polyaniline thermoreversible gel composites

polyaniline thermoreversible gel composites

Synthetic Metals 159 (2009) 1710–1716 Contents lists available at ScienceDirect Synthetic Metals journal homepage: www.elsevier.com/locate/synmet M...

1MB Sizes 0 Downloads 18 Views

Synthetic Metals 159 (2009) 1710–1716

Contents lists available at ScienceDirect

Synthetic Metals journal homepage: www.elsevier.com/locate/synmet

Multiwalled carbon nanotube/polyaniline thermoreversible gel composites Ashesh Garai, Arun K. Nandi ∗ Polymer Science Unit, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India

a r t i c l e

i n f o

Article history: Received 28 January 2009 Received in revised form 9 April 2009 Accepted 11 May 2009 Available online 6 June 2009 Keywords: Polyaniline Carbon nanotube Gel composite Photoluminescence Mechanical property and conductivity

a b s t r a c t Composites of multiwalled carbon nanotubes (MWCNTs) in a dinonylnaphthalene disulphonic acid (DNNDSA)-doped polyaniline (PANI) thermoreversible gel were prepared from a formic acid medium. A three-dimensional fibrillar network and a reversible first order phase transition characterize the systems as thermoreversible gels. Transmission electron micrographs indicate that the MWCNTs are well dispersed in the gel and PANI-DNNDSA wraps the MWCNT surface unevenly. ␲–␲, CH–␲ and acid–base interactions are evident from Fourier transform infrared spectroscopy. Thermal stability increases with increasing MWCNT content and the storage modulus of the composites increases dramatically. Photoluminescence increases significantly in the composites showing a red shift of the emission peak with increasing MWCNT content. The ␲ band-polaron band transitions show a red shift and the dc conductivity increases two orders of magnitude over that of the PANI-DNNDSA gel with the addition of MWCNTs. The current–voltage characteristic curves are Ohmic in nature and the current increases appreciably with increasing MWCNT concentration. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Composites of polymers and carbon nanotubes [1] (P/CNTs) have received considerable research interest because of dramatic improvement of mechanical, thermal, optical and conducting properties extending their applications in nanoscale devices, field emitters, bio-sensors etc. [2,3]. P/CNTs can be made with different polymeric matrices (e.g. insulating, ferroelectric and conducting) and the combined synergistic and conducting properties using multiwalled carbon nanotubes (MWCNTs) make them attractive for technological use [4–19]. Apart from the increase of conductivity, thermal stability, mechanical strength and new electronic/optoelectronic properties can be found in P/CNTs that use a conducting polymer [4–19]. Polyaniline (PANI) is an important polymer for its high conductivity [20], unique redox properties [21], ease of synthesis [22], and good environmental stability [23,24]. But it is difficult to process because of its infusibility and insolubility [25,26]. Presently, long chain sulphonic and phosphoric acids are used as both doping and processing agents for PANI by dispersion polymerization [27,28], emulsion polymerization [29], and by solution mixing [30]. A slightly different way of doping with long chain sulphonic acids, e.g. dinonylnaphthelene sulphonic acid, dinonylnaphthelene disulphonic acid (DNNDSA), camphor sulphonic acid etc. using the swelled PANI lattice in a formic acid medium produced thermore-

∗ Corresponding author. Tel.: +91 033 2473 4971x561; fax: +91 033 2473 2805. E-mail address: [email protected] (A.K. Nandi). 0379-6779/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.synthmet.2009.05.011

versible gels at high concentration of the acid (>50%, w/w) [31–36]. Gels, being intermediate of liquid and solid behavior, do not have adequate mechanical property required for its external use. The nanocomposites of gels are expected to have better mechanical properties and if they are produced from MWCNTs, additional conductivity and optoelectronic properties are expected to arise. Besides the property of reversible fusibility with temperature (i.e. thermoreversible) of the gel nanocomposite may find the composite to be easily processable. Here, we report the preparation and properties of composites of thermoreversible PANI-DNNDSA gel and MWCNTs. PANI/MWCNT composites are prepared by different in situ and ex situ methods. In the former methods aniline is polymerized in HClmediated MWCNT dispersion and in the latter the PANI (emeraldine base, EB) is mixed with MWCNT dispersion in a common solvent [4–19]. The PANI coated MWCNTs are mainly synthesized by in situ processes and yet no report exists to produce PANI coated MWCNTs in gel form. The fibrillar network structure of the gel may yield some new property than that in simple PANI/MWCNT composite and its thermoreversible nature might be useful for easier processing including inkjet printing. Here, we produce the composites of MWCNTs/PANI-DNNDSA gel in formic acid which breaks the PANI lattice and also oxidizes defect sites of MWCNTs producing –COOH groups on its surface. The mechanism of wrapping MWCNTs by PANI is evaluated from transmission electron microscope (TEM), wide-angle X-ray scattering (WAXS) and Fourier transform infrared (FTIR) spectroscopy and a model of the wrapping mechanism is proposed. The mechanical, thermal, optical and conductivity properties of this system are expected to be better than those of in

