Polypyrrole–ferric oxide conducting nanocomposites

Polypyrrole–ferric oxide conducting nanocomposites

European Polymer Journal 35 (1999) 1985±1992 Polypyrrole±ferric oxide conducting nanocomposites I. Synthesis and characterization Rupali Gangopadhyay...

169KB Sizes 3 Downloads 82 Views

European Polymer Journal 35 (1999) 1985±1992

Polypyrrole±ferric oxide conducting nanocomposites I. Synthesis and characterization Rupali Gangopadhyay, Amitabha De* Nuclear Chemistry Division, Saha Institute of Nuclear Physics, 1/AF, Bidhannagar, Calcutta 700064, India Received 29 July 1998; received in revised form 4 November 1998; accepted 30 November 1998

Abstract Conducting polymer nanocomposites have drawn the attention of scientists over last few years. The present work reports the preparation of nanocomposites in which colloidal ferric oxide particles have been combined with a wellknown conducting polymer polypyrrole (PPy). The colloid has been prepared using an established technique and its successful combination with PPy has been optimized in course of the present work. Particle dimensions were measured and the nature of association between the components was observed using transmission electron microscopic technique. The XRD and TGA patterns of PPy and the composites were also studied and explained. Direct current electrical conductivity values of all the samples have been measured; the e€ect of higher temperature and ambient condition on DC conductivity of the samples has also been studied and relevant results are analysed and reported. # 1999 Elsevier Science Ltd. All rights reserved. Keywords: Colloidal ferric oxide; Polypyrrole; Nanocomposite; DC conductivity

1. Introduction Research in the ®eld of conducting polymers aims mainly at some suitable modi®cation of existing polymers so that their applicability can be improved. Out of the di€erent modi®cation techniques available, the one most widely studied and applied in this respect is the formation of composites of di€erent origins. Conducting polymer composites are, in fact, some suitable composition of a conducting polymer with one or more insulating materials so that their desirable properties are combined successfully. Over the last few years, composites of a special category, termed `nanocomposites', have been studied with growing interest. These materials are especially important owing to their bridging role between the world of conducting poly-

* Corresponding author. Fax: +91-033-337-4637.

mers and that of nanoparticles. There are many examples describing the preparation and properties of these nanocomposites which are in fact some hybrid materials in which organic materials, namely polystyrene [1,2], poly (vinyl chloride) [3] etc., and inorganic oxides or salts of di€erent metals, namely CuO [4], SiO2 [5], ZrO2 [6], SnO2 [7], BaSO4 [8] etc., combine in some special fashion with the conducting polymers to give rise to these nanocomposites. In almost all the cases some speci®c nature of association between the two components has been observed. Insulating materials, rather than being simply blended or mixed up, are encapsulated or entrapped into the conducting polymer core, resulting in some signi®cant improvement in di€erent physical properties of the conducting polymers. Polypyrrole (PPy) is an important conducting polymer with high electrical conductivity and appreciable air stability. This polymer has been successfully utilized

0014-3057/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 1 4 - 3 0 5 7 ( 9 9 ) 0 0 0 0 8 - 7


