The influence of the interface between mica and epoxy matrix on properties of epoxy-based dielectric materials with high thermal conductivity and low dielectric loss

Abstract

Achieving a better interface between epoxy and mica particles is the key issue to realize desirable thermal conductivity and low dielectric loss in epoxy-based dielectric materials. In this paper, we report two strategies to improve the interface between synthetic mica and epoxy resin. Polymer coated mica and organically modified mica were prepared by coating polydopamine (PDA) films and grafting a silane coupling agent (KH 560) and fluoro-chemicals via self-polymerization and in situ grafting. The thermal conductivity, dielectric loss and mechanical properties of the epoxy/functionalized mica dielectric materials were markedly improved in comparison with that of the epoxy/mica dielectric materials. The results suggests that the mechanical and dielectric properties of the epoxy/functionalized mica dielectric materials both depended on a good interface and good adhesion between mica and the epoxy matrix. The thermal conductivity of the epoxy/functionalized mica dielectric materials results from a balance of interfacial thermal resistance and quality of the interface. The better overall performance of the epoxy/mica@PDA and mica@PDA@F dielectric materials is attributed to the better interface between mica and epoxy matrix, and provided a new route for preparing epoxy-based dielectric materials with high thermal conductivity and low dielectric loss for use in electronics and the electrical industry.

---

1. Introduction

Epoxy-mica composites have been widely used for materials for marine coatings,^1,2 electronic packaging³ and high voltage applications^4,5 because of their good anticorrosion properties, high thermal stability and excellent insulating properties.⁶ However, epoxy-mica insulation systems, especially insulating materials employed in high-voltage equipment, are always exposed to thermal, electrical and mechanical stresses during operation.⁷ The simultaneous application of these stresses results in aging which leads to changes in the physical and chemical properties of the materials.^8–10 Even small alterations in the properties of the insulation materials are harmful to the service operation and reliability of the equipment which leads to final insulation rupture. The delamination of mica layers, cracks between mica grains and voids within composites are the major causes that deteriorate the electrical and mechanical properties of epoxy-mica insulation systems,¹¹ and the thermal aging of deteriorating epoxy-mica insulating materials is a chemical process of chemical structure change, such as molecular decomposition and oxidation of epoxy resin materials, which has a strong influence on the mechanical fatigue.^12,13

Epoxy-mica insulation materials are an essential part of electric utilities and their failure severely affects their reliability and determines their lifetime in most cases.¹⁴ Therefore, it is essential to investigate the performance of epoxy-mica insulation materials, not only to extend the lifetime of electric utilities, but also to improve the quality of the epoxy-mica insulation materials. In recent years, in order to produce reliable and economical electric utilities and epoxy-mica insulation systems, extensive researches have been conducted in testing of epoxy-mica insulation materials¹⁵ and have led to many diagnostic tools to assess the deterioration of the epoxy-mica insulation materials.^16–20 To our knowledge, little attention has been given to the thermal conductivity of the epoxy-mica insulation materials and the interface between mica and the epoxy matrix.

Synthetic mica is an expandable layered silicate with high crystallinity, purity and reproducibility,^21,22 which is widely used to prepare polymer/mica nanocomposites for improving electrical, mechanical and physical properties.^23–25 Although many attempts have been focused on exfoliation of synthetic mica into random platelets,^26–29 the exfoliation and fine dispersion of synthetic mica may be high cost and is not easily achieved.

A large amount of previous research has proved that an aqueous solution of dopamine (DA) at a weak alkaline pH can form thin, surface-adherent polydopamine (PDA) films by self-polymerization; such films show excellent ability to bind strongly to virtually all substrates and provide secondary reactivity with various functional groups, including thiol, amine and quinone itself, to form covalently grafted functional layers because of its rich amino groups and the quinone structure.^30–36 Therefore, dopamine (DA) has been widely used to prepare functional inorganic fillers for enhanced interfacial interaction between polymer matrix and filler, so as to enhance their thermal, dielectric and mechanical properties.^37–44

In this investigation, we prepared four functionalized synthetic micas with dopamine, KH 560 and fluoro-chemicals, and fabricated the epoxy-mica composites. The mechanical, thermal conductivity and dielectric properties of the epoxy-mica composites were investigated, and it was found that the interfacial adhesion of composites was greatly enhanced by hydrogen bonding between the epoxy matrix and mica. It is observed that the epoxy-mica composites have high thermal conductivity, excellent dielectric and mechanical properties.

2. Experimental

2.1 Materials

Synthetic mica (mica) and 4-(trifluoromethyl)benzylamine (98%) were obtained from Huajing Mica Co., ltd and Aladdin Chemistry Inc, respectively. Glycidoxypropyltrimethoxysilane (KH 560), methyl tetrahydrophthalic anhydride (MTPHA) and N-benzyldimethylamine (BDMA) were both obtained from Guangzhou Lunliqi synthetic resins co., Ltd. DA–HCl (98%) and tris(hydroxymethyl)aminomethane (Tris, 99%) were acquired from Sigma-Aldrich Co., Ltd. DHDP modified bisphenol F epoxy resin (25 wt%) was synthetized by an industrial method.⁴⁵ Ethanol and other solvents were obtained from Shanghai Reagent Co., Ltd. All materials were not purified further due so as simulate industrial processing.

