Introduction

The scientific and technological importance of nanostructured materials in advanced technology systems is very high and globally recognized because of their potential applications, based on their thermo-mechanical behaviour, their electrical properties, etc. [1,2,3,4]. Advanced materials are urged to respond to the ever-changing social needs, moving towards multifunctional performance. Nowadays, behaviour of engineering materials should not only meet the criteria of specific nominal values of various properties (i.e. Young’s modulus, strength, conductivity etc.), but also be able to execute functions or operations in response to an external stimulus or control signal. Materials’ multifunctionality refers to the combination of various desirable properties in a system, aiming to develop a single material/system which will be able to exhibit all necessary responses under various loading conditions at service. Mechanical sustainability, suitable thermal response, tunable electric conductivity, variable electric polarization and dielectric permittivity, magnetic properties, thermally induced phase changes could be parts of the overall multifunctional behaviour [1,2,3,4,5,6,7,8,9,10]. Polymer matrix nanocomposites exhibit multifunctionality because are able to combine light weighting, adequate static and dynamic mechanical performance, electrical, magnetic and optical properties, and ease processing [11]. Moreover, polymer nanocomposites achieve their performance at a much lower level of filler loading compared to the corresponding microcomposites [12]. Employed reinforcing phase could be in the form of nanoparticles, nanofibers or nanoplatelets, being organic or inorganic in nature, and electrically conductive or insulating. 2D materials constitute an extremely interesting class of materials, since are ultra-thin materials approaching the atomic length-scale, with structure almost free of defects that lend to the development of innovative properties [13]. 2D materials can be employed into 3D heterostructures providing, to the fabricated systems, additional functionalities for future applications [14], constituting thus a very promising reinforcing phase. Graphene is the leading figure of this class due to its advanced mechanical and electrical properties. Previous studies on graphene/nanocomposites provide methods of processing, properties characterization and potential applications [15,16,17,18,19].

Boron nitride (BN) is a two-dimensional (2D) sp2-hybridized material of great interest because of its structural similarities to graphite and its allotropic form of graphene, and its lattice difference is limited to 1.7% [20, 21] with high chemical and heat resistance. The majority of the existing studies on BN nanocomposites refer to thermal conductivity or mechanical properties [21,22,23,24], and only scarce attention has been given to their electrical behaviour [25]. Epoxy resins (ER) are suitable matrices for the development of high tech nanocomposites because of their thermo-mechanical stability, low shrinkage, resistance to corrosion, strong adhesion with nanoparticles and low cost.

The purpose of the present study is to fabricate and investigate the electrical and dynamic mechanical properties of novel nanocomposite materials based on epoxy resin matrix reinforced with boron nitride nanopowder, targeting to elucidate their electrical performance, occurring relaxation processes and energy storage ability.

Experimental

Materials and fabrication

Commercially available materials were employed for the fabrication of nanocomposites. In particular, a low viscosity epoxy resin (ER) (Epoxol 2004 A, Neotex S.A., Athens, Greece), and a slow-acting hardener (Epoxol 2004 B, Neotex S.A., Athens, Greece) were used for the preparation of the matrix, while boron nitride (BN) nanoparticles, supplied by Thomas Swan & Co. Ltd. (UK), were used as the reinforcing phase. The content of the added BN nanoinclusions in the samples was: 0, 1, 3, 5, and 10 phr (parts per hundred resin per mass). The %w/w content for each system is listed in Table 1. The preparation procedure is constituted by the following four steps:

  1. 1.

    Slow rate stirring of epoxy prepolymer and BN nanoparticles in ultra sonicator at T = 50 °C.

  2. 2.

    After the homogenation of the mixture the hardener was added at ambient. The mixing rate of epoxy prepolymer and hardener was 2:1 w/w.

  3. 3.

    Then the filler/matrix mixture was stirred and subsequently poured into silicon moulds.

  4. 4.

    The curing took place at ambient temperature for 7 days, followed by post-curing at 100 °C for 4 h.

It is worth mentioning that during the fabrication procedure special care was given in order to avoid the formation of aggregates, voids and the inclusion of air, since all three parameters are detrimental to the performance of the nanocomposites, leading to possible early failure of the systems and to degradation of their properties.

Table 1 Filler content in phr and weight per weight percentage, fitting parameters of Eqs. (2) and (4) for conductivity data at 0.1 Hz and the α-relaxation process
Full size table

Methods of characterization

Scanning electron microscopy (SEM) (Carl Zeiss EVO MA 10) was used in order to ascertain the absence of agglomerates and the successful dispersion of the nanoparticles into the epoxy resin. Structural studies of the BN nanocomposites were performed through X-ray diffraction (XRD) (Bruker AXS D8 Advance with Bragg-Bretano geometry) method. Additional structural characterization was performed via Fourier transform infrared spectroscopy (FTIR, Shimadzu IR Tracer-100 spectrometer) in the range of 500–4000 cm−1.

