Here, we demonstrate a simple and low-cost solution-casting process to improve the compatibility between the filler and the matrix of polymer-based nanocomposites using surface modified BaTiO3 fillers, which can effectively enhance the energy density of the composites.
In this work, an easy, low-cost, and widely applicable method was developed to improve the compatibility between the ceramic fillers and the polymer matrix by adding 3-aminopropyltriethoxysilane (KH550) as a coupling agent during the fabrication process of BaTiO3-P(VDF-CTFE) nanocomposites through solution casting. Results show that the use of KH550 can modify the surface of ceramic nanofillers; therefore, good wettability on the ceramic-polymer interface was achieved, and the enhanced energy storage performances were obtained by a suitable amount of the coupling agent. This method can be used to prepare flexible composites, which is highly desirable for the production of high-performance film capacitors. If an excessive amount of coupling agent is used in the process, the non-attached coupling agent can participate in complex reactions, which leads to a decrease in dielectric constant and an increase in dielectric loss.
The dielectrics applied in electrical energy storage devices are mainly characterized using two important parameters: the dielectric constant (εr) and the breakdown strength (Eb)1,2,3. In general, organic materials such as polypropylene (PP) exhibit a high Eb (~102 MV/m) and a low εr (mostly <5)4,5,6 while inorganic materials, especially ferroelectrics such as BaTiO3, exhibit a high εr (103-104) and a low Eb (~100 MV/m)6,7,8. In some applications, flexibility and the ability to withstand high mechanical impacts are also important for fabricating dielectric capacitors4. Therefore, it is important to develop methods for preparing polymer-based dielectric composites, especially for the development of low-cost methods to create high-performance 0-3 nanocomposites with high εr and Eb9,10,11,12,13,14,15,16,17,18. For this purpose, preparation methods based on ferroelectric polymer matrices such as the polar polymer PVDF and its correlated copolymers are widely accepted due to their higher εr (~10)4,19,20. In these nanocomposites, particles with high er, especially ferroelectric ceramics, have been widely used as fillers6,20,21,22,23,24,25.
When developing methods for manufacturing ceramic-polymer composites, there is a general concern that dielectric properties can be significantly influenced by the distribution of fillers26. The homogeneity of dielectric composites is not only determined by the preparation methods, but also by the wettability between the matrix and fillers27. It has been proven by many studies that the non-uniformity of ceramic-polymer composites can be eliminated by physical processes such as spin-coating28,29 and hot-pressing19,26. However, neither of these two processes change the surface connection between fillers and matrices; therefore, the composites prepared by these methods are still limited in improving εr and Eb19,27. Additionally, from a manufacturing point of view, inconvenient processes are undesirable for many applications because they can lead to much more complex fabrication processes28,29. In this regard, a simple and effective method is needed.
Currently, the most effective method to improve the compatibility of ceramic-polymer nanocomposites is based on the treatment of ceramic nanoparticles, which modifies the surface chemistry between fillers and matrices30,31. Recent studies have shown that coupling agents can be easily coated on ceramic nanoparticles and effectively modify the wettability between fillers and matrices without affecting the casting process32,33,34,35,36. For surface modification, it is widely accepted that for each composite system, there is a suitable amount of coupling agent, which corresponds to a maximal increase in energy storage density37; excess coupling agent in composites may result in a decline in the performance of products36,37,38. For dielectric composites using nano-sized ceramic fillers, it is speculated that the effectiveness of coupling agent mainly depends on the surface area of fillers. However, the critical amount to be used in each nano-sized system is yet to be determined. In short, further research is required to use coupling agents to develop simple processes for manufacturing ceramic-polymer nanocomposites.
In this work, BaTiO3 (BT), the most widely studied ferroelectric material with high dielectric constant, was used as fillers, and the P(VDF-CTFE) 91/9 mol% copolymer (VC91) was used as the polymer matrix for the preparation of ceramic-polymer composites. To modify the surface of the BT nanofillers, the commercially available 3-aminopropyltriethoxysilane (KH550) was purchased and used as a coupling agent. The critical amount of the nanocomposite system was determined through a series of experiment. An easy, low-cost, and widely applicable method is demonstrated to improve the energy density of nano-sized composite systems.
