- Email: [email protected]
Interfaced conducting polymers
Synthetic Metals 224 (2017) 109–115
Contents lists available at ScienceDirect
Synthetic Metals journal homepage: www.elsevier.com/locate/synmet
Interfaced conducting polymers J. Stejskala,* , P. Bobera , M. Trchováa , D. Nuzhnyyb , V. Bovtunb , M. Savinovb , J. Petzeltb , J. Prokešc a b c
Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, 162 06 Prague 6, Czech Republic Institute of Physics, Academy of Sciences of the Czech Republic, 182 21 Prague 8, Czech Republic Charles University, Faculty of Mathematics and Physics, 180 00 Prague 8, Czech Republic
A R T I C L E I N F O
Article history: Received 30 November 2016 Received in revised form 21 December 2016 Accepted 29 December 2016 Available online xxx Keywords: Polyaniline Polypyrrole Poly(p-phenylenediamine) Conducting polymers Broad-band dielectric spectroscopy Effective medium approximation
A B S T R A C T
The materials composed of pairs of conducting polymers, polyaniline, polypyrrole and non-conducting poly(p-phenylenediamine), were prepared by the coating of one polymer with the other. The course of polymerizations and morphology of the resulting composites have been recorded and the products were characterized by FTIR and Raman spectroscopies, DC and broad-band AC conductivity and permittivity measurements. The interfacial interaction between conducting polymers does not introduce any new effects concerning the conductivity. On the other hand, using composites where the conducting polymer is embedded in non-conducting polymer matrix allows for the control of structure at nanoscale and for the design of materials with new conductivity properties. This is illustrated by coating the conducting polymers with the poly(p-phenylenediamine) which has the conductivity by eight orders of magnitude lower, which yields composites with an intermediate conductivity, by three orders of magnitude lower than that of the conducting polymers. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Polyaniline (PANI) and polypyrrole (PPy) are probably the most studied conducting polymers. They are closely related to each other and both are prepared by the chemical oxidation of respective monomers or by their electropolymerization. They were usually investigated separately, and only several comparative reports on both polymers have been published [1–7]. The copolymerization of aniline and pyrrole may be regarded as a route to combine both polymers [8–16]. The molecular structure of copolymers, however, is complex. Monomer reactivity ratios widely differ [10] and, consequently, the copolymers are chemically heterogeneous by definition [17], i.e. they are composed of chains differing in the participation of aniline and pyrrole constitutional units. The copolymerization leads to the reduction in the conductivity due to the decrease in chain conjugation and, consequently, most copolymers are non-conducting [8]. They may be useful, however, in applications that do not require high conductivity [11,12].
* Corresponding author. E-mail address: [email protected] (J. Stejskal). http://dx.doi.org/10.1016/j.synthmet.2016.12.029 0379-6779/© 2017 Elsevier B.V. All rights reserved.
There is probably a little interest in the simple blending of both polymers. The polymerization of one polymer, however, on the top of the other is a challenge. This approach is based on the principle that virtually any object immersed in the reaction mixture used for the preparation of conducting polymers becomes coated with a thin film of this polymer. If one conducting polymer is used as a substrate for the coating with another one, such approach affords good interfacial contact between both polymers, because the chains of the second polymer grow on the chains of the first one. The depositions of PANI onto PPy nanotubes [18–22], PPy microspheres [23] or electropolymerized PPy films [24] may serve as examples. PANI coating led to enhance the microwave absorption [23], or improved the barrier properties and thus enabled a better corrosion protection of copper [24]. The reverse process, the deposition of PPy on PANI nanofibres [25], PANI-coated steel wire [26], PANI-coated carbon fibers [27], PANI suspension [28], PANI-coated carbon nanotubes [29] or electropolymerized PANI films [30–33] has also been reported. The composites had improved properties in corrosion protection [33], when applied in supercapacitors [18,21,34], in analytical sciences [19,20,26], in monitoring temperature [32] or in sensing of alcohol [6]. PPy/PANI coaxial nanoarrays were used in flexible supercapacitors.
110
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
The combination of both conducting polymers has recently been tested in the coating of cotton textile, which has subsequently been used as electrodes in the monitoring of carnivorous plant response to mechanical stimulation [35]. Cotton was coated with one polymer, PANI or PPy, and then alternatively with the second. The sensing function was similar in single or double-coated textile, but the conductivity of the latter type was improved, due to the higher content of the conducting component. The combination of PANI with another conducting polymer, poly(3,4-dioxythiophene), has also been recently reported [36]. The composites had improved thermoelectric properties. The present study concentrates on classical globular forms of conducting polymers, i.e. on powders. Another polymer, poly(pphenylenediamine) (PPDA), has also been included in the present study. This polymer is closely related to PANI, but its conductivity is lower by about eight orders of magnitude [37]. One of the polymers was used as a template, which was coated by the second polymer. The former polymer can thus be regarded as filler and the latter as a matrix. All combinations of the three polymers have been prepared. The presence of PANI or PPDA accelerates the oxidation of aniline and may affect the morphology of resulting materials and, subsequently, their electrical properties [38]. The course of the oxidations has also been recorded. 2. Experimental part 2.1. Synthesis of conducting polymers PANI was prepared by the oxidation of 0.2 M aniline hydrochloride with 0.25 M ammonium peroxydisulfate in aqueous medium [39]. The precipitate was collected on filter, rinsed with 0.2 M hydrochloric acid, with acetone, and dried in air, then over silica gel at room temperature.
PPy was similarly prepared by the oxidation of 0.2 M pyrrole with 0.5 M iron(III) chloride, and the PPDA by the oxidation of 0.2 M p-phenylenediamine dihydrochloride with 0.25 M ammonium peroxydisulfate [37,40] in water, and the oxidation products were treated as above. 2.2. Coating of conducting polymers Individual polymers were used as templates for the coating with the same or another polymer. Freshly prepared 50 mL reaction mixture used for the preparation of PANI or PPDA or 100 mL that for PPy was poured over 1 g of template polymer. In this way, about 1 g of polymer was produced over 1 g of the template polymer. 2.3. Characterization The powders were compressed at 70 kN by a manual hydraulic press to the pellets of 13 mm in diameter and 1 mm thick. The room temperature conductivity was determined by a four-point method in the van der Pauw arrangement using a Keithley 220 Programmable Current Source, a Keithley 2010 Multimeter as a voltmeter, and a Keithley 705 Scanner equipped with a Keithley 7052 matrix card. A Thermo Nicolet NEXUS 870 FTIR Spectrometer with a DTGS TEC detector recorded Fourier-transform infrared (FTIR) spectra of the samples dispersed in potassium bromide pellets. A Renishaw InVia Reflex Raman microspectrometer was used to collect the spectra excited with a HeNe laser (633 nm) focused with a research-grade Leica DM LM microscope on the sample. A Peltiercooled CCD detector (576 384 pixels) registered the dispersed light which was analyzed with a spectrograph using a holographic grating of 1800 lines mm1.
Fig. 1. Temperature profile during the oxidation of (a) aniline, (b) p-phenylenediamine and (c) pyrrole in the absence (squares) and in the presence of respective polymers: PANI (circles), PPDA (up-triangles) and PPy (down-triangles).
