Arti Gupta*a and
Ram Pal Tandonb
aDepartment of Physics & Astrophysics, University of Delhi, Delhi 110007, India. E-mail: artigupta80@gmail.com; Fax: +91-27-667061; Tel: +91-2766-7725 ext. 1367
bDepartment of Physics & Astrophysics, University of Delhi, Delhi 110007, India
First published on 6th January 2015
In this paper, we illustrate the significant optical and magnetodielectric properties of the polyvinylidene difluoride–Co0.6Zn0.4Mn0.3Fe1.7O4 nanocomposite film, containing the optimum weight content (0.5%) of CZFMO nanoparticles (average particle size ∼ 20 nm). The photoluminescence spectra of the nanocomposite film clearly demonstrated stable and sharp luminescence peaks at 430 nm (indigo region), 458 nm (blue region) and 490 nm (green region), accompanied by considerably higher intensity, compared to the broad emission peak (450–550 nm) observed for pristine polyvinylidene difluoride film. The results of fluorescence spectroscopy confirmed those obtained in photoluminescence. Polymer nanoparticle surface interaction could be the possible mechanism behind the enhanced optical properties. Additionally, this nanocomposite film exhibited frequency independent, linear and negative magnetodielectric effect ∼0.08% at the maximum applied field of 5 Tesla.
Notably, in literature inorganic nanofillers leading to the enhanced optical properties1,3 of organic polymer matrix, are limited to the semiconductor quantum dots of PbSe, CdS, CdSe, etc. Moreover, these semiconductor quantum dots are always used in core–shell structure along with surface modification/engineering to protect their PL features. The major concern with these semiconductor based polymer nanocomposites is the toxicity of Pb, Se and Cd (doped ZnS1 is the only known example of nontoxic semiconductor, used in fabrication of polymer–inorganic nanocomposites), which restrains their commercial use. On the other hand, investigations1,3 on iron oxide based (Fe2O3, Fe3O4, (Ni, Zn)Fe2O4) polymer nanocomposites are typically focused on their applications in electrical/magnetic shielding, nonlinear optics, magnetic electrocatalysis, microwave absorption etc.,due to unique combination of electrical and magnetic properties in these materials.
Inorganic nanosized ferrites exhibit interesting optical properties, compared to their bulk counterparts and, addition of even a very small amount of inorganic ferrite nanoparticles to host organic polymer matrix can substantially enhance the optical properties of resulting nanocomposite. However, this is still away from practical realization and, thus, merits investigations.
In our earlier publication,17 we reported novel as well as interesting magnetic/magnetostrictive properties of bulk Co0.6Zn0.4Mn0.3Fe1.7O4 (CZFMO). Here, we report the enhanced optical properties of PVDF–CZFMO nanocomposite film, containing the optimum weight content (0.5%) of CZFMO nanoparticles. This nanocomposite film displayed stable and sharp luminescence peaks at 430 nm (indigo region), 458 nm (blue region) and 490 nm (green region) accompanied by higher intensity, compared to broad emission peak (450–550 nm) observed for pristine PVDF film. Furthermore, magnetodielectric (MD) studies on nanocomposite film illustrated frequency independent, linear and negative MD effect (∼0.08% at H = 5 T). Clearly, PVDF–CZFMO nanocomposite film has the crucial advantages of non-toxicity and simplicity of preparation procedure (discussed in the Experimental section).
Notably, the state of dispersion and the nature of the interface between nanoparticle and host matrix largely affect the polymer nanoparticle interaction and overall performance of polymer nanocomposite. Consequently, we found that higher particle loading (>0.5 wt%) of CZFMO nanoparticles in PVDF–CZFMO nanocomposite film, with agglomeration and inhomogeneous dispersion, results in broad emission peak with low intensity, almost identical to that of pristine PVDF film. Here onwards, the results obtained on optimum PVDF–CZFMO nanocomposite film are discussed unless otherwise specified.
