Introduction

Polyaniline (PANI) is environmentally stable and highly tuneable conducting polymer, which can be produced as bulk powder, cast films, or fibres (MacDiarmid and Epstein 1989). Polymeric chain consists of a succession of reduced benzenic nucleus and oxidised quinoidic nucleus. Both insulating emeraldine base (EB) and conducting emerladine salt (ES) are the stable forms. (Cao et al 1992; MacDiarmid and Epstein 1994). ZnO is semiconducting in nature having unique properties like large exciton binding energy (60 meV) and relatively large direct band gap of ~3.3 eV at room temperature (Andeen et al 2003; Chen et al 1998). Advantage associated with these properties includes higher breakdown voltages, ability to sustain large electric fields, low electric noise and high temperature and high power operations (Ozgur et al. 2005). It has been studied extensively for making optical and electronic devices like: light emitting diodes, solar cells, transducers, varistors, photo detectors, etc., (Saito et al. 2002; Koch et al 1995; Liang et al 2001). The incorporation of nanoparticles into the polymer offers enhanced performance for both the host and the guest. The polymer nanocomposites are promising candidates based on the fact that the small-sized particles enhance the properties while the polymer matrix offers flexible functionalities to control host–guest interaction to ensure the growth and distribution of nanoparticles (Gangopadhyay 2000). Hence by choosing particular choice of the matrix and the filler particles, one can modify the properties of the composite for a particular application.

Some study has been done on the bulk PANI–ZnO composites for its dielectric behaviour in microwave region (Sharma et al 2009). As nanoparticles have large surface to volume ratio, they lead to change the properties as compared to the bulk material. Hence, the incorporation of the ZnO nanoparticles in the EB form may lead to certain different results as compared to the bulk particles.

However to the best of our knowledge no study has been reported on dielectric behaviour of EB–ZnO nano-composite in this frequency region (1 kHz–1 MHz). The dielectric properties of such a material are much relevant for telecommunication applications in aerospace and for EMI shielding. A low and stable dielectric constant is often desirable for application such as high-speed integrated package and satellite communication. In this paper we report synthesis of PANI–ZnO nanocomposite films and their dielectric behaviour in the frequency range (1 kHz–1 MHz).

Experimental

Materials and synthesis

High purity zinc acetate dihydrate (98%), hydrochloric acid (35% GR), ammonium peroxidisulphate (98%), methanol (99.8%), aniline (99.5%), N-methyl-2-pyrolidone (NMP) (99%), and ammonia (30% GR) were purchased from Merck & Co., Inc., and were used as received. Distilled water was employed as a medium to polymerize aniline.

EB was synthesized by chemical oxidative polymerization method (Stejskal 2002). In a typical experiment to polymerize aniline, 5 ml of aniline monomer was mixed with 100 ml of 2 M HCl solution in a glass beaker. The mixture was continuously stirred for half an hour at 0 °C. 8.0 g ammonium peroxidisulphate was dissolved in water by its minimum quantity to prepare a solution, which is then added drop-wise to the above mixture with constant stirring. The colour of the reaction mixture changed to dark green. This solution was kept for overnight. The solution was filtered in a Buchrer funnel and the residue obtained was washed several times with distilled water till no acid remained. To make PANI soluble in organic solvent, the residue was treated with 10% ammonia solution to form PANI EB form . The solution and precipitates turned blue. The precipitates were filtered, washed with distilled water and the residue was dried in oven at 60 °C for 5 h. It was grounded with a mortar and pestle, sieved through 20-μm sieve and stored in a glass container. It is an insulating form.