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

situ PANI composites due to its network structure and are reported here. 2. Experimental section 2.1. Synthesis of PANI PANI was prepared by polymerizing aniline in HCl medium with ammonium persulphate initiator at 0 ◦ C described elsewhere [22]. It was converted to EB form by digesting with NH4 OH solution. The intrinsic viscosity [] of PANI (EB) in H2 SO4 (97%) was 0.972 g/dl. The molecular weight of PANI was 15,500 calculated by the approximation of taking K and ˛ values of poly (p-phenylene terephthalamide) [K = 1.95 × 10−6 and ˛ = 1.36] in H2 SO4 medium [37,38]. 2.2. Dopant DNNDSA (55% concentrate in isobutanol) was gifted by King Industries, (Norwalk, CT, USA; trade name Nacure 155). 2.3. MWCNTs MWCNTs [Aldrich, U.S.A. product no. 636487-2G lot no. 03426TB] have outer diameter 20–30 nm, inner diameter ∼5–10 nm, length 0.5–200 ␮m and >95% purity as mentioned by the company. The MWCNTs, characterized by micro Raman spectroscopy using 632.8 nm laser, have D band at 1302 cm−1 and G bands at 1557 and 1588 cm−1 (suppl. Fig. 1). The doublet of the G band may arise for the splitting of degenerate E2g state or for the existence of a different energy state [39]. The spectrum clearly indicates that MWCNTs have a low degree of graphitization [15,40]. 2.4. Solvent Formic acid (synthesis grade, E. Merck, India) and H2 SO4 (97%) (E. Merck, India) were used as received. The N-methyl pyrrolidone (NMP) (G.R. Merck, India) used for spectrophotometric measurements was distilled before use. 2.5. Preparation of MWCNTs/PANI gel composites PANI (EB) and DNNDSA in the weight ratio (15: 85) (the weight of DNNDSA was counted from its 55% (w/w) concentration in isobutanol) and MWCNTs were mixed in formic acid medium [total concentration 2.5% (w/w)] in a round bottomed flask. The weights of MWCNTs were chosen for having MWCNT concentrations 0.5, 1 and 3% with respect to the weight of PANI-DNNDSA (w/w). The mixture was stirred at 65 ◦ C for 24 h and was dried on flat dishes at 60 ◦ C by a mild flow of air and finally at 60 ◦ C in vacuum for a week. The absence of 3114 and 3337 cm−1 peaks in the FTIR spectra of dried samples suggests the complete removal of formic acid and isobutanol [33,41]. 2.6. Morphology For TEM study, a drop of the mixture (24 h) was placed on a carbon coated copper grid, dried and was observed through a highresolution TEM (JEOL, 2010 EX) operated at an accelerated voltage of 200 kV. A charge coupled display (CCD) camera was used to take the pictures. The surface morphology of the dried gels was recorded in a field emission scanning electron microscope (FE-SEM) instrument (JEOL, JSM-6700F). The samples were observed at a voltage of 5 kV after platinum deposition.