The electrical response of the specimens was examined via broadband dielectric spectroscopy (BDS). The Alpha-N ac-bridge and the BDS 1200 dielectric cell were used for studying the electrical response of all systems. Data recording was conducted automatically via windeta software. Temperature control was achieved via Novotherm system with an accuracy of ± 0.1 °C. All devices and software used for electrical studies were purchased by Novocontrol Technologies. The samples were examined, under isothermal conditions, in the temperature range from 30 to 160 °C with a temperature step of 5 °C and in the frequency range from 10−1 to 107 Hz.

Structural integrity and mechanical properties of the nanocomposites were investigated by means of dynamic mechanical thermal analysis (DMTA, Q800, TA Instruments) from ambient to 160 °C. Tests were conducted in the three point bending configuration, while frequency was kept constant at 1 Hz.

Results and discussion

Morphological-structural characterization (SEM-XRD-FTIR)

Figure 1 provides evident for the fine dispersion of BN within the epoxy matrix, for the lowest (1 phr) and the highest (10 phr) filler content. At the specimen with the highest filler content, limited clusters can be detected in tandem with the nanodispersion. From SEM images, it can be concluded that the dispersion of nanoinclusions in the epoxy resin is, at least, satisfactory and large agglomerates have been avoided. Finally, small size nanoparticle’s aggregates (approximately 300 nm) are detected at the system with the highest filler content.

Fig. 1
figure 1

SEM images for the nanocomposites with: a 1 and b 10 phr BN content

Full size image

In Fig. 2, XRD spectra of BN nanopowder and all tested systems are presented. Figure 2a depicts the XRD pattern of BN. Its crystalline structure is evident by the formed peaks at 26.8°, 41.7°, 43.7°, 50.1° and 54.9° which correspond to the diffraction of (002), (100), (101), (102) and (004) planes, respectively. The successful implementation of the BN nanoparticles is apparent via the formation of peaks related to BN, the intensity of which increases with filler content (Fig. 2b). The amorphous matrix gives a wide halo with no diffraction peaks. Additional structural characterization was performed by means of FTIR (Fig. 3). BN nanopowder exhibits two characteristic vibrational modes, the in-plane axial B-N-B vibration and the out-of-plane B-N vibration, located at approximately 1390 cm−1 and 800 cm−1, respectively. The same modes are also observed in the spectrum of the 10 phr BN/epoxy specimen, verifying the successful integration of BN nanoparticles in the matrix.

Fig. 2
figure 2

X-ray diffraction spectra of: a BN nanopowder and b all examined systems

Full size image
Fig. 3
figure 3

FTIR spectra of BN nanopowder, unfilled epoxy and 10 phr BN/epoxy nanocomposite

Full size image

Electrical characterization (BDS)

Figure 4a depicts the variation of the real part of dielectric permittivity (ε′) as a function of frequency and temperature for the nanocomposite with 10 phr content of BN. The values of ε′ decrease with frequency since permanent and/or induced dipoles fail progressively to be oriented with the field. At low frequencies, the real part of dielectric permittivity attains the highest values, since the slow alternation of the applied field provides the necessary time for dipoles orientation. Moreover, ε′ increases further with the temperature because of the dipoles’ enhanced mobility. Figure 4b depicts comparative isochronal spectra of ε′ for all studied systems as a function of temperature, at 1 kHz. Reinforced systems show higher values of ε′ than those of the neat ER over the entire temperature range, since BN exhibits higher values than the ER and because of the heterogeneity of the composite systems. The influence of filler content upon the values of permittivity is more evident at high temperatures because of the enhanced effect of interfacial polarization. In the 3D graph of tanδ versus frequency and temperature (Fig. 5a) of the 10 phr BN/epoxy system, three relaxations are recorded. At high frequencies, the fastest and the weakest observed process is attributed to motions of lateral polar groups of the main polymeric chain, named β-relaxation. At the intermediate frequencies, the recorded relaxation process is characterized as relatively slow and is attributed to the transition from the glassy to the elastomeric phase of the amorphous polymeric matrix and is called α-relaxation. This transition is characterized by the glass transition temperature (Tg) which is related to the location of the loss peak. The slowest process is observed at the low-frequency and high-temperature range. This phenomenon is attributed to the interfacial polarization (IP) and is also known as Maxwell–Wagner–Sillars (MWS) effect. IP is observed in heterogeneous materials because of the accumulation of charges at the interface of the system, where large dipoles are formed. These dipoles attempt to be oriented with the field, however exhibit inertia, due to their large size, leading to increased relaxation times and thus induced dipoles need sufficient time and thermal agitation [3, 26]. Figure 5b presents the variation of tanδ with frequency at T = 160 °C for all studied systems. Loss spectra reveal the presence of all three dielectric relaxation processes discussed previously, namely IP, α- and β-relaxation. The faster process (β-relaxation) is recorded only as a tendency in the dielectric loss versus frequency spectra and no further analysis can be conducted. Inset of Fig. 5b depicts comparative plots of tanδ as a function of temperature, at 1 kHz, for all studied nanocomposites. All systems acquire low values, which increase with temperature, while the effect of filler content can be considered as negligible.