2.2 Preparation of polymer coated mica particles (mica@PDA)

The preparation of functionalized mica particles was performed as follows: 10 g of mica particles were dispersed into 200 mL, 10 mM Tris–HCl aqueous solution (pH = 8.5). The mixture was then treated in an ice-bath by ultrasonication for 30 min, and after addition of 1 g DA–HCl, the mixture was treated in an ice-bath by ultrasonication for another 10 min. Finally, the above mixture was stirred vigorously for 24 h at 25 °C and the brown functionalized mica particles were collected by centrifugation followed by washing with ethanol three times, and dried under vacuum at 60 °C for 24 h. The obtained functionalized mica particles were denoted as mica@PDA.

2.3 Fluoro-chemical functionalization of polymer coated mica particles (mica@PDA@F)

2000 mg mica@PDA was dispersed in 300 mL ethanol solution and the mixture was treated in an ice-bath by ultrasonication for 15 min, 53 mg 4-(trifluoromethyl)benzylamine was added to the mixture which was then treated by ultrasonication for another 10 min. Finally, the mixture was allowed to stir at room temperature overnight. The fluoro-chemical functionalized polymer coated mica particles were collected by centrifugation followed by washing with ethanol three times, and dried under vacuum at 60 °C for 24 h. The obtained brown fluoro-chemical functionalized polymer coated mica particles were denoted as mica@PDA@F.

The schematic illustration for the preparation process of mica@PDA and mica@PDA@F is shown in Fig. 1.

Fig. 1 Schematic illustration for the preparation of mica@PDA and mica@PDA@F.

2.4 Preparation of in situ grafted mica particles

20 g of mica was added into a three-neck flask with a stirrer and reflux condensers, which was put in a thermostatted oil-bath and protected by nitrogen. 0.2 g KH 560 was added via a dropping funnel over a period of 10 minutes when the temperature reached 120 °C. The reaction was continued for 120 min and then cooled down to 25 °C, and the collected white product was denoted p-mica. Alternatively, 0.14 g 4-(trifluoromethyl)benzylamine was added to the three-neck flask within 5 min when the oil temperature was 130 °C, the reaction continued for 60 min, and the collected white product was denoted f-mica when the temperature was decreased to 25 °C. None of the products were purified further before preparing epoxy-based dielectrics materials. Samples of mica, p-mica and f-mica used for further characterization were dispersed in ethanol for centrifuging and this cycle was repeated three times, and then the samples were dried under vacuum at 60 °C for 24 h. Fig. 2 shows the schematic illustration for the preparation process of p-mica and f-mica.

Fig. 2 Schematic illustration for the preparation of p-mica and f-mica.

2.5 Preparation of epoxy-based dielectric materials

20 g the DHDP modified bisphenol F epoxy resin (25 wt%), 8 g MTHPA and 0.10 g BDMA were added into a 100 mL beaker, and the mixture was stirred to a give a uniform liquid at 120 °C. Then 28 g mica particles were gradually added to the epoxy matrix and stirred continuously until the mixture became homogeneous. The ternary mixtures were poured into the different molds and cured in an oven at 135 °C for 2 h and at 150 °C for 12 h. The samples were polished suitably for further testing.

2.6 Characterization

Fourier-transform infrared spectroscopy (FTIR) was recorded by a Perkin-Elmer Paragon1000 instrument over range of 4000–500 cm⁻¹. Surface area (BET) isotherms were obtained at liquid-nitrogen temperature of 77 K using a Micrometrics ASAP 2460 automated gas adsorption system on samples degassed at 150 °C for 10 h under reduced pressure. The specific surface areas were determined using multipoint BET analysis. X-Ray diffraction (XRD) patterns were collected on a powder diffractometer (D/max-2200/PC, Rigaku, Japan) using Cu-Kα radiation (40 kV, 20 mA). Diffraction patterns were collected from 5 to 30° at a speed of 1° min⁻¹. Field emission scanning electron microscopy (FE-SEM, JSM-7401F, JEOL Ltd., Japan) was used to observe the morphology of mica and the fractured surfaces of epoxy-based dielectric materials after impact testing. All samples are sputtered with a layer of gold to avoid charge accumulation. Transmission electron microscopy (TEM, JEM-2100, JEOL Ltd., Japan) was performed to observe the morphology of all mica particles, the mica samples being prepared by dropping a sample solution onto an ultra-thin carbon coated copper grid and air-drying. Thermogravimetric analysis (TGA) was carried out with a NETZSCH TG 209 F3 at a heating rate 20 °C min⁻¹ in a nitrogen flow of 20 mL min⁻¹. Dynamic mechanical thermal analysis (DMTA) was performed on a TA Instruments DMA Q 800 equipped with a liquid-nitrogen apparatus in a single cantilever mode. Samples were heated to 200 °C with a rate of 3.0 °C min⁻¹ and the frequency was 1.0 Hz. The specimen dimension was 25 × 5.0 × 2.0 mm³. The thermal conductivity (λ, W m⁻¹ K⁻¹) of bulk specimens was measured on disk samples in the temperature range from 25 to 165 °C with 20 °C intervals by a laser flash method (NETZSCH LFA 467).