R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992

in preparation of di€erent nanocomposites. In a series of recent publications Armes et al. have described the preparation and characterization of stable colloidal dispersions of PPy and polyaniline (PAn) using ultra®ne SiO2 and SnO2 as particulates in aqueous media; using TiO2, ZrO2, SbO2, YtO2 colloids, nanocomposite formation essentially occurs but macroscopic precipitation of black solid composite cannot be prevented [9±11]. Gan et al. have synthesized ultra®ne PAn± BaSO4 composite dispersion having composite particles of 10±20 nm via inverse microemulsion route [8]; particles are subjected to precipitation on centrifugation. Matijevic and coworkers have described the preparation of several polymer coated inorganic nanocomposites by utilizing the oxide core as colloidal oxidant. Using colloidal SiO2 and CuO, respectively, they have synthesized PPy- and PAn-coated nanocomposites with immeasurably low electrical conductivity [4,12]. In one of our previous communications preparation of PPy coated ZrO2 nanoparticles with high electrical conductivity (up to 15 S/cm) has been reported [6]. In situ preparation of iron oxide colloid and its encapsulation in PAn core has been successfully performed by Wan et al. [13]. In a recent publication Armes et al. have reported the preparation of functionalized nanoparticles; starting from pyrrole and pyrrole-3-acetic acid, they have prepared carboxylated PPy±SiO2 nanoparticles [14]. In addition to all these, incorporation of di€erent oxides, viz. TiO2, WO3, MnO2 etc. into electrochemically synthesized PPy ®lms has also been reported by Yoneyama and co-workers [15]. Insertion of PAn into V2O5 xerogel is described by Kanatzidis et al. [16] and electrochemical intercalation of Li into PAn/V2O5 nanocomposite has been performed by Leroux et al. [17]. Utilization of WO3 as the insulating material has also been reported by Yoneyama and Shiji [18]. In the above-mentioned articles authors have studied di€erent physical properties of the resultant nanocomposites. Particle size measurement as well as observation of the nature of association between two components have been made using TEM. Di€erent morphologies namely `raspberry' morphology, `stranded' morphology etc. are revealed from TEM pictures. However, almost everyone agrees that the inorganic particles are somehow `glued' or `bound' by the polymeric chain. DC electrical conductivity, thermogravimetric analysis (TGA) and other morphological studies (XPS, SEM) are also reported. In the present communication we report the easy chemical synthesis of a Fe2O3 encapsulated PPy nanocomposite. Being a well-studied colloid with prolonged stability and easy availability Fe2O3 was expected to be a suitable material for nanocomposite formation. At neutral medium it was found to be highly stable even on boiling; its particle size and particle content

were also found to be appropriate for encapsulation. Earlier Butterworth et al. [19,20] reported the preparation of conducting nanocomposites having some magnetic property using `magnetite' as the core material. However, as reported by the previous authors and supported by the XRD pattern, the colloidal ferric oxide used in this work is a-Fe2O3; the TEM studies were undertaken to determine the particle size as well as the nature of association between the two components. Thermal stability, crystallinity, electrical conductivity and environmental stability of the samples have also been studied in the course of work.

2. Experimental 2.1. Materials Pyrrole (AR) and anhydrous FeCl3 (AR) both were obtained from E. Merck, Germany. The monomer was distilled under nitrogen ¯ow and was kept in the dark prior to use. 2.2. Methods 2.2.1. Preparation of the colloid Aiming at the generation of Fe2O3 particles of a particular dimension, the colloid has been prepared using a standard procedure as described elsewhere [21]. 12 ml of a 32% ferric chloride solution were poured into 750 ml boiling water. Dark red coloured colloid readily formed, was dialysed until free from ions and stored. 2.2.2. Preparation of composites A ®xed volume (50 ml) of the colloid was taken from time to time and the volume was reduced to about 15 ml on evaporation. The selected volume of pyrrole was added to it with constant stirring. After some time, FeCl3 solution maintaining a pyrrole:FeCl3 mole ratio of 1:2.5 was added dropwise to it with continuous stirring. The resultant nanocomposite comes out of the solution and settles down as black solid which is ®ltered, washed thoroughly and vacuum dried at 508C. 2.3. Characterizations done 2.3.1. Particle content of colloid Ferric oxide content in the colloid was measured. 50 ml of colloid were taken in round bottom ¯ask and were frozen into solid. Afterwards the solvent was taken out and subjected to sublimation at ÿ358C using a lyophilizer. The dried particles were weighed until a constant weight was attained.

R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992


2.4. TEM studies Measurements of particle sizes of both the bare colloid and the composites and observation of the nature of association between the insulating and conducting components were performed using a Hitachi 600 Transmission Electron Microscope. 2.5. Elemental microanalysis Elemental microanalysis (C, H, N etc.) was carried out using Perkin-Elmer (USA) 2400 Series II CHN Analyser. From the percentage of the elements PPy content in the nanocomposites was determined. 2.6. Thermogravimetric analyses Thermogravimetric properties of the bare polypyrrole and the nanocomposites were studied in a Shimadzu DT 30 Thermal Analyser over a temperature range of 30±6008C. 2.7. X-Ray Di€raction Studies X-ray di€raction patterns of polypyrrole and the nanocomposites were taken using a Philips Di€ractometer (PW 1710). 2.8. DC electrical conductivity DC conductivity of pressed pellets of bare polymer and composites were measured over a temperature range from 25 up to 1258C using the van der Pauw technique [22] using a programmable DC voltage/current generator (Advantest. R6142) and a multimeter (Solatron, SI 7071).