Fig. 4
figure 4

a 3D plot of the real part of dielectric permittivity versus frequency and temperature for the nanocomposite with 10 phr BN content and b comparative spectra of the real part of dielectric permittivity as a function of temperature, at 1 kHz, for all studied systems

Full size image
Fig. 5
figure 5

a 3D plot of tanδ versus frequency and temperature for the nanocomposite with 10 phr BN content and b variation of tanδ with frequency at 160 °C for all studied systems. Inset depicts tanδ as a function of temperature, at 1 kHz, for all studied systems

Full size image

The dielectric response of the nanocomposites was also studied in terms of the ac-conductivity formalism. In Fig. 6a, the 3D plot of σac as a function of frequency and temperature for the nanocomposite with 10 phr BN content, is shown. Apparently, in the low frequency edge, ac-conductivity approaches constant values, which correspond to the dc ones. This tendency is temperature dependent and constant plateau are formed above 90 °C. On the other hand, at high frequencies the conductivity presents an exponential dependence on frequency. The recorded behaviour is considered as characteristic for the hopping conduction mechanism [27]. Hopping conductivity is related to charge carriers migration such as, electrons, polarones and ions [27,28,29]. At low frequencies, charges tend to cover relatively long distances because of the slow alternation of the field. However, only a limited number of charges is able to overcome the potential barriers exerted by the insulating matrix, leading thus to low conductivity values. At high frequencies, the fast rotation of the field allows an increased number of charges to move (hop) between adjacent sites, covering thus short distances although contributing to the recorded values of conductivity. Therefore, the value of σac acquires higher values in the high frequency range [27, 28]. This behaviour is in accordance with the Universal Law of ac-conductivity, expressed by Eq. (1) [30]:

$$\sigma_{\text{ac}} (\omega ) = \sigma_{\text{dc}} + A(\omega )^{\text{S}}$$
(1)

where σdc is the conductivity at dc conditions and A, S parameters depended on temperature and filler’s content. At temperatures higher than 90 °C, conductivity values, at 0.1 Hz, reach a plateau approaching their dc value. Figure 6b presents the variation of conductivity, at the lowest measured frequency (f = 0.1 Hz), as a function of reciprocal temperature, for all tested nanocomposites. In all cases, conductivity attains low values being in accordance with the insulating nature of both matrix and reinforcing phase. Between 80 and 90 °C, inset of Fig. 6b, a variation of the slope of the σ = f(T) curves is observed, which could be related to the enhanced mobility of macromolecules above the glass transition temperature of the matrix [31]. The high temperature (> 90 °C) experimental data of Fig. 6b are fitted via Eq. (2) which is in the form of Vogel–Fulcher–Tammann (VFT) function for conductivity. Equation (2) is typically used for dc conductivity, although it can be employed at isochronal conditions in the low frequency range [32]. In our case, it is applied at the lowest measured frequency and for the systems in the range of high temperatures. These conditions constitute a concession of approaching dc conductivity values. Values of fitting parameters are listed in Table 1.

$$\sigma_{{0.1\,{\text{Hz}}}} = \sigma_{\text{o}} {\text{e}}^{{ - \frac{\text{D}}{{\text{T} - \text{T}_{\text{o}} }}}}$$
(2)

where σ0 is a pre-exponential factor, D a parameter associated with activation energy, and T0 the Vogel temperature.

Fig. 6
figure 6

a 3D plot of σac versus frequency and temperature for the nanocomposite with 10 phr BN content and b variation of conductivity with reciprocal temperature at 0.1 Hz, for all studied systems, at temperatures higher than 90 °C. Inset provides the performance of ac-conductivity at 0.1 Hz in the entire examined temperature range

Full size image

The functional behaviour of the systems was studied via the dielectric reinforcing function (DRF) which is defined according to Eq. (3) and is considered as a measure of the normalized polarization [33].

$$G(f,T) = \frac{{\varepsilon_{\text{comp}}^{{\prime }} (f,T)}}{{\varepsilon_{\text{matrix}}^{{\prime }} (f,T)}}$$
(3)

where \(\varepsilon_{\text{comp}}^{{\prime }}\) and \(\varepsilon_{\text{matrix}}^{{\prime }}\) are the frequency and temperature dependent real parts of dielectric permittivity of the composite and the matrix, respectively.