1. Surface modification of BT fillers
2. Preparation of BT-VC91 nanocomposites
3. Characterization and measurement
The free-standing nanocomposite films with different contents of fillers were successfully fabricated as described in the protocol, and were labeled as xBT-VC91, where x is the volume percentage of BT in the composites. The effect of KH550 (coupling agent) on the morphology and microstructure of these BT-VC91 films was studied by SEM and shown in Figure 1. The SEM images of 30BT-VC91 nanocomposites with 1 and 5 wt% coupling agent are shown in Figure 1a and Figure 1b. The filler distribution of BT-VC91 nanocomposites with 1 wt% KH550 is much denser and more uniform than that of BT-VC91 nanocomposites with 5 wt% KH550, suggesting that ceramic nanoparticles treated with a suitable amount of coupling agent could be uniformly distributed in the nanocomposites during casting, while the excessive amount of coupling agent may cause interactions between ceramic nanoparticles, and leading to the aggregation of fillers. The image of cross section (i.e., the ceramic-polymer interface) of 30BT-VC91 nanocomposites using as-received (unmodified) BT fillers is shown in Figure 1c, while the cross section of 30BT-VC91 nanocomposites containing 1 wt% of KH550 is shown in Figure 1d. For the nanocomposites using uncoated BT, although most of the nanoparticles are tightly encapsulated in polymer, there are still some separation between the fillers and matrix, which means there is no linkage between the matrix and fillers. For the nanocomposites using KH550-coated BT, there is no separation between BT nanoparticles and VC91 matrix, which indicates that the coupling agent could acts as a bridge between filler and matrix.
The dielectric properties of nanocomposites with different amounts of coupling agent were then tested and shown in Figure 2. The dielectric content vs. amount of coupling agent at 1 kHz and 100 kHz was plotted in Figure 2a,b. For the nanocomposites with a low filler content (i.e., 5, 10 and 15 vol%), the εr of the composites was basically unchanged when a small amount of coupling agent is used, and slightly decreases with the increasing coupling agent amount. For the nanocomposites with a high filler content, in particular the nanocomposites with a filler content of 30 vol%, the εr of the composites increases obviously with a small amount of coupling agent, and sharply decreases with the further increasing coupling agent amount. When a suitable amount of KH550 was coated on the surface of BT filler, the maximum εr could be achieved. For example, a εr of 51 was achieved from 30BT-VC91 with 2 wt% of KH550 (Figure 2a), which is much larger than that of 30BT-VC91 without KH550 (about 40). In this composite system, the increase of εr for the nanocomposites with a small amount of coupling agent is due to the increase of wettability on the ceramic-polymer interface, and the possible percolation from the additives6,10,33; the decrease of εr for BT-VC91 using BT nanoparticles coated with a large amount of KH550 is due to the formation of VC91-KH550 polymer blends with a low dielectric constant. The difference in dielectric properties between low filling and high filling nanocomposites can be attributed to the actual amount of KH550 used in the sample preparation. The dielectric loss vs. amount of coupling agent at 1 kHz and 100 kHz was plotted in Figure 2c,d. BT-VC91 with KH550 has a higher dielectric loss than that of BT-VC91 without KH550.
The breakdown strengths of BT-VC91 nanocomposites were also recorded and shown in Figure 3. To determine the critical amount of the coupling agent, the breakdown strength vs. amount of coupling agent and breakdown strength vs. content of filler were shown in Figure 3a and Figure 3b, respectively. As expected, the Eb of BT-VC91 decreased with increasing filler content (Figure 3b) due to the formation of the ceramic-polymer interface. A maximal Eb of 30BT-VC91 was observed for composites produced using fillers treated with 2 wt% KH550 (Figure 3b). If a KH550 amount exceeding 2 wt% was used, the Eb of BT-VC91 was further decreased (Figure 3a). By adding 2 wt% KH550, the Eb of 30BT-VC91 could be increased to 200 MV/m.
The charge-discharge efficiency and discharge energy density of nanocomposites with different amount of coupling agent were calculated from their P-E loops. As an example of the enhanced energy density due to the usage of coupling agent, the energy storage properties of 15BT-VC91 with different amount of KH550 are shown in Figure 4. The maximal energy densities of BT-VC91 nanocomposites with a small amount of coupling agent (1 – 2 wt%) apparently increased compared with those of nanocomposites without coupling agent (Figure 4b), which could be mainly attributed to the enhanced breakdown strength and a relatively high charge-discharge efficiency (η). Due to the higher loss under high electric filed, the η of BT-VC91 nanocomposites decreased at relatively high electric fileds (Figure 4a). Adding 1 – 2 wt% of KH550 increased η of nanocomposites under a fixed electric field (Figure 4a), which was attributed to the introduced bridge-linking effect. In summary, for nanocomposites prepared in this work using BT nanoparticles with ~200 nm in diameter, the critical amount of KH550 is smaller than 2 wt%.
