The role of interfacial H-bonding on the electrical properties of UV-cured resin filled with hydroxylated Al₂O₃ nanoparticles

The Al₂O₃ nanoparticles with an average diameter of about 100 nm, a purity of 99.9 %, and an average surface area of 50–80 m² g⁻¹ were obtained from Shanghai Macklin Biochemical Co., Ltd. (China). Acrylate resin (High Temp, Formlabs Inc., USA), which is sensitive to 405 nm light radiation wa used as the polymer matrix. Aqueous solutions of H₂O₂ (30 wt%), acetone and ethanol were supplied by Shanghai Reagents Co. Ltd (China). All of the chemicals were used as received. 2 cm thick quartz glass sheets , PDMS films and PET films are commercially available.

To hydroxylate the Al₂O₃ nanoparticles, the synthetic procedure reported by S Chang et al was followed [19]. Initially, 10 g c-Al₂O₃ nanoparticles and 100 ml aqueous solution of H₂O₂ (30 wt%) were added to a round bottomed flask. Then, the mixture was sonicated for 30 min to guarantee a good dispersion and refluxed at 106 °C for 4 h. Then the h-Al₂O₃ nanoparticles were recovered by centrifugation at 2000 rpm for 20 min and were washed with distilled water at least 3 times. Finally, the h-Al₂O₃ nanoparticles were dried under vacuum at 80 °C for 12 h.

The nanocomposites were fabricated by the UV-cured method, as illustrated in scheme 1. Firstly, the required amount of aforementioned Al₂O₃ nanoparticles were ultrasonically dispersed in acetone for 1 h. At the same time, UV-cured resin was diluted in acetone by magnetic stirring. The suspension of nanoparticles was then added into the UV-cured resin/acetone solution. Then, the mixture was sonicated and stirred for 6 h at 50 °C to get homogeneous nanocomposites and evaporate the residual solution. Subsequently, the mixture was degassed for 30 min to remove internal air bubbles. The mixed material was then injected into a transparent mold and cured for 2 h under 405 nm UV exposure (75 mW cm⁻²) at 60 °C. After which, films with a thickness of 500 ± 10 μm were obtained for test. The cured samples filled with c-Al₂O₃ and h-Al₂O₃ were respectively named as UV/c-Al-x and UV/h-Al-x, where x represents the percentile weight fraction of the ceramic fillers. Pure resin films were obtained by using the same procedure.

Zoom In Zoom Out Reset image size

The Chemical structures of c-Al₂O₃ and h-Al₂O₃ nanoparticles were characterized by Fourier transform infrared spectroscopy (FT-IR, IN10 + IZ10, USA) in the wavenumber range from 400 to 4000 cm⁻¹. The elemental compositions of the nanoparticles were identified by x-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, USA) with a monochromatic Al Kα x-ray (1486.6 eV photons) source. Survey scans over a binding energy range of 0–1200 eV were first taken for each sample with a constant detector pass energy range of 200 eV and then the high-resolution XPS measurements (pass energy 40 eV) were carried out to quantitatively measure the binding energy and atomic concentration.

The volume resistivity was investigated by Keithley 6517B test system. Each sample was measured eight times under a fixed 500 V DC voltage. According to GB/T 1408.1–2016 [20], the AC breakdown strength of the films with average thickness of 500 ± 10 μm was measured under a sphere-sphere electrode configuration. The dielectric properties of the samples were measured by broadband dielectric spectroscopy (Concept 80, Novocontrol Technology) over the frequency range of 10^–1–10⁶ Hz at 300 K under a 1.0 −V alternating-current voltage.The space charge distribution of the samples was measured using the pulsed electroacoustic (PEA) method under the applied DC electric field of 40 kV mm⁻¹ at room temperature for 30 min after the electric field was applied.