ZnO nanoparticles was synthesised as described earlier (Seow et al. 2009). In a typical experiment, zinc acetate dihydrate (3.35 mmol) was first dissolved in methanol (31.25 ml) and another solution of potassium hydroxide (6.59 mmol) was prepared by dissolving it into methanol (16.25 ml). The potassium hydroxide solution was added drop-wise to the zinc acetate solution at 60 °C under vigorous stirring. After 1.5 h, nanoparticles started to precipitate and the solution became turbid. The heating and stirring were discontinued after 2 h and the solution was allowed to stand for another 2 h. The ZnO nanoparticles settled at the bottom and the excess mother liquor was removed. Precipitates were washed twice with methanol (12.5 ml). A portion of precipitate was dispersed in a mixture of 12.5 ml of methanol and 2.5 ml of chloroform for TEM imaging. The dispersed solution was translucent and stable for about 2 weeks. The other portion of precipitate was air dried and stored as ZnO nanoparticles.

Growth of films

The known amount of EB powder (1 g) was dissolved in NMP (10 ml). The solution was stirred for 4 h to make EB completely soluble. This solution was filtered using a Whatman filter paper having pore size of 6 μm to get clear solution. A film of this solution was deposited by solution casting technique on the ITO-coated glass substrates. These films were allowed to dry at room temperature for 72 h.

For the EB–ZnO nanocomposites film (10%), ZnO nanoparticles (0.01 gm) were dispersed in 1 ml of EB solution in NMP (containing 0.1 g EB). Similarly, 20 and 30% ZnO nanoparticles were dispersed in EB–NMP solution. Each solution was stirred for 5 h continuously. The films from these composites solutions were prepared on an ITO-coated glass substrate by the solution casting technique. These films were allowed to dry at room temperature for 72 h. All characterisations and dielectric behaviour were studied using film specimens having uniform thickness of 5.0 ± 0.02 μm.

Characterisation techniques

The thickness of EB as well as composite films was measured with the help of a mechanical profilometer (From Talysurf Series 2 Instrument). X-ray diffraction studies were carried out using a PANalytical’s X’Pert PRO X-ray diffractometer system in 2θ range of 20–80°. XRD patterns were analyzed by matching the observed peaks with the standard pattern provided by the JCPDS file. The lattice parameters for ZnO were calculated for nanoparticles and for composite. EB and EB–ZnO (20 wt%) composite was studied in the frequency range of 500–2000 cm−1 by Fourier Transform Infra Red (FTIR) spectroscope (Model: Perkin Elmer RX-1). A transmission electron microscopy (TEM model: Hitachi H7600) was used to investigate the dispersion of ZnO particles in EB. Dielectric measurements of films of EB and EB–ZnO composites were carried out by Network analyzer (model: HIOKI3532-50 LCR HiTESTER) in the frequency range of 1 kHz–1 MHz. Figure 1 shows the schematic presentation of the measurement of the dielectric constant.

Fig. 1
figure 1

Schematic diagram used for the dielectric constant measurement of EB and EB–ZnO nanocomposite films on ITO coated glass

Full size image

To measure the dielectric properties, one half area of the glass substrate was coated with EB or composites and another half was uncoated to form a capacitor. Conducting silver paste was used to form the electrodes in contact with coated and uncoated side. Silver-based conducting paste is widely used. They have reasonably high conductivity and provide good contact to the materials (in this case polymer films). Silver paste is not directly soldered and conduction relies on point to point contact between polymer film and ITO surface. Moreover, polymer composite as a whole is a neutral species therefore it does not react with silver paste. Similar configuration has been reported in literature (Harsányi 1995). The dielectric constant (εr) of each film was calculated using the equation:

$$ C = \varepsilon_{\text{r}} \varepsilon_{\text{o}} \, a/d $$

Where εo is the vacuum permittivity, which equals 8.85 × 10−12 F/m, a is the electrode area and d is the thickness of film.

Results and discussion

Structure and morphology

The XRD patterns of EB, ZnO nano-crystal and EB-ZnO (30%) composite films are as shown in Fig. 2. The XRD pattern of EB is in conformation with the earlier reported results as only a broad peak was observed at 2θ = 23° (Fig. 2a) showing the amorphous nature (Pouget et al. 1991).