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
The broad-band dielectric and conductivity spectroscopies have used three techniques. In the terahertz range (0.2–2 THz), the complex dielectric response was obtained by using the timedomain THz transmission spectrometer based on a Ti:sapphire femtosecond laser [41]. The generation of the THz pulses was performed by an interdigitated photoconducting GaAs switch. Electro-optic sampling with a plate of the (110) ZnTe crystal was used to detect the transmitted THz pulses. In the microwave range (200 MHz–20 GHz), the pellets without electrodes were characterized by an Agilent 85070 open-end coaxial probe with an Agilent E8364B vector network analyzer. In the low frequency range (102 Hz–1 MHz), the dielectric and conductivity spectra were obtained using the Novocontrol Alpha-AN High-performance frequency analyzer. Au electrodes were sputtered on the planeparallel sample surfaces and the contacts were provided by silver wires fixed by the silver paste. The applied AC electric field was 1 V/mm. 3. Results and discussion Conducting polymers, such as PANI or PPy, and from the formal point of view also PPDA despite its low conductivity, coat any substrate immersed in the reaction mixture used for their preparation. This has been established in detail [42] and used for the coating of various substrates [43,44]. Such strategy has been used for the coating of the conducting polymers with the same or another polymer. This means that the original template polymers were incorporated in the matrix of the polymer used for coating. As the template and matrix polymers may differ in the conductivity, and there may be interfacial interaction between both polymers, the resulting microstructured composite may have new properties.
111
In addition, it is known that PPDA or PANI accelerate the oxidation of aniline [38] and affect the morphology of produced polymer [45]. For that reason, the course of polymerizations and morphology were investigated in addition to the conductivity of composites. 3.1. The course of polymerization The oxidation of aniline is an exothermic reaction. Its course can thus be conveniently followed by monitoring the reaction temperature. An induction period, typical of aniline oxidation, is followed by the polymerization, when the temperature increases (Fig. 1a). The induction period disappears when any of conducting polymers is present in the reactions mixture. This effect has so far been described only for acceleration by PANI [38,46,47]. On the other hand, any substrate introduced in the reaction mixture removes the induction period [43]. The oxidation of PPDA starts immediately after mixing of reactants and is only marginally affected by the presence of conducting polymers (Fig. 1b). The oxidation of pyrrole proceeds easily but it is less exothermic, and the temperature increase is smaller compared to the oxidation of aniline (Fig. 1c). 3.2. Morphology The changes in morphology are illustrated for the PANI template (Fig. 2a). PANI alone is composed of fused extended globules. After the polymerization of aniline in the presence of PANI, no marked modification of morphology has been observed (Fig. 2b). It cannot be decided whether the original PANI particles were coated with newly created polymer or if such polymerization
Fig. 2. (a) Original PANI, and PANI coated with (b) PANI, (c) PPDA, and (d) PPy.
112
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
produced new PANI particles in addition to the template, although the former hypothesis is favoured. Also with PPDA coating, the morphology does no display any features of interest (Fig. 2c). On the other hand, the coating with PPy leads to globules (Fig. 2d) and the features of original PANI template are absent. 3.3. DC conductivity The DC conductivity of polyaniline and polypyrrole is of the units S cm1 (Table 1). The coating with PANI again with PANI does not result in any significant change in conductivity as expected, and the same applies also to the deposition of PPy on PPy. PPDA has conductivity by eight orders of magnitude lower and is often regarded as non-conducting. The coating with nonconducting PPDA was anticipated to reduce the conductivity of the composite significantly, because the conducting PANI moiety would be separated by insulating barriers. The decrease in conductivity, however, was only to the order of 103 S cm1. This means that PPDA barriers do not hinder the electron transfer between PANI areas completely and their appreciable percolation through PPDA appears. The same conclusion holds also for the coating of PPy with PPDA (Table 1). This will be more quantified later. When PANI or PPy is deposited on PPDA, the latter nonconducting polymer acts as a filler in the composite. The conductivity of the composite is thus controlled by the conductivity of PANI or PPy matrix. All above results do not suggest any specific interfacial phenomena that would influence the overall conductivity of composites. 3.4. Infrared spectra The structure of composites was further analyzed by spectroscopic methods. Powdered samples combining all three polymers were analyzed by FTIR spectroscopy in transmission mode. Under such conditions, both polymers are expected to be detectable in the spectra, but the contribution of the coating polymer should dominate over that of the template polymer. This is especially true for particles of micrometer size and larger, while for smaller objects the contribution of both components would be more balanced. When PANI was modified with PPDA, however, the spectral features of PANI are still prevailing in the spectrum, but some peaks of PPDA can also be distinguished. Contrary, after subsequent polymerization with PPy only the bands of PPy are observed in the spectrum of the resulting sample, as expected (Fig. 3a). When PPDA is modified with PANI or PPy, the spectra of composites correspond completely to the spectra of PANI or PPy, respectively (Fig. 3b). Finally, when PPy was modified with PANI, the spectrum of the resulting sample is close to the spectrum of PANI (Fig. 3c). After modification with PPDA, the features of both polymers, PPy and PPDA, can be distinguished in the spectrum of the final product.
Table 1 DC conductivity (in S cm1) of template polymers before and after coating. Template
Uncoated
Coated with polyaniline
Polyaniline Poly(p-phenylenediamine) Polypyrrole
2.18 5.0 109 4.24
3.26 2.60 3.16
PPDA
polypyrrole 3
7.3 10 4.6 108 9.5 103
7.25 2.88 2.95
Fig. 3. FTIR spectra of (a) PANI, (b) PPDA, and (c) PPy coated with the corresponding polymers. The spectra of template polymers are included for comparison.
3.5. Raman spectra The Raman spectra were recorded in the back-scattering geometry with 633 nm excitation laser line. In the contrast to FTIR spectroscopy in transmission mode, Raman spectroscopy reflects the molecular structure of particle surfaces rather than the composition of particles. This is also the technique that can prove the completeness of coating of one polymer with another one.
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
113
Fig. 5. Broad-band complex dielectric spectra of PPDA (shell) and their composites with PANI and PPy (core). Spectra of neat PANI from [48] are shown for comparison. Experimental data are depicted by the symbols and lines show the fits and modelling – see text. The PPy-PPDA composite was not modelled because the dielectric function of the neat PPy was not determined.
latter polymers. After modification of PPy with PANI or PPDA, the Raman spectra of the final products correspond to PANI or PPDA, respectively. Also in this case the coating of template is again confirmed.
Fig. 4. Raman spectra of (a) PANI, (b) PPDA and (c) PPy coated with the corresponding polymers. The spectra of template polymers are included for comparison.
When PANI was modified with PPDA, Raman spectrum (Fig. 4) is very close to the spectrum of PPDA with only small features corresponding to the spectrum of PANI. In case of combination with PPy we detect practically only the bands of PPy in the spectrum of the final product, as expected in the case of complete coating. When PPDA is coated with PANI or PPy, only the bands of the latter polymers are observed in the spectra of the resulting samples. This again illustrates a good coating of template with the
Fig. 6. Broad-band AC conductivity spectra of PPDA (shell) and their composites with PANI and PPy (core). Spectra of neat PANI from [48] are shown for comparison. Experimental data are depicted by the symbols and lines show the fits and modelling – see text. CS stands for coated spheres. The PPy-PPDA composite was not modelled because the AC conductivity of the neat PPy was not determined due to its high DC conductivity.