The isothermal surface potential decay (ISPD) method was employed to analyze the charge transport behaviors of test samples. The corona triodes consist of a needle, a mesh grid and a grounded planar electrode [21]. Both the corona system and the test samples were placed into a test chamber where the temperature and humidity was approximately kept constant as 50 °C and 55%. The specimen was firstly charged with a −15 kV DC corona voltage and a −6 kV DC bias voltage for 10 min. Then the sample was moved to the probe and the surface potential of samples was measured immediately after charging by the probe of an electrostatic voltmeter for 50 000 s.

The trap level distribution was obtained by thermally stimulated depolarization current (TSDC) measurements. Electrodes with radii of 30 mm were used. The detailed TSDC procedure is as follows: The samples were firstly polarized under 250 V DC electric field at 150 °C for 30 min. Then the samples were rapidly cooled to −20 °C and stabilized at −20 °C for 3 min. After that, the poling voltage was removed and the samples were short-circuited for 3 min. Finally, the depolarization current was recorded from −20 °C to 180 °C with a heating rate of 3 °C min⁻¹.

FTIR spectroscopy was used to demonstrate the effect of the H₂O₂ treatment on the surface of the c-Al₂O₃ particles. Figure 1(a) showed the FTIR spectra of c-Al₂O₃ and h-Al₂O₃ nanoparticles. The broad absorbance band peaked at 3443 cm⁻¹ represents the stretching mode of surface -OH groups on the particles [22]. It is clearly found that the c-Al₂O₃ and h-Al₂O₃ nanoparticles demonstrated exactly similar spectra in shape. However, as illustrated in the inset of figure 1(a), after being treated with H₂O₂, a stronger hydroxyl absorbance peak appeared in the h-Al₂O₃ nanoparticles, which signified denser surface hydroxyl groups. Overall, the results implied that the hydroxyl groups are successfully introduced onto the surface of c-Al₂O₃ nanoparticles by H₂O₂ treatment.

Zoom In Zoom Out Reset image size

Additionally, the effect of the H₂O₂ treatment on the surface of the Al₂O₃ nanoparticles was further verified by XPS. As described in figure 1(b), the XPS general spectra of c-Al₂O₃ and h-Al₂O₃ nanoparticles were obtained. It can be seen that there is no longer a significant difference between both nanoparticles. Figures 1(c) and 1(d) revealed the high-resolution XPS spectra of the c-Al₂O₃ and h-Al₂O₃ nanoparticles, respectively. It is clearly that the h-Al₂O₃ nanoparticles have a stronger -OH peak than c-Al₂O₃ nanoparticles. These results indicate that the surface treatment of the c-Al₂O₃ nanoparticles with H₂O₂ could endow the c-Al₂O₃ nanoparticles with -OH groups.

The dispersion of fillers in composites can be directly observed by using the SEM. The SEM micrographs to describe the fractured surfaces of the UV/c-Al-10 and UV/h-Al-10 composites are shown in figure 2. Large phase separations and agglomerations of Al₂O₃ particles in UV-cured resin matrix can be observed as demonstrated in figure 2(a) and the inset of figure 2(a). As the hydrogen bonds formed by the hydroxyl groups of h-Al₂O₃ and the carboxyl groups of UV-cured resin molecular chain, there is strong interfacial strength between filler and matrix, leading to a good dispersion of h-Al₂O₃ particles in the resin matrix (as shown in figure 2(b) and the inset of figure 2(b)).