A universal testing machine (MTS-SANS CMT 4304, Shenzhen SANS Testing Machine Co, Ltd) was used to measure the flexural modulus of the nanodielectrics according to ISO 178:2001. The specimens were bended with a support span of 66.0 mm at a loading speed of 10 mm min⁻¹ for the entire flexural tests. Charpy impact tests were determined according to ISO 179-1:2000 on an impact tester (RAY-RAN Test Equipment Ltd., UK) with hammer weight of 0.818 kg and hammer rate of 3.5 m s⁻¹. The size of the specimens was 80 × 10 × 4 mm³, at least ten specimens of the same sample were successfully tested for the flexural and Charpy impact tests. In addition, the fracture surface of the broken specimens tested by the Charpy impact testing was examined with SEM.

The dielectric properties were measured using a Novocontrol Alpha-N high-resolution dielectric analyzer (Concept 40, Germany) at room temperature and in the frequency range 10⁻¹ to 10⁶ Hz, applying 1 V voltage across two opposite sides of the samples and a layer of gold was evaporated on both surfaces to serve as electrodes. The AHDZ-10/100 dielectric strength tester (Shanghai Lanpotronics Co., China) was used to measure the breakdown strength. The specimens with the thickness of around 0.5 mm were placed between two 10 mm diameter copper ball electrodes and the electrode system containing the measured sample was immersed in pure silicone oil in order to prevent surface flashover. A 50 Hz ramp voltage was applied across two ball typed electrodes and was increased with a rate of 2 kV s⁻¹ until the sample was punctured. Fourteen specimens for each sample were measured.

3. Results and discussion

3.1 Characterization

Fig. 3 shows digital photographs of mica, mica@PDA, mica@PDA@F, p-mica and f-mica. It is obvious that mica@PDA and mica@PDA@F were brown while the particles of p-mica and f-mica were still white. The particles of the functionalized micas were smooth and did not tend to aggregate. However, the unmodified mica particles tended to aggregate together. The macro forms of the micas provided direct visual evidence of successfully functionalization.

Fig. 3 Digital photographs of mica and functionalized micas.

FTIR analysis of all micas was carried out to identify whether the micas were functionalized or not. Fig. 4(a) showed peaks at about 2800–3000 cm⁻¹ which were related to the –CH₂ and –CH₃ antisymmetric stretching vibrations. The peaks at about 1600 cm⁻¹ of p-mica and f-mica were assigned to the flexural vibrations associated with –OH. Peaks at about 1700 cm⁻¹, expected for –C=O of PDA, were not observed for mica@PDA, mica@PDA@F in Fig. 4³¹.

Fig. 4 FTIR spectra and TGA curves of mica and functionalization micas.

Fig. 4(b) shows the thermal weight loss for mica, mica@PDA, mica@PDA@F, p-mica and f-mica. It can be seen from Fig. 4(b) that the degradation of mica, mica@PDA, mica@PDA@F, p-mica and f-mica is small and the degradation of mica@PDA, mica@PDA@F, p-mica and f-mica is slightly larger than that of mica. The initial degradation and final degradation temperature of p-mica and f-mica are 620, 633.7 °C and 337.1, 361.7 °C, respectively, while the initial degradation and final degradation temperature of mica@PDA and mica@PDA@F are 494.4, 713.2 °C and 649.4, 761.1 °C, respectively. The result illustrates that the binding between chemicals and mica is through chemical reaction rather than physical adsorption.² The above FTIR and TGA analyses evidenced that the functionalization of mica was successful.

3.2 XRD analysis

The change in the interlayer spacing of the lattice structure of the layered silicates is usually monitored by XRD. The synthetic fluorinated mica (mica) has a similar layered structure with montmorillonite (MMT), and many previous investigations have shown that mica can be intercalated with organic surfactants and lead to significantly improved physical properties of polymer/mica composites.^29,50 Fig. 5 shows the XRD patterns of all mica particles from which it was observed that the functionalization process did not lead to intercalation of the layered structure of mica.

Fig. 5 XRD patterns of mica and functionalization micas.

3.3 Morphology of mica particles

Fig. 6 presents the SEM images of all mica particles. It is noteworthy that the mica tended to aggregate to form clusters, while functionalized micas with different chemicals showed a better dispersion and less agglomeration. The SEM images showed that the particle sizes of the functionalized micas covered by a polymer layer, such as mica@PDA and mica@PDA@F were smaller than the in situ micas such as p-mica and f-mica.

Fig. 6 SEM images of micas: (a) mica, (b) mica@PDA, (c) mica@PDA@F, (d) p-mica, (e) f-mica.