3. Results and discussion 3.1. Preparation of composites In some of the previous communications, including one of ours, the amount of the insulating material was varied during the preparation of the composite, keeping the amount of the conducting polymer component ®xed. This technique could not be followed in this context because ferric oxide particles, once separated from the solution and dried up, could not be redispersed. Therefore, the colloid having ferric oxide content 1.8 g/ l was used directly. When pyrrole is polymerized in the presence of the colloid, and the nanocomposite forFig. 1. Transmission electron micrographs obtained from (a) Fe2O3 colloid, (b) nanocomposite P2, (c) nanocomposite P4.


R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992

Fig. 2. TGA curves obtained from (a) polypyrrole (P0), (b) nanocomposite P2, (c) nanocomposite-P4.

mation is complete, colloid particles get encapsulated in the polymer chain and the colour of the supernatant liquid changes to greenish yellow from brownish red, but the colour of the colloid is partially retained in the liquid when the composite formation or encapsulation of colloid particles is left incomplete. A signi®cant volume e€ect was shown by the system, a larger volume requires a larger volume of pyrrole. All these inconveniences were avoided using a ®xed volume of colloid and reducing the water content by evaporation. During the synthesis of nanocomposites, the volume of pyrrole was gradually lowered from 0.4 to 0.25 ml, below which the encapsulation remains incomplete. The pyrrole:FeCl3 mole ratio was kept constant at 1:2.5. This is an experimentally determined ratio reported elsewhere [23] relating to the optimization of polymerization of pyrrole with respect to the percentage of conversion and conductivity. Signi®cant lowering of the moles of FeCl3 either leaves the encapsulation incomplete or the product formation becomes too low to handle conveniently. This inconvenience guided the synthesis of only heavily doped samples. The volume fraction of conducting polymer in these composites lies above the percolation threshold. 3.2. TEM studies The apparent physical nature of PPy changes remarkably after composite formation; the initial ¯akes like texture changes to distinctly shaped granular form. This visual change is explained by an underlying phenomenon taking place during the composite formation as revealed and supported by the transmission electron micrographs of the bare colloid and the composites P2 and P4, shown in Fig. 1(a)±(c), respectively. The colloid particles, to a

®rst approximation, are polydisperse spheres lying in the size range from 25 to 50 nm. After composite formation, these particles (dark shaded) are found to be entrapped in PPy chain (light shaded). Therefore, the colloid particles are not simply mixed up or blended with the PPy; they are rather `glued' or `bound' by the PPy chains. In other words, as observed earlier by Armes et al., the colloid particles are acting here as the supports to the growing polymer chains. This observation is further supported by the fact that the encapsulated Fe2O3 particles are very dicult to dissolve even on heating with dilute acid. 3.3. Thermogravimetric analyses Themogravimetric patterns of polypyrrole and two of the nanocomposites are shown in Fig. 2. All the samples follow the similar degradation curves. Degradation starts at around 2008C and continues up to 400±4508C. As reported earlier [24], PPy shows a gradual degradation trend while the composites degrade relatively sharply. In composites, the amount of residue increases as the conducting fraction is lowered, which in e€ect raises the relative amount of thermally stable Fe2O3. That is why almost 56.5% residue is left with P4, while P2 leaves 51.4% and bare PPy (P0) leaves 49.3%. 3.4. X-Ray di€raction studies In some earlier reports polypyrrole has been described as an amorphous polymer [24]. However, Ouyang and Li [25] have recently reported two broad peaks centred around 7.18 and 22.48 in the XRD patterns of electrochemically synthesized PPy ®lms doped