114
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
3.6. Dielectric and AC conductivity spectra In Figs. 5 and 6 we present our room-temperature broad-band (102–1012 Hz) complex dielectric and AC conductivity spectra, respectively. The spectra of neat conducting PANI are taken from earlier reports [48,49]. Neat PPy was not measured due to its high absorption and DC conductivity. For the discussion of PANI spectra, cf. the previous discussion [49]. Considering the non-conducting PPDA, the dominant feature is the low-frequency plateau in AC conductivity, which should reveal the DC conductivity and roughly corresponds to it (Table 1). Striking feature is the pronounced increase in permittivity on decreasing frequency at the lowfrequency end, which should be caused by a slightly increasing low-frequency part on increasing frequency in the AC conductivity spectra [48], which gives evidence for some mesoscopic inhomogeneity in the conductivity of PPDA, presumably interfacial layers between the sample and electrode (Maxwell-Wagner effect) [50]. For evaluating and discussing the spectra of the composites, we have modelled their dielectric response eef f and AC conductivity spectra by using several models based on the effective medium approximation (EMA). EMA assumes homogeneous components of the composite with ideal sharp boundaries among them and a homogeneous probing field inside the individual fractions. It requires small grains of the fractions, much smaller than the wavelengths of the probing electric field and its penetration length into the absorbing grains, the knowledge of the complex dielectric functions of neat fractions (ePANI , ePPDA ) and their concentrations (xPANI , xPPDA ). Of course, any interaction among the fractions is neglected. We have used three models for the discussion: (1) Coated-spheres (CS) model introduced by Hashin and Shtrikman [51] assumes that the whole sample volume is completely filled up in a fractal way with spherical core (PANI)– shell (PPDA) particles of arbitrary but fixed core–shell volume ratio:
eef f ¼ ePPDA þ
3xPANI ePPDA ðePANI ePPDA Þ 3ePPDA þ xPPDA ðePANI ePPDA Þ
ð1Þ
In this model there is no percolation of the core response through the shells and the depolarizing fields on the core–shell boundaries are very strong if the responses strongly differ. (2) Lichtenecker model [52]:
eaef f ¼ ð1 xPANI ÞeaPPDA þ xPANI eaPANI ;
1a1
ð2Þ
In this model the topology of the composite is strongly dependent on the exponent a [53]. For a = 1, it describes parallel-plate-capacitor geometry with no depolarizing field and complete percolation of both components. For a = 0, it represents the logarithmic law with a uniform distribution of all possible depolarization factors and percolation just at the threshold of both fractions independently of the concentration x. For a = 1, it corresponds to serial plate-capacitor geometry with maximum depolarizing field and no percolation of any fraction at all. For the positive a, both components are partially percolated whereas for the negative a neither of the fraction is percolated, in all cases independently of x. (3) In the Bruggeman model [54], spherical grains of both fractions are embedded into a matrix consisting of the averaged effective medium:
ePPDA eef f ePANI eef f þ xPANI ¼0 ePPDA þ 2eef f ePANI þ 2eef f
ð1 xPANI Þ
The complex dielectric responses of neat fractions for EMA modelling were obtained from the fits of experimental data of the complex dielectric functions of PPDA and conducting PANI [48] pellets in the whole measured frequency range. We used a sum of several phenomenological Cole–Cole relaxations (Rj), one damped harmonic oscillator (HO) and a classical Drude term to model the DC conductivity:
Dej DeHO v2HO þ 2 1aj vHO v2 þ ig HO v j 1 þ ðiv=vRj Þ
eðvÞ ¼ e1 þ S
s DC e0 vði vt Þ
where Dej is the dielectric strength of the j-th Cole–Cole relaxation (its contribution to the static relative permittivity), e1 is the highfrequency (infrared) contribution, which includes vibrational and electronic contributions to the dielectric function, v is the linear frequency, vRj is the mean relaxation frequency of the j-th relaxation and 0 aj 1 describes the degree of broadening of the relaxation in the loss spectra (aj = 0 corresponds to the Debye relaxation and aj approaching unity corresponds to the limiting frequency-independent loss spectra); DeHO , vHO and g HO denote the dielectric strength, frequency and damping of the THz oscillator, respectively; s DC is the DC conductivity, e0 is the permittivity of free space and t = 1/(2pG ) is the scattering (life) time of the free carriers and G the plasma damping. The real part of the AC conductivity s 0 ðvÞ (Fig. 6) was calculated from the imaginary part of the complex dielectric function e00 ðvÞ as
s 0 ðvÞ ¼ e0 ve00 ðvÞ
ð5Þ
The parameters used for the fits of PANI and PPDA spectra in Figs. 5 and 6 are listed in Table 2. The HO is needed for fitting the THz spectra, which are mainly of multi-phonon origin (Boson peak, see [49]), whereas the low-frequency Cole–Cole relaxation R3 is needed mainly to fit the increasing part of the low-frequency permittivity on decreasing frequency and its origin consists in the conductivity inhomogeneities, presumably at the sample–electrode interface [50]. It is clear that the coated-spheres model should be the most appropriate for the core–shell topology of investigated composites, but it cannot account for the effective DC conductivity by four orders of magnitude higher than that of the neat PPDA. This is a clear signature that the percolation through the PPDA shells is not fully negligible. By using the Lichtenecker model, the positive a close to zero (a = 0.053) can fit the low-frequency AC conductivity
Table 2 Parameters used for the fitting of PANI and PPDA spectra. No parameter was fixed during the fitting. For meaning of the symbols see Eq. (4) and the text. Model
Parameter
PANI
PPDA
Drude
G (Hz) sDC (S cm1)
1 1015 0.11
1 1015 1 108
Cole-Cole
De1 vR1 (Hz) a1 De2 vR2 (Hz) a2 De3 vR3 (Hz) a3
3.2 103 0.5 109 0.17
9.7 1 1012 0.92 62 480 0.50 13 106 1.5 103 0.02
Harmonic oscillator
DeHO vHO (Hz) g HO (Hz)
20 2.2 1012 9.9 1012
0.9 2.1 1012 1 1012
ð3Þ
This model shows a sharp percolation threshold for the template (PANI) fraction content xPANI = 1/3 with the non-negligible depolarizing field.
ð4Þ
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
plateau and loss spectra, but yields too high permittivity values. To estimate the percolated fraction of PANI, which should be responsible for the low-frequency conductivity values of the PPDA (shell)-PANI (core) composite, the best appears to use the Bruggeman model with the sharp percolation threshold. However, the PANI volume fraction xPANI of 50% is too high, yielding DC conductivity only by about one order of magnitude smaller than that of the neat PANI. Adjustment of the xPANI value to fit the lowfrequency conductivity of the composite yields xPANI = 0.3347, only 0.0014 above the percolation threshold, and the overall fit is the best among all the used models (Figs. 5 and 6). This can be used to estimate roughly the percolated volume fraction of PANI within the framework of the Bruggeman model to 0.1–0.2% only. Such a small value cannot influence the conclusions obtained from the FTIR and Raman spectra indicating that the coating of the PANI particles with PPDA appears to be complete. The good fit of broadband spectra of the PANI-PPDA composite and their FTIR and Raman spectra clearly indicate that any interaction between the components in the composites can be neglected. 4. Conclusions Conducting polyaniline, polypyrrole and non-conducting poly (p-phenylenediamine) have been coated with each other by in-situ polymerization. The template polymer thus acted as the filler (a core) in the composite and the deposited polymer as the matrix (a shell). FTIR and especially Raman spectroscopies indicate complete coating of the template polymer. DC conductivity of the composites is controlled by the DC conductivity of the matrix which was of the order of units S cm1 for the conducting polymer PANI and PPy matrix, but for the non-conducting PPDA coating it reduced the DC conductivity of conducting templates to the order of 103 S cm1. The modelling of the broad-band dielectric and AC conductivity spectra using various EMA models allowed us to estimate the volume fraction of the percolated conducting polymer through the PPDA shells to be only of the order of 0.1%. The composites of conducting polymers embedded into a nonconducting matrix could be of interest in the design of composites with some particular electrical properties. Acknowledgments The authors thank the Czech Science Foundation for financial support (16-02787S) and to P. Magdziarz for technical assistance with sample preparations. References [1] N.V. Blinova, J. Stejskal, M. Trchová, J. Prokeš, M. Omastová, Eur. Polym. J. 43 (2007) 2331–2341. [2] M. Omastová, K. Mosná9 cková, M. Trchová, E.N. Konyushenko, J. Stejskal, P. Fedorko, J. Prokeš, Synth. Met. 160 (2010) 701–707. [3] T.Y. Liu, L. Finn, M.H. Yu, H.Y. Wang, T. Zhai, X.H. Lu, Y.X. Tong, Y. Li, Nano Lett. 14 (2014) 2522–2527. [4] Y.L. Tao, J.C. Li, A.J. Xie, S.K. Li, P. Chen, L.P. Ni, Y.H. Shen, Chem. Commun. 50 (2014) 12757–12760. [5] M. Joulazadeh, A.H. Navarchian, M. Niroomand, Adv. Polym. Technol. 33 (2014) 21461 E1–E10. [6] M. Joulazadeh, A.H. Navarchian, IEEE Sensors J. 15 (2015) 1697–1704.