Zoom In Zoom Out Reset image size

The dielectric properties of the UV-cured resin nanocomposites were measured to find out the effect of hydrogen bonds formed by the carboxyl groups of UV-cured resin and the hydroxyl groups of h-Al₂O₃ nanoparticles. Figures 3(a) and (b) display the frequency dependence of the dielectric constant and dielectric loss tangent in pure UV-cured resin, UV/c-Al-x, and UV/h-Al-x composites. We can see that the dielectric constant for UV/c-Al-x and UV/h-Al-x composites increased with the increasing content of fillers relative to that of pure UV-cured resin. The main reasons are the larger dielectric permittivity of inorganic Al₂O₃ fillers and the greater interfacial polarization gained by the large amount of filler particles [23]. Beyond that, the dielectric constant of UV/h-Al-x is higher than that of UV/c-Al-x composites and the dielectric constant and loss tangent increased significantly with elevating weight of the Al₂O₃ nanoparticles especially when the content of Al₂O₃ reached 10 wt%. It may be explained by two reasons as followed. The first one is that the orientation polarization could be promoted when polar groups (-OH groups) were introduced into the composites. Second, the presence of hydrogen bonding enhanced the size of the interface region between the nanoparticles and the resin, which contributed to the improvement of the interface polarization at relatively low frequency [24].

Zoom In Zoom Out Reset image size

The volume resistivity of UV/c-Al-x and UV/h-Al-x composites with different contents of fillers are shown in figure 3(c) and listed in table 1. It is indicated that the volume resistivity decreased with the increasing content of c-Al₂O₃ or h-Al₂O₃ nanoparticles, meanwhile the volume resistivity of UV/h-Al-x is significantly higher than that of UV/c-Al-x composites. In addition, the results presented show that volume resistivity of UV/h-Al-0.5 composites increased 4.68 times (from 2.18 × 10¹⁶ Ω cm⁻¹ to 10.2 × 10¹⁶ Ω cm⁻¹), while that of UV/c-Al-0.5 composites increased 2.51 times (from 2.18 × 10¹⁶ Ω cm⁻¹ to 5.47 × 10¹⁶ Ω cm⁻¹) compared to pure UV-cured resin.

Table 1. Dielectric properties, volume resistivity and Weibull parameters of AC breakdown strength.

	r,	ρ (1016	Weibull parameters of AC breakdown strength
Specimen	103 Hz	Ω cm−1)	α (kV mm−1)	Shape, β
Pure UV	3.54	2.18	58.89	43.13
UV/c-Al-0.5	3.69	5.47	64.82	36.91
UV/c-Al-1	3.89	2.24	58.73	27.00
UV/c-Al-10	4.06	0.21	53.84	21.77
UV/h-Al-0.5	3.71	10.20	68.00	45.43
UV/h-Al-1	3.94	3.54	61.56	36.09
UV/h-Al-10	4.26	2.04	57.65	35.45

Breakdown strength (E_b) is a crucial parameter for practical applications as it determines the operational electric field. The Weibull statistical distribution is employed to analyze the AC breakdown strength of the nanocomposites according to the following formula [25, 26]:

$$\begin{equation}P(E) = 1 - \exp \left[ { - {{({E_{\text{b}}}/\alpha )}^\beta }} \right]\end{equation} \tag{ 1 }$$

where P(E) is the failure probability, E_b is the measured breakdown strength, α is the scale parameter which denotes a 63.2% probability of failure and β is the shape parameter which reflects the dispersion of data.

Figure 3(d) illustrates the AC breakdown strength of the nanocomposites with different contents of fillers and the parameters α and β are summarized in table 1. It is clear that the breakdown strength decreased with the increasing content of c-Al₂O₃ or h-Al₂O₃ particles. Moreover, the breakdown strength of UV/h-Al-x is remarkably higher than that of UV/c-Al-x composites. Specifically, the AC breakdown strength of UV/h-Al-0.5 composites increased by 15.5% (from 58.89 kV mm⁻¹ to 68 kV mm⁻¹) and 5% (from 64.82 kV mm⁻¹ to 68 kV mm⁻¹) compared with pure UV-cured resin and UV/c-Al-0.5 composites, respectively.

Figure 4(a) demonstrates the space charge behaviors and corresponding electrical field profiles of pure UV-cured resin. It can be obviously seen that large amounts of homogenous charges were injected from both electrodes, especially the cathode. The injected homogenous charges accumulated at the vicinity of the cathode electrode and moved towards the bulk of the sample with increasing time. The largest electric field distortion in pure UV-cured resin is about 9 kV mm⁻¹, which is nearly 22.5% of the average applied electric field of 40 kV mm⁻¹. Such serious electric field distortion is very unfavorable for insulation applications [27, 28].