Surface topographical images of nanocomposite and pristine PVDF films are shown in Fig. 3(a) and (b), respectively. Topography of nanocomposite film is significantly different from pristine PVDF film due to embedded CZFMO nanoparticles. Root mean square roughness (RMS) or Sq. are 137.64 nm and 288.11 nm for pristine PVDF and nanocomposite films, respectively (in the spin coating method high humidity and low substrate temperature lead25 to highly rough PVDF films). Higher surface roughness for nanocomposite film could be an outcome of stress/phase transformation induced by particle loading. Surface SEM images for pristine PVDF and nanocomposite films are shown in Fig. 4(a) and (b), respectively. For pristine PVDF film, tree like spherulites were observed, which grew from their primary nucleation sites, developed during the spin coating process. For nanocomposite film, CZFMO nanoparticles can be identified as bright dots embedded in the dark (appearing dark due to high resistivity) PVDF matrix. This further confirms the formation of nanocomposite film without any detectable porosity. Comparison of surface SEM images for pristine PVDF and nanocomposite films indicate higher roughness for nanocomposite film, consistent with AFM data. Particle size of CZFMO estimated from this SEM image is roughly in agreement with that determined using the TEM micrograph.
Raman spectra for pristine PVDF and nanocomposite films are shown in Fig. 5. Broad Raman bands with low intensity are characteristics of semicrystalline PVDF. For both samples, three Raman bands at 564 cm−1 (ν1), 801 cm−1 (ν2) and 1094 cm−1 (ν3), were observed. Band (ν1) at 564 cm−1 appeared at an intermediate position between the band at 538 cm−1, related to the CF2 deformation mode and band at 609 cm−1, related to the CF2 wagging mode.26,27 In literature,28,29 band at 794 cm−1 is attributed to α phase of PVDF, whereas, band at 811 cm−1 is attributed to γ phase of PVDF. For our samples, the appearance of the Raman band at an intermediate position of 801 cm−1 possibly indicates the presence of mixture of α and γ phases. Band at 1094 cm−1 (ν3)28,29 can originate due to a combination of different phases of PVDF. Two noticeable changes found in the spectra for nanocomposite film, were as following: (i) intensity ratio (Iν3/Iν1) increases (ii) peak (ν2) becomes sharp and intense (without any detectable change in the position). These changes signify the effectiveness of the polymer nanoparticle surface interaction.
The optical absorption spectra in the wavelength range (200–600 nm) for pristine PVDF and nanocomposite films are shown in Fig. 6(a). The absorption edge was found at ∼322 nm for pristine PVDF and at ∼335 nm for nanocomposite film. Inclusion of CZFMO nanoparticles in PVDF matrix increases absorption and shifts the position of absorption edge towards higher wavelength or lower frequency side, demonstrating red shift in spectra. Transmittance (%) vs. wavelength (nm) plots for PVDF and nanocomposite films are shown in Fig. 6(b). Significant reduction in transmittance for nanocomposite film is evident. For polymer nanocomposites, transmittance is largely affected by the surface roughness, particle size, dispersion of nanoparticles, polymer nanoparticle interface, refractive index, etc. For nanocomposite film, decrease in transmittance can be caused4,30 by Rayleigh scattering (by CZFMO nanoparticles) and relatively high surface roughness (evident from the AFM and surface SEM images). The optical band gap energy was estimated using the following Tauc relation31
(αhν)1/n = A(hν − EG) | (1) |
![]() | ||
Fig. 6 UV-visible spectra for pristine PVDF and nanocomposite films (a) absorbance and (b) transmittance. |
The PL spectra (excitation wavelength = 400 nm) for pristine PVDF and nanocomposite films are shown in Fig. 8. Nanocomposite film displayed stable and sharp luminescence peaks at 430 nm (indigo region), 458 nm (blue region) and 490 nm (green region), accompanied by higher intensity, compared to broad32 emission peak (450–550 nm) observed for pristine PVDF film. The first band at 430 nm (∼3 eV) is close to the direct band gap energy value (estimated from optical absorption spectra) and could be attributed to the band edge emission. The second band at 458 nm (2.7 eV) could be ascribed to the surface trap states associated with nanoparticles, whereas, the third band at 490 nm (2.5 eV) could be attributed to the deep trap states. The PL spectra for CZFMO nanopowder (inset of Fig. 8) demonstrated a broad emission peak (420–440 nm). This broad emission peak arises due to possible overlap of d–d transitions 4E, A1(4G) 6A1(6S). For nanocomposite film, PL features like sharp emission peaks with high intensity are unlike those observed for both PVDF and CZFMO. Enhanced PL efficiency for nanocomposite film should be caused by the polymer nanoparticle surface interaction. This interaction involves the electric field produced by the surrounding highly electronegative CF2 dipoles of polymer chains, that could promote the 3d–4s 4p orbital (4s and 4p orbitals are hybridized) coupling. Such coupling gives rise to populated 4s and 4p orbitals and delocalization of electrons.33 Consequently, electrons from 4s and 4p orbitals take part in the band edge transitions. This resulted in stronger PL intensity for nanocomposite film.