115
[7] Z. Stempien, T. Rybicki, E. Rybicki, M. Kozanecki, M.I. Szynkowska, Synth. Met. 202 (2015) 49–62. [8] J. Stejskal, M. Trchová, I.A. Ananieva, J. Jan9 ca, J. Prokeš, S. Fedorova, I. Sapurina, Synth. Met. 146 (2004) 29–36. [9] M. Bhaumik, R. McCrindle, A. Maity, Chem. Eng. J. 228 (2013) 506–515. [10] H.R. Wasnik, D.S. Kelkar, V.Y. Ganvir, J. Polym. Eng. 35 (2015) 99–103. [11] H. Asiabi, Y. Yamini, S. Seidi, A. Esrafili, F. Rezaei, J. Chromatogr. A 1397 (2015) 19–26. [12] S. Pramodini, P. Poornesh, Opt. Mater. 46 (2015) 186–194. [13] V. Chaudhury, A. Kaur, J. Ind. Eng. Chem. 26 (2015) 143–148. [14] S. Dhibar, P. Bhattacharya, G. Hatui, C.K. Das, J. Alloys Compd. 625 (2015) 64– 75. [15] B.L. Liang, Z.Y. Qin, T. Li, Z.J. Dou, F.X. Zeng, Y.M. Cai, M.F. Zhu, Z. Zhou, Electrochim. Acta 177 (2015) 335–342. [16] W. Wang, F.H. Zhu, Y.T. Dai, H. Zhang, J.H. Lei, Int. J. Electrochem. Sci. 11 (2016) 4000–4006. [17] J. Stejskal, Polym. Int. 54 (2005) 108–113. [18] H.G. Mi, X.G. Zhang, X.G. Ye, S.D. Yang, J. Power Sources 176 (2008) 403–409. [19] J. Wilson, S. Radhakrishnan, C. Sumathi, D. Dharuman, Sens. Actuat. B: Chem. 171–172 (2012) 216–222. [20] S. Radhakrishnan, C. Sumathi, V. Dharuman, J. Wilson, Anal. Methods 5 (2013) 1010–1015. [21] Z.L. Wang, X.J. He, S.H. Ye, Y.X. Tong, G.R. Li, ACS Appl. Mater. Interfaces 6 (2014) 642–647. nková, J. Kovárová, J. Kopecká, J. [22] J. Stejskal, I. Sapurina, M. Trchová, I. Šede Prokeš, Chem. Pap. 69 (2015) 1341–1349. [23] C.H. Tian, Y.C. Du, P. Xu, R. Qiang, Y. Wang, D. Ding, J.L. Xue, J. Ma, H.T. Zhao, X.J. Han, ACS Appl. Mater. Interfaces 7 (2015) 20090–20099. [24] T.J. Pan, Z.W. Zuo, T. Wang, J. Hu, Z.D. Chen, Y.J. Ren, J. Power Sources 302 (2016) 180–188. [25] B.L. Liang, Z.Y. Qin, J.Y. Zhao, Y. Zhang, Z. Zhou, Y.Q. Lu, J. Mater. Chem. A 2 (2014) 2129–2135. [26] S.S. Zhao, M.A. Wu, F.Q. Zhao, B.Z. Zeng, Talanta 117 (2013) 146–151. [27] R. Rajagopalan, J.O. Iroh, Electrochim. Acta 47 (2002) 1847–1855. [28] Y. Wang, R.G. Wang, S.C. Xu, Q. Liu, J.X. Wang, Desalin. Water Treat. 54 (2015) 3248–3256. [29] S.T. Hussain, F. Abbas, A. Kausar, M.R. Khan, High Perform. Polym. 25 (2012) 70–78. [30] Z.Q. Gao, J. Bobacka, A. Ivaska, J. Electroanal. Chem. 364 (1994) 127–133. [31] Z.Q. Gao, J. Bobacka, A. Ivaska, Synth. Met. 55–57 (1993) 1477–1482. [32] M. Karakişla, M. Saçak, J. Polym. Sci., Part A: Polym. Chem. 38 (2009) 51–59. [33] J.Q. Xu, Y.Q. Zhang, D.Q. Zhang, Y.M. Tang, H. Cang, Prog. Org. Coat. 88 (2015) 84–91. [34] W. Lei, P. He, S.S. Zhang, F.Q. Dong, Y.J. Ma, J. Power Sources 266 (2014) 347– 352. [35] V. Bajgar, M. Penhaker, L. Martinková, A. Pavlovi9 c, P. Bober, M. Trchová, J. Stejskal, Sensors 16 (2016) 498 1–12. [36] X.Y. Wang, C.Y. Liu, L. Miao, J. Gao, Y. Chen, J. Electron. Mater. 45 (2016) 1813– 1820. [37] J. Stejskal, Prog. Polym. Sci. 41 (2015) 1–31. [38] J. Stejskal, P. Kratochvíl, M. Špírková, Polymer 36 (1995) 4135–4140. [39] J. Stejskal, R.G. Gilbert, Pure Appl. Chem. 74 (2002) 857–867. [40] J. Trlica, P. Sáha, O. Quadrat, J. Stejskal, Physica A 283 (2000) 337–348. mec, F. Kadlec, C. Kadlec, Opt. Express 18 (2010) 15338–15348. [41] P. Kužel, H. Ne [42] J. Stejskal, I. Sapurina, J. Prokeš, J. Zemek, Synth. Met. 105 (1999) 195–202. [43] S. Fedorova, J. Stejskal, Langmuir 18 (2002) 5630–5632. [44] J. Stejskal, I. Sapurina, M. Trchová, Prog. Polym. Sci. 35 (2010) 1420–1481. [45] J. Stejskal, M. Špírková, A. Riede, M. Helmstedt, P. Mokreva, J. Prokeš, Polymer 40 (1999) 2487–2492. [46] H.D. Tran, I. Norris, J.M. D'Arcy, H. Tsang, Y. Wang, B.R. Mattes, R.B. Kaner, Macromolecules 41 (2008) 7405–7410. [47] Z.D. Zujovic, Y. Wang, G.A. Bowmaker, R.B. Kaner, Macromolecules 44 (2011) 2735–2742. [48] J. Petzelt, D. Nuzhnyy, V. Bovtun, M. Savinov, M. Kempa, I. Rychetský, Phys. Status Solidi A 210 (2013) 2259–2271. [49] D. Nuzhnyy, J. Petzelt, I. Rychetský, M. Trchová, J. Stejskal, Synth. Met. 209 (2015) 561–569. [50] J. Petzelt, D. Nuzhnyy, Phase Transit. 89 (2016) 651–666. [51] Z. Hashin, S. Shtrikman, J. Appl. Phys. 33 (1962) 3125–3131. [52] K. Lichtenecker, Phys. Z. 27 (1926) 115–158. [53] T. Zakri, J.P. Laurent, M. Vauclin, J. Phys. D: Appl. Phys. 31 (1998) 1589–1594. [54] D.A.G. Bruggeman, Ann. Phys. (Leipzig) 24 (1935) 665–679.