Zoom In Zoom Out Reset image size

As shown in figure 4(b), there is only a small amount of charge injected and accumulated in the UV/c-Al-1 composites. Up to 1800 s, the maximum electric field in the UV/c-Al-1 composites is less than that in pure UV-cured resin. Due to the introduction of the h-Al₂O₃ nanoparticles, as shown in figure 4(c), the nanocomposites possessed the best suppression performance of space charge, thus there is almost no space charge accumulation. The maximum electric field distortion detected in UV/c-Al-1 is about 3 kV mm⁻¹, which is only one third of that in pure UV-cured resin. It suggests that the injection and migration of charge is significantly suppressed, and the distortion of the electric field is effectively improved.

Though the PEA is able to provide a spatial distribution of space charge across the specimen, its spatial resolution is limited (typically 5–10 μm). Therefore, this technique cannot be used to distinguish space charges located in the subsurface region of the specimen from capacitive charges on the electrodes (shown in figures 4(b) and 4(c)), which confuses the understanding of the effect of deep traps on the mechanism of space charge suppression [29]. Fortunately, ISPD is a common method to characterize the traps of sample in the subsurface region (1–2 μm). Based on the isothermal decay current theory, the relationship between the trap energy E_t and the trap density N_t is shown as following [30, 31].

$$\begin{align}\begin{cases}\begin{array}{*{20}{l}} \!\!\! {E_{\text{t}}} = kT\ln (\gamma t) \\ \!\! {N_t}({E_t}) = \displaystyle {\frac{{\varepsilon _0}{\varepsilon _r}t}{kT{f_0}({E_t})\delta Lq}\frac{d{V_s}}{dt}} \end{array}\end{cases}\end{align} \tag{ 2 }$$

Where k is the Boltzmann's constant, T is the absolute temperature, γ is the electron vibration frequency, set as 10¹² s⁻¹, N_t(E_t) is the trap density corresponding to the energy level of E_t, ₀ is the vacuum permittivity and _r is the relative permittivity of dielectric, δ is the thickness of top charge layer, usually set as 2 μm, L is the sample thickness, and f₀(E_t) is the initial occupation probability of traps, set as 1/2 [32].

The ISPD experimental surface potential decay curve and corresponding calculated trap level distributions of pure UV-cured resin, UV/c-Al-1, and UV/h-Al-1 composites are described in figures 5(a) and 5(b), respectively. As can be seen from figure 5(a), the surface potential decay rate of UV/c-Al-1 composites is bigger than that of pure UV-cured resin, while obviously less than that of UV/h-Al-1 composites. Trap level E_t and trap density N_t(E_t) were calculated by using the ISPD theoretical equation (2), as shown in figure 5(b). It indicates that large amounts of deep traps are introduced by doping c-Al₂O₃ and h-Al₂O₃ particles in pure UV-cured resin. In addition, the trap density and depth of UV/h-Al-1 composites are higher than that of UV/c-Al-1 composites.

Zoom In Zoom Out Reset image size

To further characterize the trap level distributions of pure UV-cured resin, UV/c-Al-1, and UV/h-Al-1 composites, the TSDC discharge current with temperature change was transformed into trap level density and trap level according to the calculation program in the previous study [33], as shown in figure 6. Meanwhile, the TSDC experimental current spectrum was inset into figure 6. The calculated trap level parameters are summarized in table 2. It is found that doping c-Al₂O₃ or h-Al₂O₃ particles into the UV-cured resin did not induce a new trap level peak. All the samples have two peaks, including deep trap peak α located at high temperature and shallow trap peak β located at a relatively lower temperature. By comparing the trap level distributions of pure UV-cured resin, UV/c-Al-1, and UV/h-Al-1 composites, we found that the trap level and trap density of peak α in UV/c-Al-1 composite are bigger than pure UV-cured resin, while the trap level of peak β is lower than pure UV-cured resin. However, when doped with 1 wt% h-Al₂O₃ particles, the trap level and density of peak α and peak β are significantly higher than pure UV-cured resin and UV/c-Al-1 composites. It indicates that large amounts of deep traps are introduced by doping h-Al₂O₃ in pure UV-cured resin, which is in good agreement with the results of ISPD.