![]() | ||
Fig. 8 PL spectra for pristine PVDF and nanocomposite films. inset shows PL spectra for CZFMO nanopowder. |
The fluorescence spectra for PVDF and nanocomposite (solution form) are shown in Fig. 9(a). For nanocomposite, the spectra illustrated three emission bands at 430 nm, 458 nm and 490 nm (appearing as a shoulder), accompanied by significantly higher intensity, compared to broad emission peak (450–550 nm) observed for PVDF. Clearly, the results of fluorescence spectroscopy are in complete agreement with those obtained in PL. In order to probe the dynamic behavior of involved electronic transitions, time resolved fluorescence spectroscopy (TRFS) (Fig. 9(b)) was employed. Notably, the mean lifetime of exciton, determined using TRFS, gives useful information regarding the quality of material and its device performance. The decay profile was fitted using three exponentials for PVDF and using two exponentials for nanocomposite. Using this data, the mean life time was determined to be ∼1.61 × 10−9 s (χ2 = 1.03) and 1 × 10−10 s (χ2 = 0.96) for PVDF and nanocomposite, respectively. Since, the efficiency of radiative recombination is inversely proportional to the decay time, more (roughly by an order of magnitude) efficient recombination could be anticipated for nanocomposite (attractive feature for device applications).
![]() | ||
Fig. 9 (a) fluorescence spectra for pristine PVDF and nanocomposite (solution form) (b) time resolved fluorescence data. |
The XRD patterns for pristine PVDF and nanocomposite (sheet form) are shown in Fig. 10. For pristine PVDF, four peaks at ∼17.6°, 18.2°, 19.8° (most intense) and 26.7° corresponding to (100), (020), (110) and (021) crystalline planes of α phase of PVDF,29 were noted. Whereas, the XRD pattern for nanocomposite showed a striking difference; peak at 18.2° corresponding to (020) plane has the highest intensity, unlike that observed for pristine PVDF (peak positions are essentially unchanged). Interestingly, Steinhart et al.34 first reported such observation for nanotubes consisting of α-PVDF; therein, XRD pattern revealed the b-axis oriented crystallites of PVDF inside the wall of the nanotube (b-axis of unit cell was oriented parallel to the long axis of the nano tube). For PVDF–CZFMO nanocomposite, the substantial change in XRD pattern signifies the strength of polymer nanofiller surface interaction. We believe that the pores on the surface of spherical CZFMO nanoparticle can act as templates and promote the oriented growth of α-PVDFcrystallites (depending on the geometry of the pores, i.e. shape, size etc.) in nanocomposite. Evidently, the growth took place along the b-axis (major growth axis), giving a manifestation of anisotropy in nanocomposite, unlike the isotropic reflections observed for pristine PVDF.
Magnetization (M) vs. temperature (T) plots for nanocomposite in ZFC and FC conditions in the temperature range (2 K ≤ T ≤ 300 K) at H = 1 T are shown in Fig. 11. At high magnetic field of 1 T, complete alignment of individual magnetic moments of CZFMO nanoparticles in the direction of applied field takes place. Consequently, in the entire temperature range of investigation, ZFC and FC plots completely overlapped each other, demonstrating the existence of long range magnetic ordering. Magnetization decreases with the increase in temperature due to the added thermal energy, which breaks the collinear arrangement of magnetic spins. Clearly, the M–T plot showed the normal expected ferrimagnetic35 behavior, confirming the magnetic ordering of dispersed CZFMO nanoparticles (the superparamagnetic behavior of CZFMO nanoparticles is reported in our recent publication36).
![]() | ||
Fig. 11 Magnetization (M) vs. temperature (T) plots for nanocomposite sheet in ZFC and FC conditions at H = 1 T. |
The dielectric constant (ε′) vs. frequency (f) plots at room temperature in the frequency range (100 Hz–110 MHz) for pristine PVDF and nanocomposite are shown in Fig. 12. In the entire frequency range, ε′ and tanδ for nanocomposite are higher than those of pristine PVDF. Similar observations were reported in literature for other PVDF based nanocomposite systems.37,38 For present instance, enhancement in ε′ could be attributed to the presence of oriented crystallites of PVDF in nanocomposite. Similarly, higher tan
δ for nanocomposite reflects the difference between oriented and unoriented form of α-PVDF crystallites. Here, the ε′–f plots also revealed the thickness resonance mode for pristine PVDF and nanocomposite. The resonance (fR) and antiresonance (fA) frequencies for pristine PVDF were found to be 83 MHz and 99 MHz, respectively, with bandwidth (Δf = fA − fR) value of 13 MHz. For nanocomposite, a considerable increase in value of tan
δ close to resonance, was noted, however, the values of fA, fR and Δf are essentially same as pristine PVDF.