Contents lists available at ScienceDirect
Synthetic Metals journal homepage: www.elsevier.com/locate/synmet
Interfaced conducting polymers J. Stejskala,* , P. Bobera , M. Trchováa , D. Nuzhnyyb , V. Bovtunb , M. Savinovb , J. Petzeltb , J. Prokešc a b c
Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, 162 06 Prague 6, Czech Republic Institute of Physics, Academy of Sciences of the Czech Republic, 182 21 Prague 8, Czech Republic Charles University, Faculty of Mathematics and Physics, 180 00 Prague 8, Czech Republic
A R T I C L E I N F O
Article history: Received 30 November 2016 Received in revised form 21 December 2016 Accepted 29 December 2016 Available online xxx Keywords: Polyaniline Polypyrrole Poly(p-phenylenediamine) Conducting polymers Broad-band dielectric spectroscopy Effective medium approximation
A B S T R A C T
The materials composed of pairs of conducting polymers, polyaniline, polypyrrole and non-conducting poly(p-phenylenediamine), were prepared by the coating of one polymer with the other. The course of polymerizations and morphology of the resulting composites have been recorded and the products were characterized by FTIR and Raman spectroscopies, DC and broad-band AC conductivity and permittivity measurements. The interfacial interaction between conducting polymers does not introduce any new effects concerning the conductivity. On the other hand, using composites where the conducting polymer is embedded in non-conducting polymer matrix allows for the control of structure at nanoscale and for the design of materials with new conductivity properties. This is illustrated by coating the conducting polymers with the poly(p-phenylenediamine) which has the conductivity by eight orders of magnitude lower, which yields composites with an intermediate conductivity, by three orders of magnitude lower than that of the conducting polymers. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Polyaniline (PANI) and polypyrrole (PPy) are probably the most studied conducting polymers. They are closely related to each other and both are prepared by the chemical oxidation of respective monomers or by their electropolymerization. They were usually investigated separately, and only several comparative reports on both polymers have been published [1–7]. The copolymerization of aniline and pyrrole may be regarded as a route to combine both polymers [8–16]. The molecular structure of copolymers, however, is complex. Monomer reactivity ratios widely differ [10] and, consequently, the copolymers are chemically heterogeneous by definition [17], i.e. they are composed of chains differing in the participation of aniline and pyrrole constitutional units. The copolymerization leads to the reduction in the conductivity due to the decrease in chain conjugation and, consequently, most copolymers are non-conducting [8]. They may be useful, however, in applications that do not require high conductivity [11,12].
* Corresponding author. E-mail address: [email protected] (J. Stejskal). http://dx.doi.org/10.1016/j.synthmet.2016.12.029 0379-6779/© 2017 Elsevier B.V. All rights reserved.
There is probably a little interest in the simple blending of both polymers. The polymerization of one polymer, however, on the top of the other is a challenge. This approach is based on the principle that virtually any object immersed in the reaction mixture used for the preparation of conducting polymers becomes coated with a thin film of this polymer. If one conducting polymer is used as a substrate for the coating with another one, such approach affords good interfacial contact between both polymers, because the chains of the second polymer grow on the chains of the first one. The depositions of PANI onto PPy nanotubes [18–22], PPy microspheres [23] or electropolymerized PPy films [24] may serve as examples. PANI coating led to enhance the microwave absorption [23], or improved the barrier properties and thus enabled a better corrosion protection of copper [24]. The reverse process, the deposition of PPy on PANI nanofibres [25], PANI-coated steel wire [26], PANI-coated carbon fibers [27], PANI suspension [28], PANI-coated carbon nanotubes [29] or electropolymerized PANI films [30–33] has also been reported. The composites had improved properties in corrosion protection [33], when applied in supercapacitors [18,21,34], in analytical sciences [19,20,26], in monitoring temperature [32] or in sensing of alcohol [6]. PPy/PANI coaxial nanoarrays were used in flexible supercapacitors.
110
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
The combination of both conducting polymers has recently been tested in the coating of cotton textile, which has subsequently been used as electrodes in the monitoring of carnivorous plant response to mechanical stimulation [35]. Cotton was coated with one polymer, PANI or PPy, and then alternatively with the second. The sensing function was similar in single or double-coated textile, but the conductivity of the latter type was improved, due to the higher content of the conducting component. The combination of PANI with another conducting polymer, poly(3,4-dioxythiophene), has also been recently reported [36]. The composites had improved thermoelectric properties. The present study concentrates on classical globular forms of conducting polymers, i.e. on powders. Another polymer, poly(pphenylenediamine) (PPDA), has also been included in the present study. This polymer is closely related to PANI, but its conductivity is lower by about eight orders of magnitude [37]. One of the polymers was used as a template, which was coated by the second polymer. The former polymer can thus be regarded as filler and the latter as a matrix. All combinations of the three polymers have been prepared. The presence of PANI or PPDA accelerates the oxidation of aniline and may affect the morphology of resulting materials and, subsequently, their electrical properties [38]. The course of the oxidations has also been recorded. 2. Experimental part 2.1. Synthesis of conducting polymers PANI was prepared by the oxidation of 0.2 M aniline hydrochloride with 0.25 M ammonium peroxydisulfate in aqueous medium [39]. The precipitate was collected on filter, rinsed with 0.2 M hydrochloric acid, with acetone, and dried in air, then over silica gel at room temperature.
PPy was similarly prepared by the oxidation of 0.2 M pyrrole with 0.5 M iron(III) chloride, and the PPDA by the oxidation of 0.2 M p-phenylenediamine dihydrochloride with 0.25 M ammonium peroxydisulfate [37,40] in water, and the oxidation products were treated as above. 2.2. Coating of conducting polymers Individual polymers were used as templates for the coating with the same or another polymer. Freshly prepared 50 mL reaction mixture used for the preparation of PANI or PPDA or 100 mL that for PPy was poured over 1 g of template polymer. In this way, about 1 g of polymer was produced over 1 g of the template polymer. 2.3. Characterization The powders were compressed at 70 kN by a manual hydraulic press to the pellets of 13 mm in diameter and 1 mm thick. The room temperature conductivity was determined by a four-point method in the van der Pauw arrangement using a Keithley 220 Programmable Current Source, a Keithley 2010 Multimeter as a voltmeter, and a Keithley 705 Scanner equipped with a Keithley 7052 matrix card. A Thermo Nicolet NEXUS 870 FTIR Spectrometer with a DTGS TEC detector recorded Fourier-transform infrared (FTIR) spectra of the samples dispersed in potassium bromide pellets. A Renishaw InVia Reflex Raman microspectrometer was used to collect the spectra excited with a HeNe laser (633 nm) focused with a research-grade Leica DM LM microscope on the sample. A Peltiercooled CCD detector (576 384 pixels) registered the dispersed light which was analyzed with a spectrograph using a holographic grating of 1800 lines mm1.
Fig. 1. Temperature profile during the oxidation of (a) aniline, (b) p-phenylenediamine and (c) pyrrole in the absence (squares) and in the presence of respective polymers: PANI (circles), PPDA (up-triangles) and PPy (down-triangles).