The morphology of mica and functionalized micas was investigated by TEM as shown in Fig. 7. It is obvious from Fig. 7(b) that a stable and dense PDA polymer shell is directly coated on the surface of mica with the thickness of the PDA polymer shell being about 5–10 nm. Fig. 7(a) and (c) do not show any polymer layer on the mica. The results further confirm that the PDA polymer was successfully grafted onto the surface of the mica particles.

Fig. 7 TEM images of mica: (a) mica, (b) mica@PDA, (c) p-mica.

3.4 BET analysis

A Micromeritics ASAP 2460 instrument was used to determine the specific surface area of the samples. The specific surface area of samples was calculated according to the Brunauer–Emmett–Teller (BET) method from the linear part of the nitrogen adsorption isotherms. The nitrogen adsorption–desorption isotherm data are listed in Table 1. To our surprise, a significant decrease of the BET surface area was observed for p-mica, from 1.3271 to 0.8274, while the BET surface area of the mica@PDA was increased to 1.5793. The functionalization process of mica reduced the structural damage of the mica surface and reduced mica agglomeration; the absence of structural damage decreased the BET surface area, while reduced agglomeration increased the BET surface area. According to this explanation, the variation of the BET surface area should be attributed to the repair of structural damage and less agglomeration. In addition, it is found in Table 1 that the fluoro-chemicals decreased the BET surface area of functionalized mica. The results could be explained in terms of the electronegativity of fluorine.

Table 1 BET surface area of mica and functionalized micas

	BET surface area/m^2 g^−1
Mica	1.3271
p-Mica	0.8274
f-Mica	0.6286
Mica@PDA	1.5793
Mica@PDA@F	1.3120

---

3.5 Morphology of epoxy-based dielectric materials

Fig. 8 shows the SEM images of the fracture surface of epoxy-based dielectric materials after impact testing. It is observed from Fig. 8(a) that there was a clear interface and the large gaps between mica and epoxy matrix due to their weak affinity; the results indicated the poor dispersion of mica in the epoxy matrix due to weak interactions between polar unmodified mica and the organic matrix. In contrast, Fig. 8(b)–(e) illustrated a good dispersion and the smooth fracture of functionalized micas and epoxy matrix, which suggests a better compatibility and affinity between functionalized micas and epoxy matrix.

Fig. 8 SEM micrographs of epoxy-based dielectric materials: (a) epoxy/mica, (b) epoxy/mica@PDA, (c) epoxy/mica@PDA@F, (d) epoxy/p-mica, (e) epoxy/f-mica.

3.6 Dynamic mechanical behavior of epoxy-based dielectric materials

DMA was performed to measure the response of a given material to an oscillatory deformation with temperature. Storage modulus, loss modulus and tan δ are three main parameters; storage modulus and loss modulus correspond to the elastic and plastic response to the deformation, respectively. The ratio of storage modulus and loss modulus is tan δ and is used to determine the occurrence of molecular mobility transitions. Here, DMA analysis has been studied to track the temperature dependence of storage modulus, loss modulus and tan δ of epoxy-based dielectric materials. Fig. 9 shows the storage modulus, loss modulus and tan δ of epoxy-based dielectric materials as a function of temperature. It is surprising (Fig. 9(a)) that the storage modulus of epoxy/functionalized micas dielectric materials was lower than that of epoxy/mica dielectric materials at temperatures below T_g; this behavior has been ascribed to the increased rubbery properties in the organic–inorganic interface neighborhood of functionalized micas. Above T_g, the storage modulus of epoxy/functionalized mica dielectric materials exhibited higher values except the epoxy/f-mica dielectric materials. The epoxy thermosets became soft above T_g, the reinforcement effect of the mica particles became prominent, and strong enhancement of storage modulus appeared due to the restricted movement of the epoxy chains within the epoxy/functionalized mica dielectric materials.⁴⁹ However, the storage modulus of the epoxy/f-mica dielectric materials was lower than that of all other epoxy based dielectric materials over the whole range. The possible reason might be that there was the longest chemical bonding between mica and epoxy matrix within the epoxy/f-mica dielectric materials, which provided the best rubbery properties below T_g and less restricted movement above T_g.

Fig. 9 Dynamic mechanical spectra of epoxy-based dielectric materials: (a) storage modulus, (b) tan δ.

Fig. 9(b) reveals that the presence of functionalized micas led to a significant shift and broadening of the tan δ curves for all epoxy/functionalized mica dielectric materials compared to that of epoxy/mica dielectric materials. The broadening of the tan δ curves was due to the unrestricted segmental motions at the better interfaces between functionalized micas and epoxy matrix. In addition, a significant shift of the tan δ curves is attributed to the action of the polymer layer, the better organic/inorganic interface and the fine dispersion of functionalized micas in the epoxy matrix. To our surprise, Fig. 9(b) shows that the epoxy/f-mica dielectric materials showed the highest T_g, the reasons being no polymer layer between f-mica and epoxy matrix, a good organic/inorganic interface and the fine dispersion of functionalized micas in the epoxy matrix.