Fig. 2
figure 2

XRD pattern for (a) EB powder, (b) ZnO nanoparticles film grown over ITO-coated glass and (c) EB-ZnO (30%) nanocomposite film over ITO coated glass

Full size image

Figure 2b shows the powder XRD patterns of the ZnO nanoparticles synthesized at 60 °C after its solution casting on the glass substrate. The XRD patterns of as obtained ZnO nanoparticles are of hexagonal wurzite structure. The characteristic peaks are observed at 31.720, 34.500, 36.300, 47.760, 56.60, 63.100 and 68.000 which corresponds to miller indices (100) (002) (101) (102) (110) (103) and (201), respectively. The lattice constants are obtained as a = b= 3.2269 Å and c = 5.2539 Å using powder XRD software. All the diffraction peaks are in agreement with JCPDS data (card no. 36-1451, a = b = 3.249 Å, c = 5.206 Å). The results indicate that product consist of pure phase and no characteristic peaks for other impurity like Zn(OH)2. The sharp peak indicates that the nanoparticles are well crystallized. The small particles size makes the diffraction peaks broader than corresponding standard pattern. In contrast to the standard card, the (002) peak is stronger, revealing the (001) oriented growth of ZnO nanoparticles.

ZnO has a hexagonal closed-packed wurtzite structure having alternating planes of Zn and O atoms. The (002) plane is parallel to the planes which are containing O and Zn atoms while the (100) plane is perpendicular to the (002) plane, intersecting alternating layers of Zn and O atoms. The crystalline size in (002) and (100) plane can be essentially considered as matrices of diameter (D) and length (L), respectively.

The crystallite size of ZnO was found to be ~9.5 ± 0.5 nm as calculated from the first three prominent peaks using the Scherrer formula (L = 0.93 λ/βcosθ), where L is the average crystallite size, λ is the wave length of Cu (1.54 Å), β is the full width of diffraction line at half the maximum intensity and θ is the Bragg angle.

The XRD of the composite (Fig. 2c) has similar pattern as that of ZnO powder; however the peak intensity was modified. ZnO has retained its hcp wurtzite structure in the composite. The presence of XRD peak of emeraldine at 2θ value of 23° with lower intensity indicates the dilution of the EB concentration due to adsorption of EB chains on nanoparticle surface. From the XRD peaks, it was observed that lattice parameter for nanocomposite were a = b = 3.2442 Å, c = 5.2061 Å. This change of lattice parameter of ZnO nanoparticle on formation of composite with EB shows that the unit cell of ZnO stretches along horizontal direction and suppresses along c-axis. This might be attributed due to higher surface along the prism plane as compared to the basal plane in hexagonal close packed crystal structure. Similar results have been reported (Sharma et al 2009) for PANI–ZnO composites. The stretching of unit cell in a–b plane and suppression along c-axis suggests strong interaction between ZnO nanoparticles surface with the EB molecular chains. The crystallite size calculated from the first three major peaks found to be ~11.5 ± 0.5 nm.

The typical TEM images of as prepared ZnO nanostructure is given in Fig. 3. It is clear from the image that the spherical nanoparticles are mono dispersed with size 5–9 nm ranges and maximum particles are of 6 nm size. TEM images of EB-ZnO nanocomposite in Fig. 4 clearly shows entrapping of ZnO nanoparticles in emeraldine matrix. Moreover, the nanoparticles still remain monodispersive with average diameter around 10 nm. The increase in the size of nanoparticles in the nanocomposite indicates that the surface of nanoparticle has interaction with EB chains, which is also clear from FTIR analysis. Similarly, XRD analysis also shows the increase in size of ZnO nanoparticles in the composite. XRD average nanoparticle size is somewhat larger in comparison to TEM image, because for XRD analysis nano-powder was prepared by air drying the synthesized particles. There is a possibility of agglomeration as no surfactant has been used for capping during synthesis. Moreover, TEM image gives the size of the individual nanoparticle, whereas the size calculated from the XRD patterns represents average size.