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
The broad-band dielectric and conductivity spectroscopies have used three techniques. In the terahertz range (0.2–2 THz), the complex dielectric response was obtained by using the timedomain THz transmission spectrometer based on a Ti:sapphire femtosecond laser [41]. The generation of the THz pulses was performed by an interdigitated photoconducting GaAs switch. Electro-optic sampling with a plate of the (110) ZnTe crystal was used to detect the transmitted THz pulses. In the microwave range (200 MHz–20 GHz), the pellets without electrodes were characterized by an Agilent 85070 open-end coaxial probe with an Agilent E8364B vector network analyzer. In the low frequency range (102 Hz–1 MHz), the dielectric and conductivity spectra were obtained using the Novocontrol Alpha-AN High-performance frequency analyzer. Au electrodes were sputtered on the planeparallel sample surfaces and the contacts were provided by silver wires fixed by the silver paste. The applied AC electric field was 1 V/mm. 3. Results and discussion Conducting polymers, such as PANI or PPy, and from the formal point of view also PPDA despite its low conductivity, coat any substrate immersed in the reaction mixture used for their preparation. This has been established in detail [42] and used for the coating of various substrates [43,44]. Such strategy has been used for the coating of the conducting polymers with the same or another polymer. This means that the original template polymers were incorporated in the matrix of the polymer used for coating. As the template and matrix polymers may differ in the conductivity, and there may be interfacial interaction between both polymers, the resulting microstructured composite may have new properties.
111
In addition, it is known that PPDA or PANI accelerate the oxidation of aniline [38] and affect the morphology of produced polymer [45]. For that reason, the course of polymerizations and morphology were investigated in addition to the conductivity of composites. 3.1. The course of polymerization The oxidation of aniline is an exothermic reaction. Its course can thus be conveniently followed by monitoring the reaction temperature. An induction period, typical of aniline oxidation, is followed by the polymerization, when the temperature increases (Fig. 1a). The induction period disappears when any of conducting polymers is present in the reactions mixture. This effect has so far been described only for acceleration by PANI [38,46,47]. On the other hand, any substrate introduced in the reaction mixture removes the induction period [43]. The oxidation of PPDA starts immediately after mixing of reactants and is only marginally affected by the presence of conducting polymers (Fig. 1b). The oxidation of pyrrole proceeds easily but it is less exothermic, and the temperature increase is smaller compared to the oxidation of aniline (Fig. 1c). 3.2. Morphology The changes in morphology are illustrated for the PANI template (Fig. 2a). PANI alone is composed of fused extended globules. After the polymerization of aniline in the presence of PANI, no marked modification of morphology has been observed (Fig. 2b). It cannot be decided whether the original PANI particles were coated with newly created polymer or if such polymerization
Fig. 2. (a) Original PANI, and PANI coated with (b) PANI, (c) PPDA, and (d) PPy.
112
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
produced new PANI particles in addition to the template, although the former hypothesis is favoured. Also with PPDA coating, the morphology does no display any features of interest (Fig. 2c). On the other hand, the coating with PPy leads to globules (Fig. 2d) and the features of original PANI template are absent. 3.3. DC conductivity The DC conductivity of polyaniline and polypyrrole is of the units S cm1 (Table 1). The coating with PANI again with PANI does not result in any significant change in conductivity as expected, and the same applies also to the deposition of PPy on PPy. PPDA has conductivity by eight orders of magnitude lower and is often regarded as non-conducting. The coating with nonconducting PPDA was anticipated to reduce the conductivity of the composite significantly, because the conducting PANI moiety would be separated by insulating barriers. The decrease in conductivity, however, was only to the order of 103 S cm1. This means that PPDA barriers do not hinder the electron transfer between PANI areas completely and their appreciable percolation through PPDA appears. The same conclusion holds also for the coating of PPy with PPDA (Table 1). This will be more quantified later. When PANI or PPy is deposited on PPDA, the latter nonconducting polymer acts as a filler in the composite. The conductivity of the composite is thus controlled by the conductivity of PANI or PPy matrix. All above results do not suggest any specific interfacial phenomena that would influence the overall conductivity of composites. 3.4. Infrared spectra The structure of composites was further analyzed by spectroscopic methods. Powdered samples combining all three polymers were analyzed by FTIR spectroscopy in transmission mode. Under such conditions, both polymers are expected to be detectable in the spectra, but the contribution of the coating polymer should dominate over that of the template polymer. This is especially true for particles of micrometer size and larger, while for smaller objects the contribution of both components would be more balanced. When PANI was modified with PPDA, however, the spectral features of PANI are still prevailing in the spectrum, but some peaks of PPDA can also be distinguished. Contrary, after subsequent polymerization with PPy only the bands of PPy are observed in the spectrum of the resulting sample, as expected (Fig. 3a). When PPDA is modified with PANI or PPy, the spectra of composites correspond completely to the spectra of PANI or PPy, respectively (Fig. 3b). Finally, when PPy was modified with PANI, the spectrum of the resulting sample is close to the spectrum of PANI (Fig. 3c). After modification with PPDA, the features of both polymers, PPy and PPDA, can be distinguished in the spectrum of the final product.
Table 1 DC conductivity (in S cm1) of template polymers before and after coating. Template
Uncoated
Coated with polyaniline
Polyaniline Poly(p-phenylenediamine) Polypyrrole
2.18 5.0 109 4.24
3.26 2.60 3.16
PPDA
polypyrrole 3
7.3 10 4.6 108 9.5 103
7.25 2.88 2.95
Fig. 3. FTIR spectra of (a) PANI, (b) PPDA, and (c) PPy coated with the corresponding polymers. The spectra of template polymers are included for comparison.
3.5. Raman spectra The Raman spectra were recorded in the back-scattering geometry with 633 nm excitation laser line. In the contrast to FTIR spectroscopy in transmission mode, Raman spectroscopy reflects the molecular structure of particle surfaces rather than the composition of particles. This is also the technique that can prove the completeness of coating of one polymer with another one.
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
113
Fig. 5. Broad-band complex dielectric spectra of PPDA (shell) and their composites with PANI and PPy (core). Spectra of neat PANI from [48] are shown for comparison. Experimental data are depicted by the symbols and lines show the fits and modelling – see text. The PPy-PPDA composite was not modelled because the dielectric function of the neat PPy was not determined.
latter polymers. After modification of PPy with PANI or PPDA, the Raman spectra of the final products correspond to PANI or PPDA, respectively. Also in this case the coating of template is again confirmed.
Fig. 4. Raman spectra of (a) PANI, (b) PPDA and (c) PPy coated with the corresponding polymers. The spectra of template polymers are included for comparison.
When PANI was modified with PPDA, Raman spectrum (Fig. 4) is very close to the spectrum of PPDA with only small features corresponding to the spectrum of PANI. In case of combination with PPy we detect practically only the bands of PPy in the spectrum of the final product, as expected in the case of complete coating. When PPDA is coated with PANI or PPy, only the bands of the latter polymers are observed in the spectra of the resulting samples. This again illustrates a good coating of template with the
Fig. 6. Broad-band AC conductivity spectra of PPDA (shell) and their composites with PANI and PPy (core). Spectra of neat PANI from [48] are shown for comparison. Experimental data are depicted by the symbols and lines show the fits and modelling – see text. CS stands for coated spheres. The PPy-PPDA composite was not modelled because the AC conductivity of the neat PPy was not determined due to its high DC conductivity.