3.7 Thermal stability of epoxy-based dielectric materials

Better thermal stability of epoxy-based dielectric materials means they should withstand a higher operation temperature and maintain their properties for a given period of time, TGA was used to study the thermal stability and weight loss of epoxy-based dielectric materials and the samples were heated from 50 to 1000 °C to ensure the organic compound was entirely degraded. Two assumptions were made for the TGA study: (1) the degradation of epoxy is the only factor of the composites weight loss when the samples are subjected to a heating program under nitrogen or (2) the filler does not affect the degradation kinetics of the epoxy resin due to their high melting temperatures.³ Fig. 10 and Table 2 show the thermal weight loss data, DTG curves and characteristic thermal parameters of epoxy-based dielectric materials. It can be seen from Fig. 10(a) that the thermal weight loss of epoxy/functionalized micas are less than that of epoxy/mica, and it is seen from Fig. 10(b) that all samples displayed one step of degradation behavior. The results suggested that the improved interfacial crosslinking between mica and epoxy matrix will lead to higher thermal stability to weight loss. According to Table 2, for the epoxy/in situ mica dielectric materials, the initial thermal decomposition temperatures (T_5%), 50% maximum thermal decomposition temperatures (T_50%) and maximum thermal decomposition temperatures (T_max) are a little improved. However, those of epoxy/mica@PDA and epoxy/mica@PDA@F are a little decreased. The residue of epoxy-based dielectric materials in 1000 °C ranged from 44.14 to 57.45% with the data deviating slightly from the 50% theoretical value. The results of different thermal decomposition temperatures and the thermal weight loss suggested that the polymer layer between mica and epoxy matrix slightly reduced the thermal stability of the epoxy-based mica dielectric materials and improved the thermal weight loss of the epoxy-based mica dielectric materials due to the improved interfacial crosslinking between mica and epoxy matrix.

Fig. 10 TG and DTG curves of epoxy-based dielectric materials: (a) TG curves; (b) DTG curves.

Table 2 Characteristic thermal parameters of epoxy-based dielectric materials

	Material
	Epoxy/mica	Epoxy/p-mica	Epoxy/f-mica	Epoxy/mica@PDA	Epoxy/mica@PDA@F
T5%/°C	391.0	408.1	397.7	390.7	374
T50%/°C	427.9	425.6	431.7	433.0	417.6
Tmax/°C	430.6	431.7	435.7	437.5	420
Residue (%)	44.14	53.18	47.36	52.21	57.45

---

3.8 Mechanical properties of epoxy-based dielectric materials

Flexural modulus is an important property indicating the bending stiffness of the material,⁴⁶ and the flexural strength is used to evaluate the ability to resist deformation and warpage under load. Impact strength is the capability of impact resistance which shows the degree of toughness. Fig. 11 shows the flexural modulus, flexural strength and impact strength of the epoxy-based dielectric materials. It is interesting that flexural modulus, flexural strength and impact strength of epoxy-based dielectric materials are simultaneously improved by adding functionalized micas into the epoxy matrix. The covalent bonding of the functionalized micas provided a strengthened interaction between the micas and epoxy matrix, with the improvement of mechanical properties attributed to the better interface for load transferring from the epoxy matrix to the micas.⁴⁷ It is observed that the impact strength of epoxy-based dielectric materials with mica@PDA and mica@PDA@F increased by about 72 and 62%, respectively. It is possible that the PDA polymer layer of mica@PDA and mica@PDA@F provided a better interface between mica and epoxy matrix, which results in a higher impact strength. It is also noted that the flexural modulus and flexural strength showed little difference among all the functionalized micas except for the epoxy/f-mica dielectric materials. This suggests that functionalized micas provided improved interfacial crosslinking between mica and epoxy matrix.

Fig. 11 Mechanical properties of the epoxy-based dielectric materials.

For flake composites it is difficult to achieve their theoretical high strength because it is difficult to align all flakes perfectly and have optimum overlap of one flake over another. Flakes stacked on top of each other within the polymer matrix act as a stress concentrator and reduce the strength.⁴⁸ Also, the morphological studies revealed that the flake mica tended to sediment at the bottom of the composites. Therefore, the improvement of flexural modulus and flexural strength is below 30% even though the functionalized micas provided better interface towards the epoxy matrix. The residue data of epoxy-based dielectric materials also confirmed sediment formation.

3.9 Thermal properties of epoxy-based dielectric materials

Fig. 12 shows the thermal conductivity of the epoxy-based dielectric materials as a function of temperature. It is observed that the thermal conductivity of all samples increases with temperature and then decreases. The significant plateau temperatures of epoxy/mica, epoxy/in situ grafted mica and epoxy/mica@polymer are at 125, 85 and 65 °C, respectively. The thermal conductivity results of epoxy/mica are consistent with previous reports,⁵¹ in that the thermal conductivity of an amorphous polymer increases with increasing temperature to the glass transition temperature (T_g) and decreases above T_g. While, the lower plateau temperatures of epoxy/functionalized mica are ascribed to stronger phonon Umklapp scatterings⁵² and the influence of chemicals or polymer layers of the mica surface.