R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992


Fig. 3. XRD patterns of (a) polypyrrole (P0), (b) nanocomposite P2.

with TsOÿ. Study of XRD patterns of present samples also reveals some degree of crystallinity, with appearance of one broad peak in the region of 2y=208±308, with a maximum around 24.68 for bare polymer as shown in Fig. 3(a). The peak may be assigned to the scattering from PPy chains at the interplanar spacing [25]. For the composites P2 and P4 identical XRD pattern is observed as in Fig. 3(b), where this peak becomes sharper with increased relative peak intensity and reduced amorphous scattering region. It implies that the composite samples have a more ordered arrangement than the bare polymer, which is also re¯ected in their electrical conductivity data. 3.5. DC electrical conductivity PPy and the composites with decreasing PPy content have been characterized with respect to their DC electrical conductivity. Composition of the samples, as obtained from elemental microanalysis, and their room temperature DC conductivities, are shown in Table 1. A number of distinct features are shown by the samples at room temperature. The most important and

interesting observation is that, despite the insertion of an insulating material, the DC conductivity of the composites is found to be signi®cantly higher than the bare polymer. Conductivity goes on increasing as we decrease the conducting fraction along the series of samples as shown in Fig. 4. This apparently anomalous behaviour, however, is in accordance with the previous experiments done in our laboratory using ZrO2± PPy nanocomposites. In the disordered systems like the conducting polymers, microscopic conductivity depends upon the doping level, conjugation length or chain length etc. which should not vary widely in the samples prepared in identical conditions. The macroscopic conductivity, however, depends on some external factors like compactness of the sample, orientation of the microparticles etc. In the present samples, pyrrole being polymerized in identical conditions, the intrinsic microscopic conductivities are more or less equal, but the physical properties, viz. compactness and molecular orientation, are signi®cantly varied depending upon the polypyrrole content in the samples. Pure PPy itself is a very lightweight polymer with poor compactness; here the microparticles are

Table 1 Compositions and room temperature DC conductivity values of PPy±Fe2O3 nanocomposites Sample speci®cation

Fe2O3 contenta (gm)

Volume of pyrrole (ml)

Polypyrrole (wt%)

DC conductivity (S/cm)

P0 P1 P2 P3 P4

0.00 0.09 0.09 0.09 0.09

± 0.40 0.35 0.30 0.25

100.00 73.00 64.46 62.41 61.06

23.6 33.1 58.9 79.6 85.3


Volume of colloid was kept ®xed at 50 ml.


R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992

Fig. 4. Variation in room temperature DC conductivity of nanocomposites with change in pyrrole (%) content.

very randomly oriented and the linking among the polymer particles through the grain boundaries is very poor which results in relatively lower conductivity. On composite formation, the growing polymer chains are supported on the Fe2O3 particles and thereby acquire distinct granular shape which leads to an improvement of the compactness of the composite material. This apparent physical change runs in parallel with the improvement in the internal ordering of the polymer planes as revealed by the XRD studies. As the PPy content in the samples is lowered, the change in compactness become more signi®cant; as a result, the weak links between the grains are increasingly improved and the coupling through the grain boundaries become stronger which ultimately results in the improvement in macroscopic conductivity measured in the pelletised form. Previously Armes et al. have reported a remarkable decrease in conductivity in such nanocomposites as compared to the bare polymer (PPy) using stringy SiO2 particles [26]. This decrease has been explained as a direct consequence of increase in resistive interparticle contacts within the material. In the present case a total encapsulation or coating of the insulating particles is observed and the fraction of the conducting part being far above that of the insulating part do not give rise to the above-mentioned resistive linkages. Therefore, the rise in DC conductivity in the nanocomposites from the pure PPY, although unnatural, is not unreasonable; it is attributed to the improved coupling of the grains of materials through the grain boundaries as well as to the continuous coating of the insulating intercalates with the conducting polymer. Few other publications, reporting the decrease in conductivity are also available [11]; from that point the improvement in conductivity in the present samples is obviously an achievement.

Fig. 5. Typical I±V plots shown by (a) polypyrrole (P0), (b) nanocomposite P2.