114
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
3.6. Dielectric and AC conductivity spectra In Figs. 5 and 6 we present our room-temperature broad-band (102–1012 Hz) complex dielectric and AC conductivity spectra, respectively. The spectra of neat conducting PANI are taken from earlier reports [48,49]. Neat PPy was not measured due to its high absorption and DC conductivity. For the discussion of PANI spectra, cf. the previous discussion [49]. Considering the non-conducting PPDA, the dominant feature is the low-frequency plateau in AC conductivity, which should reveal the DC conductivity and roughly corresponds to it (Table 1). Striking feature is the pronounced increase in permittivity on decreasing frequency at the lowfrequency end, which should be caused by a slightly increasing low-frequency part on increasing frequency in the AC conductivity spectra [48], which gives evidence for some mesoscopic inhomogeneity in the conductivity of PPDA, presumably interfacial layers between the sample and electrode (Maxwell-Wagner effect) [50]. For evaluating and discussing the spectra of the composites, we have modelled their dielectric response eef f and AC conductivity spectra by using several models based on the effective medium approximation (EMA). EMA assumes homogeneous components of the composite with ideal sharp boundaries among them and a homogeneous probing field inside the individual fractions. It requires small grains of the fractions, much smaller than the wavelengths of the probing electric field and its penetration length into the absorbing grains, the knowledge of the complex dielectric functions of neat fractions (ePANI , ePPDA ) and their concentrations (xPANI , xPPDA ). Of course, any interaction among the fractions is neglected. We have used three models for the discussion: (1) Coated-spheres (CS) model introduced by Hashin and Shtrikman [51] assumes that the whole sample volume is completely filled up in a fractal way with spherical core (PANI)– shell (PPDA) particles of arbitrary but fixed core–shell volume ratio:
eef f ¼ ePPDA þ
3xPANI ePPDA ðePANI ePPDA Þ 3ePPDA þ xPPDA ðePANI ePPDA Þ
ð1Þ
In this model there is no percolation of the core response through the shells and the depolarizing fields on the core–shell boundaries are very strong if the responses strongly differ. (2) Lichtenecker model [52]:
eaef f ¼ ð1 xPANI ÞeaPPDA þ xPANI eaPANI ;
1a1
ð2Þ
In this model the topology of the composite is strongly dependent on the exponent a [53]. For a = 1, it describes parallel-plate-capacitor geometry with no depolarizing field and complete percolation of both components. For a = 0, it represents the logarithmic law with a uniform distribution of all possible depolarization factors and percolation just at the threshold of both fractions independently of the concentration x. For a = 1, it corresponds to serial plate-capacitor geometry with maximum depolarizing field and no percolation of any fraction at all. For the positive a, both components are partially percolated whereas for the negative a neither of the fraction is percolated, in all cases independently of x. (3) In the Bruggeman model [54], spherical grains of both fractions are embedded into a matrix consisting of the averaged effective medium:
ePPDA eef f ePANI eef f þ xPANI ¼0 ePPDA þ 2eef f ePANI þ 2eef f
ð1 xPANI Þ
The complex dielectric responses of neat fractions for EMA modelling were obtained from the fits of experimental data of the complex dielectric functions of PPDA and conducting PANI [48] pellets in the whole measured frequency range. We used a sum of several phenomenological Cole–Cole relaxations (Rj), one damped harmonic oscillator (HO) and a classical Drude term to model the DC conductivity:
Dej DeHO v2HO þ 2 1aj vHO v2 þ ig HO v j 1 þ ðiv=vRj Þ
eðvÞ ¼ e1 þ S
s DC e0 vði vt Þ
where Dej is the dielectric strength of the j-th Cole–Cole relaxation (its contribution to the static relative permittivity), e1 is the highfrequency (infrared) contribution, which includes vibrational and electronic contributions to the dielectric function, v is the linear frequency, vRj is the mean relaxation frequency of the j-th relaxation and 0 aj 1 describes the degree of broadening of the relaxation in the loss spectra (aj = 0 corresponds to the Debye relaxation and aj approaching unity corresponds to the limiting frequency-independent loss spectra); DeHO , vHO and g HO denote the dielectric strength, frequency and damping of the THz oscillator, respectively; s DC is the DC conductivity, e0 is the permittivity of free space and t = 1/(2pG ) is the scattering (life) time of the free carriers and G the plasma damping. The real part of the AC conductivity s 0 ðvÞ (Fig. 6) was calculated from the imaginary part of the complex dielectric function e00 ðvÞ as
s 0 ðvÞ ¼ e0 ve00 ðvÞ
ð5Þ
The parameters used for the fits of PANI and PPDA spectra in Figs. 5 and 6 are listed in Table 2. The HO is needed for fitting the THz spectra, which are mainly of multi-phonon origin (Boson peak, see [49]), whereas the low-frequency Cole–Cole relaxation R3 is needed mainly to fit the increasing part of the low-frequency permittivity on decreasing frequency and its origin consists in the conductivity inhomogeneities, presumably at the sample–electrode interface [50]. It is clear that the coated-spheres model should be the most appropriate for the core–shell topology of investigated composites, but it cannot account for the effective DC conductivity by four orders of magnitude higher than that of the neat PPDA. This is a clear signature that the percolation through the PPDA shells is not fully negligible. By using the Lichtenecker model, the positive a close to zero (a = 0.053) can fit the low-frequency AC conductivity
Table 2 Parameters used for the fitting of PANI and PPDA spectra. No parameter was fixed during the fitting. For meaning of the symbols see Eq. (4) and the text. Model
Parameter
PANI
PPDA
Drude
G (Hz) sDC (S cm1)
1 1015 0.11
1 1015 1 108
Cole-Cole
De1 vR1 (Hz) a1 De2 vR2 (Hz) a2 De3 vR3 (Hz) a3
3.2 103 0.5 109 0.17
9.7 1 1012 0.92 62 480 0.50 13 106 1.5 103 0.02
Harmonic oscillator
DeHO vHO (Hz) g HO (Hz)
20 2.2 1012 9.9 1012
0.9 2.1 1012 1 1012
ð3Þ
This model shows a sharp percolation threshold for the template (PANI) fraction content xPANI = 1/3 with the non-negligible depolarizing field.
ð4Þ
J. Stejskal et al. / Synthetic Metals 224 (2017) 109–115
plateau and loss spectra, but yields too high permittivity values. To estimate the percolated fraction of PANI, which should be responsible for the low-frequency conductivity values of the PPDA (shell)-PANI (core) composite, the best appears to use the Bruggeman model with the sharp percolation threshold. However, the PANI volume fraction xPANI of 50% is too high, yielding DC conductivity only by about one order of magnitude smaller than that of the neat PANI. Adjustment of the xPANI value to fit the lowfrequency conductivity of the composite yields xPANI = 0.3347, only 0.0014 above the percolation threshold, and the overall fit is the best among all the used models (Figs. 5 and 6). This can be used to estimate roughly the percolated volume fraction of PANI within the framework of the Bruggeman model to 0.1–0.2% only. Such a small value cannot influence the conclusions obtained from the FTIR and Raman spectra indicating that the coating of the PANI particles with PPDA appears to be complete. The good fit of broadband spectra of the PANI-PPDA composite and their FTIR and Raman spectra clearly indicate that any interaction between the components in the composites can be neglected. 4. Conclusions Conducting polyaniline, polypyrrole and non-conducting poly (p-phenylenediamine) have been coated with each other by in-situ polymerization. The template polymer thus acted as the filler (a core) in the composite and the deposited polymer as the matrix (a shell). FTIR and especially Raman spectroscopies indicate complete coating of the template polymer. DC conductivity of the composites is controlled by the DC conductivity of the matrix which was of the order of units S cm1 for the conducting polymer PANI and PPy matrix, but for the non-conducting PPDA coating it reduced the DC conductivity of conducting templates to the order of 103 S cm1. The modelling of the broad-band dielectric and AC conductivity spectra using various EMA models allowed us to estimate the volume fraction of the percolated conducting polymer through the PPDA shells to be only of the order of 0.1%. The composites of conducting polymers embedded into a nonconducting matrix could be of interest in the design of composites with some particular electrical properties. Acknowledgments The authors thank the Czech Science Foundation for financial support (16-02787S) and to P. Magdziarz for technical assistance with sample preparations. References [1] N.V. Blinova, J. Stejskal, M. Trchová, J. Prokeš, M. Omastová, Eur. Polym. J. 43 (2007) 2331–2341. [2] M. Omastová, K. Mosná9 cková, M. Trchová, E.N. Konyushenko, J. Stejskal, P. Fedorko, J. Prokeš, Synth. Met. 160 (2010) 701–707. [3] T.Y. Liu, L. Finn, M.H. Yu, H.Y. Wang, T. Zhai, X.H. Lu, Y.X. Tong, Y. Li, Nano Lett. 14 (2014) 2522–2527. [4] Y.L. Tao, J.C. Li, A.J. Xie, S.K. Li, P. Chen, L.P. Ni, Y.H. Shen, Chem. Commun. 50 (2014) 12757–12760. [5] M. Joulazadeh, A.H. Navarchian, M. Niroomand, Adv. Polym. Technol. 33 (2014) 21461 E1–E10. [6] M. Joulazadeh, A.H. Navarchian, IEEE Sensors J. 15 (2015) 1697–1704.