Fig. 12 Temperature dependence of thermal conductivity of the epoxy-based dielectric materials.

It is obvious from Fig. 12 that the thermal conductivity of epoxy/functionalized mica dielectric materials was higher than that of epoxy/mica and that of the epoxy/mica@PDA and epoxy/p-mica dielectric materials showed most enhancement, up to 42 and 37%, respectively. The improvement is attributed to the result of better phonon transmission through the better interfaces and the decrease of Kapitza resistance,⁵³ It can be also seen that the thermal conductivity of the epoxy/mica@PDA@F and epoxy/f-mica dielectric materials were lower than that of one layer epoxy-based dielectric materials, and that of the epoxy/mica@PDA@F was the lowest. The possible explanation is that the second chemical layer of functionalized mica not only provided better interfaces to the epoxy-based dielectrics materials but also increased interfacial thermal resistance, and the improved interfaces might provide higher interfacial thermal resistance.

3.10 Dielectric properties of epoxy-based dielectric materials

The results obtained for the dielectric properties from 1⁻² Hz to 10⁶ Hz at room temperature of epoxy-based dielectric materials are shown in Fig. 13(a) and (b) from which it is observed that filler in epoxy-based dielectrics materials acts as an intermolecular plasticizer for increasing the intermolecular distance between the molecules of epoxy resins, the chain separation leading to an increase of the dielectric properties of the materials. It is observed from Fig. 13(a) that the dielectric constant of epoxy-based dielectric materials were higher than that of neat epoxy resin (3.5–4.5), and the dielectric constant decreases with increasing the applied frequency. This decrease is contributed from orientation polarization.⁵⁴ It is also worth noting that the dielectric constant of epoxy/functionalized micas showed lower values and less frequency dependence than that of epoxy/mica; the results may be due to the reduced free volume and improved interfacial bonding between mica and epoxy matrix of the epoxy/functionalized micas. Additionally, the dielectric constant of epoxy/mica@PDA@F dielectric materials were observed to have lowest frequency dependence, which is owing to the best interface and smallest free volume between mica and epoxy matrix of the epoxy/mica@PDA@F dielectric materials.

Fig. 13 Frequency dependence of dielectric properties of the epoxy-based dielectric materials: (a) dielectric constant, (b) the dielectric loss tangent.

As shown in Fig. 13(b), the dielectric loss tangents of epoxy/functionalized micas were lower than that of epoxy/mica, especially in the low frequency range from 0.01 to 1000 Hz. Importantly, epoxy/mica@PDA and epoxy/mica@PDA@F show the lowest dielectric loss tangents and frequency dependence. Mica@PDA and mica@PDA@F were coated by a stable and dense polymer shell for suppressing the leakage current of the epoxy-based dielectric materials. The surface functionalized micas improve the dispersion of mica particles in the epoxy matrix and enhance the interfacial adhesion of the epoxy-based dielectric materials, which may further restrict the movement of the molecular dipoles. The surface functionalized mica also results in fewer voids within the epoxy-based dielectric materials. It is worth noting that the stronger fluorine atom electronegativity may reduce the electron and ionic polarizability of polymer, which leads to a decrease in the dielectric constant and dielectric loss tangents.^55–56 According to the above explanation, the lowest dielectric loss tangents and frequency dependence of the epoxy/mica@PDA and epoxy/mica@PDA@F should be mainly ascribed to the improved mica interaction with epoxy matrix and the stronger electronegativity of the fluorine atom, respectively.

Dielectric breakdown strength is defined in terms of the highest voltage before the samples fail electrically, and the Weibull distribution is in general used to analyze the data of dielectric breakdown strength. Fig. 14 shows the Weibull distribution of the breakdown strength of the epoxy-based dielectric materials. For epoxy/mica, epoxy/p-mica, epoxy/f-mica, epoxy/mica@PDA and epoxy/mica@PDA@F dielectric materials, the values of the characteristic breakdown strength are calculated as 19.25, 20.3, 22.25, 24.35 and 25.8 kV mm⁻¹, respectively. It is obvious that a significant improvement in the dielectric breakdown strength is observed after the functionalization of mica. Here, the results may be attributed to the better dispersion of functionalized mica in epoxy matrix and the stronger interfacial adhesion between the functionalized micas and the epoxy matrix. The highest characteristic breakdown strength of the epoxy/mica@PDA@F dielectric materials may be due to the superior interface between mica and epoxy matrix and the better dispersion of the epoxy/mica@PDA@F dielectric materials. The above SEM (Fig. 8) confirmed that the functionalized micas possessed better dispersion and good compatibility in epoxy matrix.

Fig. 14 Weibull distribution of the breakdown strength of the epoxy-based dielectric materials.