Fig. 3
figure 3

Typical TEM image of the synthesised ZnO nanoparticles

Full size image
Fig. 4
figure 4

Typical TEM of EB–ZnO (30%) nanocomposite

Full size image

FTIR

FTIR spectra of pure EB and EB–ZnO (30%) nanocomposite in KBr are shown in the Fig. 5. The characteristics peaks of PANI at 798.7, 1121.8, 1296.7, 1472.4 and 1559.8 cm−1 corresponds to the C–H out of plane, C–H in plane bending, C–N stretching vibration of secondary amine, stretching mode of the N–B–N benzoid rings, stretching mode of the N=Q=N quinoid rings, respectively (Tan et al. 2009; Zeng and Ko 1997). The nanocomposite also shows the similar characteristic peaks. But the corresponding peaks of pure EB have been shifted to 794.3, 1136.0, 1296.9, 1490.2, and 1580.2 cm−1 in nanocomposite. The shift may be described due to the formation of hydrogen bonding on the surface of ZnO nanoparticles and –NH– group of EB. Similar interaction has been also reported earlier (He 2005; Paul et al 2007).

Fig. 5
figure 5

FTIR spectra of (a) EB, (b) EB–ZnO (30%) nanocomposite

Full size image

Dielectric behaviour

Variation of dielectric constant (εr) of EB and nanocomposite films as a function of frequency (1 kHz–1 MHz) is shown in Fig. 6. The dielectric constant of EB film decreases from 76.19 to 67.63 in the frequency range 1–400 kHz and then increase to 68.27 in frequency range upto 1 MHz. For nanocomposite with 10% ZnO nanoparticles εr decreases from 45.2 to 39.7 in the frequency range 1–400 kHz and then again increases to 41.43. For nanocomposite with 20% ZnO nanoparticles, composite εr decreases from 24.86 to 23.31 in 1–500 kHz frequency range and then changes minutely to the value 23.87 upto 1 MHz frequency, indicating that with increase of nanoparticle concentration in the nanocomposite, the values εr is lowered and then tends to be saturated. However, nanocomposite with 30% ZnO nanoparticles, εr decrease from 8.57 to 6.56 in the frequency range 1 to 400 kHz. There is no further variation in the values of εr. The dielectric constant becomes constant in the frequency range 400 kHz–1 MHz which is important for many technological applications. The value of εr for the nanocomposite with 10% ZnO, 20% ZnO and 30% ZnO decrease by a factor ~2, ~3 and ~9, respectively, in comparison to pure EB. Average measurement values of dielectric constant w.r.t. frequency (for ten observations) of each sample along with error bars are given in Table 1. The variation in the measured values of the dielectric constant ranges from ±0.02 to ±0.05.

Fig. 6
figure 6

Variation of dielectric constant in frequency range of 1 kHz-1 MHz for. (a ) EB, (b ) EB–ZnO (10%) nanocomposite, (c ) EB–ZnO (20%) nanocomposite, (d ) EB–ZnO (30%) nanocomposite

Full size image
Table 1 Average measurement values of dielectric constant w.r.t. frequency (for ten observations) of each nanocomposite sample along with error bars
Full size table

The decrease in dielectric constant value with variation in concentration of ZnO nanoparticles in the nanocomposite can be explanined on the basis of lower dielectric constant of ZnO than EB and multilayered core model (Tanaka and Kozako 2005). The dielectric constant of ZnO has been reported in the range of 7.5–10 depending on processing conditions (Crisler et al 1968). The lower dielectric constant of 30% ZnO contained nanocomposite might be attributed due to low value of ZnO nanoparticles and their distribution in polymeric matrix (Dang et al 2003, Huang 2010). Multilayered core model accounts for the filler particles which are spherical in nature. Thus, this model has applied in present studies as synthesized ZnO nanoparticles are nearly spherical in shape as observed from TEM images and the ZnO nanoparticles are stable in EB matrix as indicated by XRD study. According to this model, the interface between inorganic (ZnO) filler nanoparticles and polymer EB matrix contain three different layers: (a) bonded layer, (b) bounded layer and (c) loose layer.