In terms of the frequency dependence of dielectric properties, εr and tanδ of nanocomposites versus testing frequency were also plotted. As an example, the dielectric properties of BT-VC91 with 1 wt% coupling agent are shown in Figure 5, which indicated that the frequency dependences of dielectric properties (εr and tanδ) of all BT-VC91 nanocomposites were mainly determined by their polymer matrix. The εr of nanocomposites gradually decreased with increasing frequency (Figure 5a). The tanδ gradually decreased with frequency at low frequencies but gradually increased at high frequencies (Figure 5b).
Figure 1: SEM images of cross sections. Filler distribution of (a) 30BT-VC91 with 1 wt% of KH550 and (b) 30BT-VC91 with 5 wt% of KH550. Ceramic-polymer interface of (c) 30BT-VC91 without KH550 and (d) 30BT-VC91 with 1 wt% of KH550. This figure has been modified from Tong et al.4. Please click here to view a larger version of this figure.
Figure 2: Dielectric properties of composites with different amount of coupling agent (a) εr at 1 kHz and (b) εr at 100 kHz; (c) tanδ at 1 kHz and (d) tanδ at 100 kHz. This figure has been modified from Tong et al.4. Please click here to view a larger version of this figure.
Figure 3: Breakdown strengths of nanocomposites with different amount of coupling agent (a) Eb of BT-VC91 as a function of KH550 amount (b) Eb of BT-VC91 as a function of filler content. This figure has been modified from Tong et al.4. Please click here to view a larger version of this figure.
Figure 4: Energy storage performances of nanocomposites with different amount of coupling agent (a) charge-discharge efficiency and (d) discharge energy density of 15BT-VC91 as a function of KH550 amount. This figure has been modified from Tong et al.4. Please click here to view a larger version of this figure.
Figure 5: Frequency dependence of the dielectric properties of nanocomposites (a) εr and (b) tanδ of BT-VC91 with 1 wt% of KH550. This figure has been modified from Tong et al.4. Please click here to view a larger version of this figure.
As discussed above, the method developed by this work could successfully improve the energy-storage performance of ceramic-polymer nanocomposites. To optimize the effect of such method, it is critical to control the amount of coupling agent used in ceramic-surface modification. For ceramic nanoparticles with a diameter of ~200 nm, it was experimentally determined that 2 wt% of KH550 could lead to a maximal energy density. For other composite systems, this conclusion may be used approximately when the fillers with the diameter is close to ~200 nm are adopted. If fillers with a diameter much larger than 200 nm are used, the critical amount should be determined again through a similar series of experiments.
Compared with other works that attempted to improve the uniformity and performance of dielectric nanocomposites, the method developed in this work is much simpler and has a lower cost. In addition, application of the coupling agent can be combined with other processes such as spin-coating and hot-pressing. The surface modification of ceramic nano-fillers will be widely applied in the manufacture of various of advanced dielectrics in future.
It must be pointed out that the application of coupling agent does not really change the properties of nanocomposites. Therefore, the effectiveness of a coupling agent in a composite strongly depends on the selection of fillers and matrices, and the method proposed here increases in energy storage performance only to a limited extent. In order to develop dielectrics with a dramatically increased energy density, the novel composite systems still need to be created.
The authors have nothing to disclose.
This work was supported by the Taiyuan University of Science and Technology Scientific Research Initial Funding (20182028), the doctoral starting foundation of Shanxi Province (20192006), the Natural Science Foundation of Shanxi Province (201703D111003), the Science and Technology Major Project of Shanxi Province (MC2016-01), and Project U610256 supported by National Natural Science Foundation of China.
3-Aminopropyltriethoxysilane (KH550) | Sigma-Aldrich | 440140 | Liquid, Assay: 99% |
95 wt.% ethanol-water | Sigma-Aldrich | 459836 | Liquid, Assay: 99.5% |
BaTiO3 nanoparticles | US Research Nanomaterials | US3830 | In a diameter of about 200 nm |
Ferroelectric tester | Radiant | Precision-LC100 | |
Glass substrates | Citoglas | 16397 | 75 x 25 mm |
Gold coater | Pelco | SC-6 | |
High voltage supplier | Trek | 610D | 10 kV |
Impedance analyzer | Keysight | 4294A | |
N, N dimethylformamide | Fisher Scientific | GEN002007 | Liquid |
P(VDF-CTFE) 91/9 mol.% copolymer | |||
Scanning Electron Microscopy (SEM) | JEOL | JSM-7000F | |
Vacuum oven | Heefei Kejing Materials Technology Co, Ltd | DZF-6020 |