![]() | ||
Fig. 12 Dielectric constant spectra for pristine PVDF and nanocomposite sheets in the frequency range (100 Hz–110 MHz) at room temperature. Inset shows corresponding tan![]() |
P–E loops for pristine PVDF and nanocomposite are shown in Fig. 13. Evidently, loops do not saturate up to the maximum applied electric field of 200 kV cm−1. For pristine PVDF, the maximum polarization (Pmax), remnant polarization (PR) and coercivity (EC) were observed to be 0.116 μC cm−2, 0.015 μC cm−2 and 24.9 kV cm−1, respectively. Whereas, for nanocomposite, Pmax, PR and EC were observed to 0.134 μC cm−2, 0.016 μC cm−2, 23 kV cm−1, respectively. Higher polarization values (Pmax & PR) for nanocomposite can be ascribed to the presence of oriented crystallites of α-PVDF (concurrently with the dielectric data).
![]() | ||
Fig. 13 Polarization (P) vs. electric field (E) loops at room temperature for pristine PVDF and nanocomposite sheets. |
MD plots for nanocomposite at two different frequencies 30 kHz and 300 kHz are shown in Fig. 14. MD effect basically probes the coupling between magnetization and polarization in a magnetoelectric material. Change in dielectric constant (ε′) on application of magnetic field, is measured in term of MD effect, defined as [ε′(H) − ε′(0)/ε′(0)%]. Where ε′(H) is the permittivity in the presence of field H and ε′(0) is the permittivity at H = 0 Oe. The frequency independent MD effect observed for nanocomposite confirmed the absence of extrinsic contribution from magnetoresistance together with Maxwell–Wagner polarization. MD (%) almost linearly decreases with the increase in field, attaining the maximum MD ∼0.08% at the highest applied field of 5 T. In literature, negative MD effect was demonstrated for PVDF–Ni0.5Zn0.5Fe2O4 (PVDF–NZFO) nanocomposite films.39 Whereas, the positive MD effect was shown40 for PVDF–Fe3O4 nanocomposite film.40 Considering the magnitudes, maximum MD = 5% (at the maximum applied field ∼1 T) was obtained in PVDF–NZFO nanocomposite film, containing 20 wt% loading of NZFO nanoparticles and, maximum MD = 0.6% (at the maximum applied field ∼2 T) was obtained in PVDF–Fe3O4 nanocomposite film, containing 9.09 wt% of Fe3O4 nanoparticles. Low value of MD (∼0.08% at the highest applied field of 5 T) for PVDF–CZFMO nanocomposite can be ascribed to the small loading (0.5 wt%) of CZFMO nanoparticles.
![]() | ||
Fig. 14 Room temperature MD% vs. field (H) plots in the field range (−5 T ≤ H ≤ 5 T) for nanocomposite sheet at two different frequencies of 30 kHz and 300 kHz. |
Importantly, PVDF–CZFMO nanocomposite displayed linear MD effect unlike the nonlinear MD effect observed for other nanocomposites.39,40 The mechanism behind the linear MD effect can be explained as following: when magnetic field is applied, the spins of CZFMO nanoparticles orient themselves along the direction of the field. Subsequently, the spin–charge coupling between polymer and nanoparticle redistributes the charge on the dipoles of polymer chains, leading to a change in polarization/capacitance. This spin–charge coupling between CZFMO nanoparticles and PVDF chains induces linear change in dielectric constant with the change in magnetization. Thus, it is proportional to the P2M term in the expression of free energy, unlike the dominant contribution of quadratic P2M2 term (arising from the strain mediated mechanical coupling between piezoelectric and magnetostrictive phases) considered for PVDF–Fe3O4 and PVDF–NZFO nanocomposites.39,40
One of the authors (A.G) would like to thank DST-INSA for providing Inspire Faculty Fellowship. We would like to thank Prof. S. Patnaik at JNU (India) for MD measurements.
This journal is © The Royal Society of Chemistry 2015 |