The first layer (bonded layer) is due to the hydrogen bonding between the ZnO nanoparticles and the EB chains which is a thin molecular transition layer. The second interfacial layer (bounded layer) is formed as other EB chains are strongly bound or interacted to the first layer and ZnO nanoparticles. This interaction has been explained on the basis of the defects present in the ZnO structure. In nanostructured ZnO, the small length scales and large surface-to-volume ratio mean that surface defects play a stronger role in controlling properties. The dominant defects which occur in ZnO structure is the oxygen vacancies and Zn-interstitials (Hagemark 1976). These defects donate two electrons resulting in over all negative charge on the ZnO structure. Due to these defects, the EB chains interact strongly with ZnO nanoparticles. The third layer is formed due to the loose interaction of the EB chains with second layer. This interaction leads to the change in chain conformation, chain mobility, crystallinity, and free volume of the loose layer as compared to the polymer matrix.

The reduction in permittivity has been explained by the contribution from both, bounded and loose layer.

Role of bounded layer:

The dielectric constant of the EB is due to the orientation polarization (Joseph Mathai et al. 2002). In the presence of electric field, the strongly polarizable bond (C=N) tend to orient in the direction of the field. However, this orientation of polar group is a slow process, but in lower frequency range (1 kHz–1 MHz) there is sufficient time for the complete orientation of the polar bond in the field direction. Thus this polarization may get completed in this frequency region. The dielectric constant of the EB is due to the strongly polarizable C=N bond in conjugated polymer chain. EB chains interact with polar ZnO nanoparticles in inner bounded layer in the presence of electric field. The polarity of ZnO impair the motion of the C=N dipoles resulting in decrease of dielectric constant.

Role of loose layer

The second layer is formed due to interaction of EB chains with oxygen defects present on the surface ZnO nanoparticles. As the ZnO nanoparticles have high surface energy, thus interaction becomes stronger in second layer. However, further adsorption /diffusion of EB polymer chains with second layer becomes weak resulting in reduced free volume. It has already been discussed that free volume in polymer-filler composite system decreases in nanostructure composite materials and increases in microstructure composite materials (Nelson and Hu 2004). This reduction in free volume explains the reduction in permittivity (Bal and Kothari 2009). With the increase of ZnO nanoparticle concentration in the composite, the dielectric constant attains nearly lowest and stable value.

Relation of the multi-core model with FTIR, XRD patterns and dielectric properties

It is experimentally studied that dielectric constant tends to decrease in EB with the increase of nanostructuration. This phenomenon has been interpreted by constrained movement of dipoles in the interface between ZnO nanofiller and surrounding EB polymer matrices. Spherical shape of ZnO in nanometric size as nanofiller, content in polymer matrix, inter filler distances and interfacial morphology play their role for dielectric variation which has been explained on the basis surface defects in ZnO crystal and various interactions of EB chains with ZnO nanoparticle. FTIR studies of the composite shows the presence H-bonding between ZnO nanoparticles and –NH- group of EB resulting in the formation of bonded layer. XRD patterns clearly confirm the dilution of EB peak due to adsorption/interaction between ZnO and EB. Formation of second bound layer has been explained on the basis of defect sites in nanoparticles interaction with the EB chains. Loose layer formation has been explained on the basis of high surface to volume ratio of ZnO nanoparticles. However, further studies are being carried with polypyrrole/polythiophene-metal oxide nanoparticles to prove the concept.

Conclusions

The dielectric behaviour of the EB–ZnO nanocomposites films was found to change drastically in comparison to pure EB in the region 1 kHz–1 MHz with the increase of nanoparticles concentration in the composite. This change in dielectric behaviour is due to the multilayered interface formed between ZnO nanoparticles and EB. The high energy surface of the ZnO nanoparticles leads to the decrease in dielectric constant. TEM and XRD analysis show some enlargement of the ZnO nanoparticles embedded in the EB matrix. FTIR also confirms the attachment of the EB chains with nanoparticles.