4. Conclusion

Two series of functionalized micas were successfully prepared by self-polymerization and in situ grafting. The results showed that the epoxy/functionalized mica dielectric materials exhibited significant improvement on dielectric properties, thermal conductivity and mechanical properties. For example, the thermal conductivity of epoxy/mica@PDA is 42% higher than that of epoxy/mica, the dielectric loss of the epoxy/mica@PDA@F is one tenth of that epoxy/mica, and more importantly, the dielectric loss and dielectric constant of the epoxy/mica@PDA@F showed lower frequency dependence, and the impact strength of epoxy/mica@PDA and epoxy/mica@PDA@F is increased by about 72% and 62% compared with that epoxy/mica, respectively. In summary, the mechanical and dielectric properties of the epoxy/functionalized mica dielectric materials both depended on the improved interface and the adhesion between mica and epoxy matrix. However, the thermal conductivity of the epoxy/functionalized mica dielectric materials results from the balance of interfacial thermal resistance and the quality of the interface. All these performances suggested that epoxy/mica coated by polydopamine (PDA) and fluoro-chemical dielectric materials are promising materials with excellent dielectric properties for high thermal conductivity application. This investigation also provided a new route for preparing epoxy-based dielectric materials with higher thermal conductivity, excellent dielectric and mechanical properties for use in thermal management and next generation electronic devices.

Acknowledgements

The authors gratefully acknowledge support from the National Science Foundation of China (No. 51107081, 51277117, 51477096) and the Special Fund of the National Priority Basic Research of China under Grant 2014CB239503.