The dielectric strengthening of the nanocomposites by the filler is expressed in Fig. 7a. This strengthening is related to the ability to store energy, at constant applied field, since the dispersion of nanoinclusions can act as a distributed network of nanocapacitors [1, 34, 35]. The formed peaks in Fig. 7b originate from α-relaxation, at higher frequencies, and IP process at lower frequencies. IP is expected to increase with filler content, although the common insulating nature of both constituents limits the electrical heterogeneity of matrix and filler, resulting in a rather moderate IP process.

Fig. 7
figure 7

Dielectric relaxation function versus frequency at: a 30 °C and b 160 °C for all studied nanocomposites

Full size image

The relaxation dynamics of α-mode was studied by fitting experimental data via Vogel-Fulcher-Tammann (VFT) equation, for all studied systems. VFT expresses the temperature dependence of the loss peak position of α-relaxation according to Eq. (3):

$$f_{\hbox{max} } = f_{\text{o}} {\text{e}}^{{ - \frac{\text{B}}{{{\text{T}} - {\text{T}}_{\text{V}} }}}}$$
(4)

where fmax is the loss peak frequency at a given temperature, f0 pre-exponential factor, B a parameter associated with activation energy, and TV the Vogel temperature, which is associated with the glass transition temperature (Tg). Fitting curves of experimental data via Eq. (4) are shown in Fig. 8. The frequency of the loss peak location, at every examined temperature, was determined via the clearly formed peaks of the imaginary part of electric modulus for each specimen. Inset in Fig. 8 provides a relative example for the 10 phr BN/epoxy nanocomposite. Obtained values of the fitting parameters are listed in Table 1. Unfilled epoxy displays the highest Vogel temperature, while TV values of nanocomposites do not vary remarkably, indicating indirectly that Tg appears to be almost independent from filler content.

Fig. 8
figure 8

Fitting relaxation dynamics of α-mode via VFT for all studied systems. Inset presents the imaginary part of electric modulus as a function of frequency for all studied temperatures for the 10 phr BN/epoxy nanocomposite

Full size image

At this point, it should be noted that the dynamics of the IP and β-relaxation follow an Arrhenius-type behaviour. However, the limited number of recorded loss peaks, prevented from conducting a reliable fitting process, and for this reason no relevant results are presented.

Mechanical characterization (DMTA)

The variation of storage and the loss modulus with temperature of all studied systems is shown in Fig. 9. The recorded spectra imply viscoelastic behaviour of the examined nanocomposites. Storage modulus (Fig. 9a) increases with filler content, while its curves depict a step-like transition from high to low values. This variation is related to the glass to rubber transition of the polymer matrix. Within the temperature range of this variation, loss peaks are formed in the spectra of loss modulus (Fig. 9b) verifying the occurrence of α-relaxation. Corresponding glass transition temperature to the loss modulus peak becomes evident that with the exception of the system with 3 phr BN which shifts to higher values, Tg of all other nanocomposites and unfilled epoxy does not vary significantly. The determined values of Tg are listed in Table 1.

Fig. 9
figure 9

a Storage and b loss modulus as a function of temperature for all BN/epoxy nanocomposites

Full size image

Conclusions

In the present study, polymer matrix/BN nanocomposites were fabricated and tested with filler content as parameter. Their morphological characterization was held via SEM, XRD and FTIR techniques. In all cases, the successful implementation of BN nanoinclusions as well as their fine dispersion was achieved. The dielectric response of nanocomposites was investigated in a wide frequency and temperature range. Dielectric permittivity values of the reinforced systems, as a function of frequency and temperature, are higher than the corresponding ones of neat epoxy resin. Moreover, an increase in the real part of the dielectric permittivity of the composites with increasing temperature was observed. Three relaxation mechanisms were detected, in the dielectric loss spectra, which are attributed to the transition from glass to elastomeric phase (α-relaxation) of the matrix, re-orientation of polar side groups of the polymeric chains (β-relaxation) and to interfacial polarization (IP or MWS effect) between matrix and encapsulated nanophase. The variation of ac-conductivity with frequency for all studied systems is in accordance with the Universal Law of ac-conductivity. Moreover, the functionality of the composites was studied via DRF, revealing the ability of the reinforced systems to store enhanced amounts of energy. The dynamics of α-relaxation was studied via VFT equation. Finally, the dynamic mechanical behaviour of the fabricated systems was investigated via DMTA. Storage modulus increases with filler content, while formed peaks in the loss modulus spectra indicate the transition from glassy to rubbery phase.