115
[7] Z. Stempien, T. Rybicki, E. Rybicki, M. Kozanecki, M.I. Szynkowska, Synth. Met. 202 (2015) 49–62. [8] J. Stejskal, M. Trchová, I.A. Ananieva, J. Jan9 ca, J. Prokeš, S. Fedorova, I. Sapurina, Synth. Met. 146 (2004) 29–36. [9] M. Bhaumik, R. McCrindle, A. Maity, Chem. Eng. J. 228 (2013) 506–515. [10] H.R. Wasnik, D.S. Kelkar, V.Y. Ganvir, J. Polym. Eng. 35 (2015) 99–103. [11] H. Asiabi, Y. Yamini, S. Seidi, A. Esrafili, F. Rezaei, J. Chromatogr. A 1397 (2015) 19–26. [12] S. Pramodini, P. Poornesh, Opt. Mater. 46 (2015) 186–194. [13] V. Chaudhury, A. Kaur, J. Ind. Eng. Chem. 26 (2015) 143–148. [14] S. Dhibar, P. Bhattacharya, G. Hatui, C.K. Das, J. Alloys Compd. 625 (2015) 64– 75. [15] B.L. Liang, Z.Y. Qin, T. Li, Z.J. Dou, F.X. Zeng, Y.M. Cai, M.F. Zhu, Z. Zhou, Electrochim. Acta 177 (2015) 335–342. [16] W. Wang, F.H. Zhu, Y.T. Dai, H. Zhang, J.H. Lei, Int. J. Electrochem. Sci. 11 (2016) 4000–4006. [17] J. Stejskal, Polym. Int. 54 (2005) 108–113. [18] H.G. Mi, X.G. Zhang, X.G. Ye, S.D. Yang, J. Power Sources 176 (2008) 403–409. [19] J. Wilson, S. Radhakrishnan, C. Sumathi, D. Dharuman, Sens. Actuat. B: Chem. 171–172 (2012) 216–222. [20] S. Radhakrishnan, C. Sumathi, V. Dharuman, J. Wilson, Anal. Methods 5 (2013) 1010–1015. [21] Z.L. Wang, X.J. He, S.H. Ye, Y.X. Tong, G.R. Li, ACS Appl. Mater. Interfaces 6 (2014) 642–647. nková, J. Kovárová, J. Kopecká, J. [22] J. Stejskal, I. Sapurina, M. Trchová, I. Šede Prokeš, Chem. Pap. 69 (2015) 1341–1349. [23] C.H. Tian, Y.C. Du, P. Xu, R. Qiang, Y. Wang, D. Ding, J.L. Xue, J. Ma, H.T. Zhao, X.J. Han, ACS Appl. Mater. Interfaces 7 (2015) 20090–20099. [24] T.J. Pan, Z.W. Zuo, T. Wang, J. Hu, Z.D. Chen, Y.J. Ren, J. Power Sources 302 (2016) 180–188. [25] B.L. Liang, Z.Y. Qin, J.Y. Zhao, Y. Zhang, Z. Zhou, Y.Q. Lu, J. Mater. Chem. A 2 (2014) 2129–2135. [26] S.S. Zhao, M.A. Wu, F.Q. Zhao, B.Z. Zeng, Talanta 117 (2013) 146–151. [27] R. Rajagopalan, J.O. Iroh, Electrochim. Acta 47 (2002) 1847–1855. [28] Y. Wang, R.G. Wang, S.C. Xu, Q. Liu, J.X. Wang, Desalin. Water Treat. 54 (2015) 3248–3256. [29] S.T. Hussain, F. Abbas, A. Kausar, M.R. Khan, High Perform. Polym. 25 (2012) 70–78. [30] Z.Q. Gao, J. Bobacka, A. Ivaska, J. Electroanal. Chem. 364 (1994) 127–133. [31] Z.Q. Gao, J. Bobacka, A. Ivaska, Synth. Met. 55–57 (1993) 1477–1482. [32] M. Karakişla, M. Saçak, J. Polym. Sci., Part A: Polym. Chem. 38 (2009) 51–59. [33] J.Q. Xu, Y.Q. Zhang, D.Q. Zhang, Y.M. Tang, H. Cang, Prog. Org. Coat. 88 (2015) 84–91. [34] W. Lei, P. He, S.S. Zhang, F.Q. Dong, Y.J. Ma, J. Power Sources 266 (2014) 347– 352. [35] V. Bajgar, M. Penhaker, L. Martinková, A. Pavlovi9 c, P. Bober, M. Trchová, J. Stejskal, Sensors 16 (2016) 498 1–12. [36] X.Y. Wang, C.Y. Liu, L. Miao, J. Gao, Y. Chen, J. Electron. Mater. 45 (2016) 1813– 1820. [37] J. Stejskal, Prog. Polym. Sci. 41 (2015) 1–31. [38] J. Stejskal, P. Kratochvíl, M. Špírková, Polymer 36 (1995) 4135–4140. [39] J. Stejskal, R.G. Gilbert, Pure Appl. Chem. 74 (2002) 857–867. [40] J. Trlica, P. Sáha, O. Quadrat, J. Stejskal, Physica A 283 (2000) 337–348. mec, F. Kadlec, C. Kadlec, Opt. Express 18 (2010) 15338–15348. [41] P. Kužel, H. Ne [42] J. Stejskal, I. Sapurina, J. Prokeš, J. Zemek, Synth. Met. 105 (1999) 195–202. [43] S. Fedorova, J. Stejskal, Langmuir 18 (2002) 5630–5632. [44] J. Stejskal, I. Sapurina, M. Trchová, Prog. Polym. Sci. 35 (2010) 1420–1481. [45] J. Stejskal, M. Špírková, A. Riede, M. Helmstedt, P. Mokreva, J. Prokeš, Polymer 40 (1999) 2487–2492. [46] H.D. Tran, I. Norris, J.M. D'Arcy, H. Tsang, Y. Wang, B.R. Mattes, R.B. Kaner, Macromolecules 41 (2008) 7405–7410. [47] Z.D. Zujovic, Y. Wang, G.A. Bowmaker, R.B. Kaner, Macromolecules 44 (2011) 2735–2742. [48] J. Petzelt, D. Nuzhnyy, V. Bovtun, M. Savinov, M. Kempa, I. Rychetský, Phys. Status Solidi A 210 (2013) 2259–2271. [49] D. Nuzhnyy, J. Petzelt, I. Rychetský, M. Trchová, J. Stejskal, Synth. Met. 209 (2015) 561–569. [50] J. Petzelt, D. Nuzhnyy, Phase Transit. 89 (2016) 651–666. [51] Z. Hashin, S. Shtrikman, J. Appl. Phys. 33 (1962) 3125–3131. [52] K. Lichtenecker, Phys. Z. 27 (1926) 115–158. [53] T. Zakri, J.P. Laurent, M. Vauclin, J. Phys. D: Appl. Phys. 31 (1998) 1589–1594. [54] D.A.G. Bruggeman, Ann. Phys. (Leipzig) 24 (1935) 665–679.