References

1. F. D. Meng, L. Liu, W. L. Tian, H. Wu, Y. Li, T. Zhang and F. H. Wang, Corros. Sci., 2015, 101, 139.
2. Y. He, Y. Fan, C. L. Chen, F. Zhong and D. Y. Qing, High Perform. Polym., 2015, 27(2), 191.
3. E. S. A. Rashid, K. Ariffin, H. M. Akil and C. C. Kooi, J. Reinf. Plast. Compos., 2008, 27(15), 1573.
4. J. J. Park, Trans. Electr. Electron. Mater., 2012, 13(4), 200.
5. M. Akatsuka, Y. Takezawa and H. Kamiya, J. Appl. Polym. Sci., 2001, 79, 2164.
6. P. Bajaj, N. K. Jha and K. Anand, J. Appl. Polym. Sci., 1995, 56, 1339.
7. M. A. R. M. Fernando, W. M. L. B. Naranpanawa, R. M. H. M. Rathnayake and G. A. Jayantha, IEEE Trans. Dielectr. Electr. Insul., 2013, 20, 2081.
8. V. K. Agarwal, H. M. Banford, B. S. Bernstein, E. L. Brancato, R. A. Fouracre, G. C. Montanari, J. L. Parpal, J. N. Seguin, D. M. Ryder and J. Tanaka, IEEE Electr. Insul. Mag., 1995, 11(3), 37.
9. E. David and L. Lamarre, IEEE Trans. Dielectr. Electr. Insul., 2007, 14(1), 212.
10. T. J. Lewis, IEEE Electr. Insul. Mag., 2001, 17, 6.
11. K. Tanaka, H. Kojima, M. Onoda and K. Suzuki, IEEE Trans. Dielectr. Electr. Insul., 2015, 22(2), 1118.
12. T. Tanaka, IEEE Trans. Dielectr. Electr. Insul., 2002, 9(5), 704.
13. S. H. Mansour, N. Mostafa and L. Abd-El-Messieh, Eur. Polym. J., 2007, 43(11), 4770.
14. M. Leijon, M. Dahlgren, L. Walfridsson, M. Li and A. Jaksts, IEEE Electr. Insul. Mag., 2001, 17(3), 10.
15. F. T. Emery, IEEE Trans. Dielectr. Electr. Insul., 2010, 17, 1396.
16. Y. S. Lee, J. K. Nelson, H. A. Scarton, D. Teng and S. Azizi-Ghannad, IEEE Trans. Dielectr. Electr. Insul., 1994, 1(6), 1186.
17. K. Kimura and Y. Kaneda, IEEE Trans. Dielectr. Electr. Insul., 1995, 2(3), 426.
18. H. N. Geetha, M. B. Srinivas and T. S. Ramu, IEEE Trans. Electr. Insul., 1990, 25(4), 747.
19. T. P. Hong, P. Gonon and O. Lesaint, IEEE Trans. Dielectr. Electr. Insul., 2009, 16, 11.
20. Y. Takezawa, M. Saeki, H. Yoshida and A. Saito, IEEE Trans. Dielectr. Electr. Insul., 2001, 8(1), 104.
21. D. H. S. Souza, K. Dahmouche, C. T. Andrade and M. L. Dias, Appl. Clay Sci., 2011, 54, 226.
22. D. H. S. Souza, C. T. Andrade and M. L. Dias, Mater. Sci. Eng., C, 2013, 33, 1795.
23. A. Boukerrou, J. Duchet, S. Fellahi and H. Sautereau, J. Appl. Polym. Sci., 2007, 105(3), 1420.
24. Y. Y. Wang and T. E. Hsieh, Chem. Mater., 2005, 17, 3331.
25. T. Agag, V. Taepaisitphongse and T. Takeichi, Polym. Compos., 2007, 28(5), 680.
26. S. S. Ray, K. Yamada, M. Okamoto, A. Ogami and K. Ueda, Chem. Mater., 2003, 15(7), 1456.
27. A. Boukerrou, J. Duchet, S. Fellahi and H. Sautereau, J. Appl. Polym. Sci., 2006, 102, 1380.
28. C. W. Chiu, W. T. Cheng, Y. P. Wang and J. J. Lin, Ind. Eng. Chem. Res., 2007, 46(22), 7384.
29. C. W. Chiu, C. C. Chu, W. T. Cheng and J. J. Lin, Eur. Polym. J., 2008, 44(3), 628.
30. L. Haeshin, S. M. Dellatore, W. M. Miller and P. B. Messersmith, Science, 2007, 318(5849), 426.
31. K. Yang, X. Y. Huang, J. L. He and P. K. Jiang, Adv. Mater. Interfaces, 2015, 2(17), 1500361.
32. Y. Yang, P. K. Qi, Y. H. Ding, M. F. Maitz, Z. L. Yang, Q. F. Tu, K. Q. Xiong, Y. Leng and N. Huang, J. Mater. Chem. B, 2015, 3(1), 72.
33. F. Wu, J. A. Li, K. Zhang, Z. K. He, P. Yang, D. Zou and N. Huang, ACS Appl. Mater. Interfaces, 2016, 8, 109.
34. L. Q. Xu, W. J. Yang, K. G. Neoh, E. T. Kang and G. D. Fu, Macromolecules, 2010, 43, 8336.
35. L. Zhang, J. J. Wu, Y. X. Wang, Y. H. Long, N. Zhao and J. Xu, J. Am. Chem. Soc., 2012, 134(24), 9879.
36. L. A. Burzio and J. H. Waite, Biochemistry, 2000, 39(36), 11147.
37. X. Liu, S. J. He, G. Song, H. N. Jia, Z. Z. Shi, S. X. Liu, L. Q. Zhang, J. Lin and S. Nazarenko, J. Membr. Sci., 2016, 504, 206.
38. S. Mallakpour and A. Zaclehnazari, New Carbon Mater., 2016, 31, 18.
39. S. Mallakpour and M. Madani, Prog. Org. Coat., 2015, 85, 131.
40. L. Wang, L. J. Hu, S. B. Gao, D. T. Zhao, L. Q. Zhang and W. C. Wang, RSC Adv., 2015, 5(12), 9314.
41. Q. P. Xin, H. Wu, Z. Y. Jiang, Y. F. Li, S. F. Wang, Q. Li, X. Q. Li, X. Lu, X. Z. Cao and J. Yang, J. Membr. Sci., 2014, 467, 23.
42. M. Zhou, Y. H. Li, C. He, T. X. Jin, K. Wang and Q. Fu, Compos. Sci. Technol., 2014, 91, 22.
43. S. L. Phua, L. P. Yang, S. Huang, G. Q. Ding, R. Zhou, J. H. Lew, S. K. Lau, X. W. Yuan and X. H. Lu, Eur. Polym. J., 2014, 57, 11.
44. Y. Song, Y. Shen, H. Y. Liu, Y. H. Lin, M. Li and C. W. Nan, J. Mater. Chem., 2012, 22(16), 8063.
45. H. L. Mo, X. Y. Huang, F. Liu, K. Yang, S. T. i and P. K. Jiang, IEEE Trans. Dielectr. Electr. Insul., 2015, 22(2), 906.
46. R. N. Ronthon, Particulate-Filled Polymer Composites, Rapra Tech. Limited, United Kingdom, 2nd edn, 2003.
47. J. W. Li, Z. X. Wu, C. J. Huang, Z. Chen, R. J. Huang and L. F. Li, Colloids Surf., A, 2013, 433(20), 173.
48. L. E. Nielsen and R. F. Landel, Mechanical properties of polymers and composites, Marcel Dekker Inc, New York, 2nd edn, revised and expanded, 1994.
49. A. Usuki, M. Kawasumi, Y. Kojima, A. Okada, T. Kurauchi and O. J. Kamiigaito, J. Mater. Res., 1993, 8, 1174.
50. A. Usuki, N. Hasegawa, H. Kadoura and T. Okamoto, Nano Lett., 2001, 1, 271.
51. P. Dashora and G. Gupta, Polymer, 1996, 37(2), 231.
52. A. A. Balandin, Nat. Mater., 2011, 10(8), 569.
53. D. G. Cahill, Rev. Sci. Instrum., 1990, 61(2), 802.
54. D. E. El-Nashar and G. Turky, Polym.-Plast. Technol. Eng., 2003, 42, 269.
55. S. J. Park, E. J. Lee and B. J. Kim, J. Colloid Interface Sci., 2008, 319(1), 365.
56. S. J. Park, H. J. Sohn, S. K. Hong and G. S. Shin, J. Colloid Interface Sci., 2009, 332, 246.

---