As discussed earlier, the present samples are all good quality semiconductors with very high conductivity. Thus I±V plots show very good linearity over a wide range of applied current, beyond which thermal ¯uctuations due to signi®cant Joule heating take place and nonlinearity, sets in. The typical I±V plots of PPy (P0) and one composite (P2) at room temperature are shown in Fig. 5. PPy shows linearity up to 40 mA, whereas the nanocomposite retains the trend up to 70 mA.

Fig. 6. Temperature dependence of DC electrical conductivity of polypyrrole and nanocomposites: (a) P0, (b) P1, (c) P2, (d) P3, (e) P4.

R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992


over a temperature range 25±1258C and the most important feature is shown by the samples at higher temperatures. The conductivity of all the systems goes on increasing as the temperature is raised; however, the nanocomposites show a remarkable increase compared to PPy. The conductivity of PPy increases steadily up to 758C, above which ¯uctuation starts and the ultimate value is lowered to some extent. The composite with the highest PPy loading (P1) also shows similar behaviour, although the increasing trend of conductivity is retained up to 1008C in this case. The conductivities of other three composites increase up to 1258C, as revealed in Fig. 6. Moreover, the conductivity of nanocomposite samples shows a good reversibility with temperature; that is, as the sample is gradually cooled down from 1258C, conductivity values approach those in the heating cycle and ultimately an almost closed loop is observed (b), but this reversibility is not observed with PPy itself (Fig. 7a). Similar thermal e€ects on the conductivity of PPy have been reported earlier [27] with BFÿ 4 -doped PPy samples (conductivity 0.1±1.0 S/cm) and Clÿ doped [28] PPy composites (conductivity 10ÿ4±10ÿ1 S/cm). In both reports the conductivity of the samples increased monotonically and showed the maxima at around 1108C. However, in these cases, the values in the cooling cycle are remarkably lower than those in the heating cycle. Therefore, it can be concluded that thermal stability of conductivity of PPy is improved on composite formation. 3.7. Ageing e€ect in ambient condition Fig. 7. (a) Thermal annealing e€ect on DC electrical conductivity of polypyrrole (*, heating cycle; w, cooling cycle). (b) Thermal annealing e€ect on DC electrical conductivity of nanocomposite P4 (*, heating cycle; w, cooling cycle).

3.6. Thermal annealing Variation of DC electrical conductivity of PPy and the composites at higher temperatures has been studied

In order to study the ageing e€ect on conductivity, pellets of the bare PPy and the nanocomposite samples were left under ambient conditions and their conductivities were remeasured at certain time intervals. The conductivity of bare PPy fell drastically with time, while the nanocomposites possessed better stability (Table 2). After 15 days PPy lost about 60% of its initial conductivity, while it was almost retained with the nanocomposites. The composite with the highest

Table 2 E€ect of ambient condition on room temperature conductivity of di€erent nanocomposites Sample speci®cations

P0 P1 P2 P3 P4 a

Conductivity at di€erent intervals (S/cm) 0 days

15 days

30 days

45 days

60 days

23.6 33.1 58.9 79.6 85.3

7.91 28.13 56.74 72.9 79.2

±a 11.6 48.5 41.6 55.8

±a ±a 36.5 32.8 42.5

±a ±a 22.5 25.6 29.5

Conductivity could not be measured.


