Supplementary Information
Sandwiched Epoxy-Alumina Composites with Synergistically Enhanced Thermal Conductivity and Breakdown Strength
Zhengdong Wang,a Yonghong Cheng,a,* Hongkang Wang,a Mengmeng Yang,a Yingyu Shao,a Xin Chen,a Toshikatsu Tanakab
a Center of Nanomaterials for Renewable Energy, State Key Laboratory of Electrical Insulation and Power Equipment, Xi'an Jiaotong University, Xi'an 710049, China.
b IPS Research Center, Waseda University, Kitakyushu, Fukuoka, Japan
* E-mail:
Table S1 Characteristic breakdown strength of nA/epoxy composites with different nA contents.
composite / E0 (kV/mm)Neat epoxy / 64.45
1 wt.% nA / 65.38
2 wt.% nA / 66.32
3 wt.% nA / 67.82
5 wt.% nA / 63.11
Fig. S1 TGA curves of the sandwiched epoxy-alumina composites with different mA content within the outer layers under nitrogen atmosphere.
The typical weight loss is observed at approximately 380 °C. The eventual stable positions of all curves are higher than that of actual filler loading, which is believed to be incomplete combustion of samples in nitrogen atmosphere. In addition, the sandwich composites show a high heat resistance and no obvious weight loss is detected before 250 °C, which would be beneficial for electronic and electric devices.
Fig. S2 The experimental and theoretical thermal conductivity of mA/epoxy composites with various mA contents.
Thermal conductivity behaviors of the mA/epoxy composite are theoretically analyzed by combining experimental data and thermal conductivity model, respectively. Three theoretical models are used in this work to further analyze the filler effect on thermal conductivity of the single layer composite. The Three models are as followed:
Maxwell model[1]
Kc=KmKf+2Km+2Φf（Kf-Km）Kf+2Km-Φf（Kf-Km） ⑴
where Kc, Km, and Kf are the thermal conductivities of the composite, matrix, and, filler respectively, and Φf are the volume fraction of the filler.
Buggeman model[2]
1-Φf=Kf-KcKf-Km(KmKc)13 ⑵
where Kc, Km, Kf, and Φf are defined as same as in eq. (1)
Y. Agari model[3]
logKc=Φf·C2·logKf+1-Φf·logC1·Km ⑶
where Kc, Km, Kf, and Φf are defined as same as in eq. (1) and eq. (2), C1 is a factor relating to the effect of the filler on the secondary structure of the polymer, and C2 is a factor relating to the ease in forming conductive chains of the filler. In eq. (3), logarithms of the thermal conductivity of the composite increases linearly with the volume fraction of the filler, constants C2 and C1 are experimentally determined.
The experimental and theoretical thermal conductivity value of composites is depicted in Figure S2. For the Maxwell model, thermal interaction between particles had not been taken into consideration, and thus lead to the lower theoretical results for higher filler fractions.[4] In other words, Maxwell model ignores the effect of filler interaction to form thermal conduction paths and contact on thermal conductivity. Both Agari model and Bruggeman model give a relatively better speculation than the Maxwell model, in the whole experimental weight fraction range. Especially for Agari model, theoretical results predicted are found to be in reasonably good accordance with experimental data, with a deviation in the percentage of only 5%. The values of C1 and C2 calculated by eq. 3 are around 0.74 and 0.70, respectively, which are relatively close to those (approximately 0.85 and 0.75, respectively) proposed by Agari. It is attributed to the shape of fillers used, because Bruggeman model is appropriate for spherical particles, while both shape and size of micro-Al2O3 particles used in the work are irregular. However, Agari model is suitable to complex systems and high filling content in composite. Hence, Agari model gives a more reasonable prediction, which is closer to the experimental data values in comparison with Bruggeman model.
1. Progelhof RC, Throne JL, Ruetsch RR (1976) Methods for predicting the thermal conductivity of composite systems: A review. POLYM ENG SCI 16 (9):615-625. doi:10.1002/pen.760160905
2. Bruggeman DAG (1935) Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen. I. Dielektrizitätskonstanten und Leitfähigkeiten der Mischkörper aus isotropen Substanzen. Annalen der Physik 416 (7):636-664. doi:10.1002/andp.19354160705
3. Agari Y, Uno T (1986) Estimation on thermal conductivities of filled polymers. J APPL POLYM SCI 32 (7):5705-5712. doi:10.1002/app.1986.070320702
4. Yung KC, Zhu BL, Wu J, Yue TM, Xie CS (2007) Effect of AlN content on the performance of brominated epoxy resin for printed circuit board substrate. J POLYM SCI POL PHYS 45 (13):1662-1674. doi:10.1002/polb.21201