Zoom In Zoom Out Reset image size

Space charge transport includes charge migration, trapping, de-trapping and recombination processes, which is closely related to the electrical properties (such as breakdown, resistivity, dielectric constant, etc) [34]. Additionally, charge transport is affected by trap characteristics [35, 36]. The interface region (interaction zone) between the nanoparticles and matrix in composite materials is a special region, different from that of nanoparticles and matrix materials. There are a large number of local states in this region, which affect the trap characteristics of composite materials. As described in [37], the interaction zone around nanoparticles consists of two regions, which are named the bonded region and transition region. Among the two regions, the bonded region will produce deeper traps and the transition region will produce a large amount of shallower traps [38]. Compared with the matrix material, the interaction zone is considered as a conductive zone, and the thickness of the interaction zone is affected by the strength of interaction between nanoparticles and polymeric matrix [39]. The stronger the interaction strength, the thicker the interaction zone becomes [40].

Figures 7(a) and 7(b) illustrated the formation mechanism of the intermolecular hydrogen-bonding system between Al₂O₃ nanoparticles and the UV-cured resin matrix. As shown in figure 7(b), the hydrogen bonding will be formed between the hydroxyl groups of Al₂O₃ nanoparticle and the carboxyl groups of UV-cured resin molecular chain. As a result, the dispersion of UV/h-Al-x composites is improved and the interfacial strength between h-Al₂O₃ and UV-cured resin matrix is enhanced (shown in figure 2). Based on the knowledge and understanding of the interaction zone, the interaction zone models of UV/c-Al-x and UV/h-Al-x composites are presented as figures 7(c) and 7(d), respectively. Compared with the UV/c-Al-x composites, a larger volume of the interaction zone existed in UV/h-Al-x composites because the interfacial bonding strength is improved by hydrogen bonding interaction between fillers and UV-cured resin matrix. Consequently, the number of deeper traps increased, which can be verified by ISPD and TSDC measurement results.

Zoom In Zoom Out Reset image size

Under high electric field stress, the carriers injected from electrodes will be captured by traps at the vicinity between the electrode and sample. A reverse electric field will be formed owing to the accumulation of homogenous charges [41, 42], resulting in the suppression of subsequent charge injection and the mitigation of electric field distortion. The deeper the trap level, the stronger the suppressing effect becomes [43]. Hence, UV/h-Al-1 composites have better suppression of space charge accumulation and the mitigation of electric field distortion than pure UV-cured resin and UV/c-Al-1 composites (shown in figure 4). Moreover, carriers need to overcome a bigger potential barrier after being trapped in a deeper trap, leading to decreased carrier mobility and energy. Thus, UV/h-Al-x composites exhibit remarkably enhanced volume resistivity, improved breakdown strength compared with UV/c-Al-x composites at the same filler content (shown in figures 3(c) and 3(d)). The space charge layer effectively hinders the movement of dipoles, causing a reduction in the dielectric constant at a certain degree. With increasing filler loading, the distance between neighboring particles becomes shorter. In this way, carriers can obtain enough energy from the electric field to hop the barrier between two neighboring particles and transport instead of being restrained in the transition region. The shorter the distance between neighboring particles, the greater the probability to hop the barrier becomes. Thus, the volume resistivity decreased with the increasing content of c-Al₂O₃ or h-Al₂O₃ particles. At a large doping concentration (10 wt%), the formation of conductive channels between the interaction areas due to the overlap of the interaction areas resulted in a lower volume resistivity and breakdown strength of the material [37, 39].