R. Gangopadhyay, A. De / European Polymer Journal 35 (1999) 1985±1992

PPy content (P1) was much less stable, as compared to those with lower PPy contents which retain signi®cant conductivity even after 60 days exposure to ambient condition. Similar experiments with PPy ®lms [29] also revealed similar fall in conductivities. For bare PPy this decrease has been assigned to the dedoping e€ect as reported earlier by Scoch et al. [30]. However, under ambient conditions this decrease may be explained as a direct consequence of aerial oxidation which results in chain scission and reduction in conjugation length of the polymer. With nanocomposites the particles are well formed and pellets are very robust and compact. Thus the inner particles are hardly exposed to the atmosphere and therefore the oxidising e€ect is minimized and delayed. 4. Conclusion Fe2O3 colloid is a highly stable colloid that can be synthesized in a very simple chemical method and particle size can be controlled. This colloid carrying particles of a particular dimension has been successfully combined with PPy to produce a series of PPy±Fe2O3 nanocomposites, varying in conducting polymer fraction. The colloid particles are entrapped or encapsulated in the core of the growing polymer chain, resulting in the formation of an inorganic-organic hybrid material. The speci®c term `nanocomposite' rather than simple blend or composite is therefore applicable here. The underlying change that has taken place owing to this subtle modi®cation has revealed itself in the improvement in di€erent physical properties of PPy such as its compactness and mechanical property, morphology, DC electrical conductivity, thermal annealing and resistance towards ambient condition etc. The mechanism of electrical transport in these materials as revealed by their low temperature conductivity (DC and AC) and thermoelectric power data is being investigated thoroughly and will be reported in one of our forthcoming publications. However, the improvements made in di€erent physical properties of the present nanocomposites are expected to enhance the application potential of the polymer without hampering its chemical properties. Acknowledgements The authors are thankful to Mr Pulak K. Roy and

Mr Ajoy Chakroborty of Biophysics Division, SINP for helping in TEM studies. R. Gangopadhyay gratefully acknowledges the ®nancial support from CSIR.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

Lascelles SF, Armes SP. Adv Mater 1995;7:864. Lascelles SF, Armes SP. J Mater Chem 1997;7(8):1339. Banerjee P, Mandal BM. Synth Met 1995;74:257. Partch RE, Gangolli SG, Matijevic E, Cai W, Arajs S. J Colloid Interface Sci 1991;27:144. Maeda S, Armes SP. J Mater Chem 1994;4:935. Bhattacharya A, Ganguly KM, De A, Sarkar S. Mater Res Bull 1996;31(5):527. Maeda S, Armes SP. Synth Met 1995;69:499. Gan LM, Zhang LH, Chan HSO, Chew CH. Mater Chem Phys 1995;40:94. Gill M, Armes SP, Fairhurst D, Emmett SN, Idzorek G, Pigott T. Langmuir 1992;8:2178. DeArmitt C, Armes SP. Langmuir 1993;9:652. Maeda S, Armes SP. Chem Mater 1995;7:171. Huang CL, Partch RE, Matijevic E. J Colloid Interface Sci 1995;170:275. Wan M, Zhou W, Li J. Synth Met 1996;78:27. McCarthy GP, Armes SP, Greaves SJ, Watts JF. Langmuir 1997;13:3686. Kawai K, Mihara N, Kuwabata S, Yoneyama H. J Electrochem Soc 1990;137(6):1793. Kanatzidis MG, Wu CG, Marcy HO, Kannewurf CR. J Am Chem Soc 1989;111:4139. Leroux F, Koene BE, Nazar LF. J Electrochem Soc 1996;143(9):L181. Yoneyama H, Shoji Y. J Electrochem Soc 1990;137:3826. Butterworth MD, Armes SP, Simpson AW. J Chem Soc Chem Comm 1994;2129. Butterworth MD, Bell SA, Armes SP, Simpson AW. J Colloid Interface Sci 1996;183:91. Chatterjee S, Sarkar S, Bhattacharyya SN. J Photochem Photobiol A: Chem 1993;72:183. van der Pauw LJ. Philips Res Rep 1958;131. Dubitsky YA, Becturov EA, Zhubanov BA. Mater Chem Phys 1993;34:306. Wang HL, Fernandez JE. Macromolecules 1993;26:3336. Ouyang J, Li Y. Polymer 1997;38(15):3997. Flitton R, Johal J, Maeda S, Armes SP. J Colloid Interface Sci 1995;173:135. Turcu R, Neamtu C. Synth Met 1993;53:325. Omastova M, KosÆ ina S, Pionteck J, Janke A, Pavlinec J. Synth Met 1996;81:49. Bhattacharya A, De A, Das S. Polymer 1996;37(19):4375. Schoch Jr KF, Byers WA, Buckley LJ. Synth Met 1995;72:13.