obtained according to reference 23.
As seen from Fig. 6 (a), the Tan δmax of NER/UiO-66 composites are much lower than that of sample P. Tan δ is the ratio of loss modulus to storage modulus. Under the same condition, the lower the Tan δmax is the less internal loss of the material. Furthermore, the peak width at half-height of Tan δ for NER/UiO-66 composites is much broader than that of sample P, indicating the lower segmental mobility in the NER/UiO-66 composites. This is because NER matrix contains more epoxy functional groups than DGEBA, and its crosslink density after curing is much higher than that of DGEBA. UiO-66 contains polar carbonyl groups, which can make it partial compatible with epoxy resin matrix. On the other hand, UiO-66 as a rigid particle has a large steric effect. Therefore, the crosslink density of NER/UiO-66 composites decreased with increasing UiO-66 content, while the internal friction increased slightly with increasing UiO-66 content. When the content of UiO-66 is not higher than 3%, UiO-66 has good compatibility with NER, the storage modulus and Tg of NER/UiO-66 composites both increased with increasing UiO-66 content. When the content of UiO-66 is higher than 3%, the compatibility of UiO-66 and epoxy resin matrix begin to deteriorate, resulting in the decrease of storage modulus and Tg.
Table.2 Tan δmax, Ve and Tg of each sample
Sample
|
P
|
A0
|
A1
|
A2
|
A3
|
A4
|
A5
|
Tan δmax
|
1.032
|
0.155
|
0.168
|
0.178
|
0.187
|
0.207
|
0.212
|
Ve/(kmol/m3)
|
9.19
|
146.19
|
84.19
|
82.74
|
71.08
|
55.60
|
52.95
|
Tg(℃)
|
53
|
154
|
169
|
172
|
178
|
165
|
164
|
3.4 Mechanical Properties Analysis
The impact strength of each sample is shown in Fig. 7. Proper crosslink density is beneficial to the improvement of polymer mechanical properties. The impact strength of sample A0 is 4.1 kJ/m2, which is 2.0 kJ/m2 higher than that of sample P. This is because the crosslink density of NER is higher than that of sample P (as shown in Table 2). The impact strength of NER/UiO-66 composites increased first and then decreased with increasingUiO-66 content. When the UiO-66 content is 4%, the impact strength of NER/UiO-66 composite (sample A4) is the maximum, which is 3.5 kJ/m2 higher than that of pure naringenin epoxy resin (sample A0). UiO-66 contains carbonyl groups, which makes it partially compatible with naringenin epoxy resin matrix when its content is not high. When the NER/UiO-66 composite is subjected to an external force, UiO-66 can induce and terminate the crazing, thereby exerting a toughening effect [22]. However, when the content of UiO-66 exceeds 4%, defects appear in NER/UiO-66 composites due to its agglomeration, which leads to the decrease of impact strength. This phenomenon is consistent with the dynamic mechanical properties discussed above.
3.5 Morphology analysis
Figure 8 shows the impact cross section of different samples. It can be seen from the figure that the cross section of sample P and A0 are smooth and show the characteristics of brittle fracture. The section of sample A4 is relatively rough, and many wrinkles can be observed, indicating that the material absorbs a lot of energy when subjected to external forces, which corresponds to its high impact strength. The cross section of sample A5 is smooth and some particles can be observed. This is due to the agglomeration of uio-66, resulting in defects in the material, which reduces the impact strength.
3.6 Thermal stability analysis
The thermal degradation curves of different samples are shown in Fig. 9. As shown in the figure, NER/UiO-66 composites have only one degradation stage. It shows that the addition of UiO-66 did not change the thermal degradation process of naringenin epoxy. The initial thermal degradation temperature (Td5, Temperature at 95% mass retention rate [24]) of samples P and A series samples are 314 ℃, 316 ℃, 314 ℃, 302 ℃, 308 ℃, 297 ℃, and 299 ℃, respectively. The Td5 of NER/UiO-66 composites decreased with the increasing UiO-66 content. This is consistent with their crosslink density. It is noted that the residual char of NER/UIO-66 composite at 800 ℃ is significantly higher than that of DGEBA. This is because the higher the crosslink density, the easier it is to form a continuous dense char layer. The char formation will insulate the polymer-air interface, reduce the heat conduction, and starve the combustion process of decomposition products.
3.7 Limiting oxygen index analysis
The Limiting oxygen index (LOI) of sample P and A series samples are 21.0%, 22.2%, 23.3%, 23.6%, 23.8%, 24.5%, 24.5%, respectively. Since no flame-retardant fillers or elements are added to the samples, the improvement of LOI is not very significant. UiO-66 can promote the formation of char layer, thereby producing a thermal insulating layer and playing an auxiliary flame-retardant effect [19]. Thus, the LOI of NER/UiO-66 composites increased with increasing UiO-66 content.