1711

2.7. Structure The structure of the composites was determined using a WAXS instrument (Seifert X-ray diffractometer, model C-3000) in reflection mode with a parallel beam optics attachment. Nickel-filtered copper K˛ radiation ( = 0.154 nm) operating at a 35 kV voltage and a 30 mA current was used. The samples were scanned from 2 = 1.5–37◦ at the step scan mode (step size 0.03◦ , preset time 2 s) and diffraction pattern was recorded using a scintillation counter detector. 2.8. Spectral characterization The UV–vis spectra of the gel composites were scanned in NMP solution from 200 to 1100 nm in a UV–vis spectrophotometer (model 8453, Hewlett-Packard) using a quartz cell with 1 mm path length. The photoluminescence (PL) study of gels was done using FluoroMax-3 (HORIBA JOBIN YVON) luminescence spectrometer. The sample films of uniform thickness was made on spreading 2 ml formic acid solution (2.5%, w/w) over equal area glass slide and then drying at 60 ◦ C and finally in vacuum. The photo-excitation was made at an excitation wavelength of 226 nm at a 60◦ angle of the film plane. The emission was detected at a right angle to the excitation beam direction. The FTIR spectra of the samples were performed from the KBr pellets of the samples in FTIR-8400S instrument [SHIMADZU]. 2.9. Thermal measurement Differential scanning calorimetry (DSC) experiments were performed in a Perkin Elmer instrument (Diamond DSC-7) under N2 atmosphere. Samples in stainless steel large volume capsules were heated from −30 to 160 ◦ C at the heating rate of 10 ◦ C/min. Cooling runs were also made after waiting for 10 min at 160 ◦ C and then cooled at the rate 5 ◦ C/min to −30 ◦ C. It was then again heated at the rate of 10 ◦ C/min to 160 ◦ C. The instrument was calibrated with indium and cyclohexane before each set of experiment. The thermal stability of the G/CNTs was measured using a thermogravimetric analysis (TGA) instrument (TA, model SDTQ600) at a heating rate of 10 ◦ C/min in nitrogen atmosphere. 2.10. Dynamic mechanical property measurement The storage modulus (G ) and loss modulus (G ) of gel composites were measured using a dynamic mechanical analyzer (DMA) (TA instruments model Q-800). Films of gel composites were prepared in the dimension (25 mm × 5 mm × 0.15 mm) by pouring the formic acid solution on a die and evaporating at 60 ◦ C in vacuum for 7 days and were installed in the tension clamp of the calibrated instrument. The sample was heated from 0 to 120 ◦ C at the heat ing rate of 5 ◦ C/min. The G and G were measured at a constant frequency of 1 Hz with a static force of 0.02 N. 2.11. dc conductivity measurement Conductivity at 27 ◦ C of the samples were measured using the gel composite pellets (diameter = 1.3 cm) by the standard springloaded pressure contact four-probe method. A constant current (I) from a dc source electrometer (Keithley, model 617) was passed through two diagonal leads of the four probes and the voltage (V) across the other two leads was measured using a multimeter (Keithley, model 2000). The conductivity () was measured from Van der Paw relation [42]  = (ln2/␲d) (I/V), where d is the thickness of the pellet. Conductivity of two pellets with two trials was measured and the average of the four measurements was taken as conductivity of the sample. I–V characteristic curves of the gel

1712

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

Fig. 1. SEM pictures of (a) G/CNT0.5 and (b) its freeze-fractured surface.

composites were measured using gel composite films by two-probe method. 3. Results and discussion All the results reported here are for PANI-DNNDSA gel at 15: 85 (w/w) composition and it’s composites, designated as G/CNT0.5, G/CNT1 and G/CNT3, where G stands for the above PANI-DNNDSA gel and the number indicates the weight percentage of MWCNTs in the composites. 3.1. Morphology and DSC study Fibrillar network morphology of G/CNT0.5 is evident from the SEM micrograph (Fig. 1 (a)) and the same morphology is present in G, G/CNT1 and G/CNT3 (Suppl. Fig. 2). So the fibrillar network structure of G [33,34] is retained in the composites with MWCNTs. The SEM pictures of freeze-fractured surface of the composite (Fig. 1b and suppl. Fig. 2) indicate that MWCNTs are well dispersed in the gel composite. The PANI-DNNDSA network and dispersion of MWCNT are more distinctly seen in the TEM picture of G/CNT1 (Fig. 2a). The inset of Fig. 2a (five times enlarged marked portion) clearly indicates from the different darkness that PANI- DNNDSA wraps MWCNTs very well. The TEM pictures of suppl. Fig. 3 also corroborate the network structure and PANI wrapped MWCNTs in the composite. In G/CNT3 (Fig. 2b) MWCNTs separately exhibit a network structure due to its high concentration and the network is embedded in PANI-DNNDSA network. For a comparison with pure MWCNT morphology, its TEM picture is also presented in Fig. 2c and certainly the MWCNTs in the G/CNTs (Fig. 2a, inset and Fig. 2b) are well dispersed and well coated in the gel matrix. So these results infer that hierarchical structure (i.e. MWCNTs wrapped by PANIDNNDSA forming core-shell structure) is present in the gel composites, the MWCNT morphology (fibril or network) depends on it’s concentration in the G/CNTs. The average thickness of MWCNTs in the G/CNTs are 23.9 ± 3.4, 27.7 ± 5.5, 30.8 ± 8 and 31.8 ± 7.5 nm for MWCNTs, G/CNT0.5, G/CNT1 and G/CNT3 respectively, indicating wrapping of MWCNTs by PANI-DNNDSA to some extent [6,15]. The heating thermogram of G/CNT0.5 (Fig. 3) of the as prepared sample [thermogram (a)] shows a broad endothermic peak that can be deconvoluted into two endothermic peaks at 69 and 89 ◦ C. In the cooling curve [thermogram (b)] and on further heating [thermogram (c)] broad peak which can be deconvoluted into two exothermic peaks at 36 and 63 ◦ C and two endothermic peaks at 36 and 66 ◦ C, respectively are observed. The higher melting points in the first heating than that of second heating are due to annealing of the solvent cast samples at 60 ◦ C for seven days after its prepa-

ration. This reversible behavior is also found in PANI-DNNDSA gel and also in other G/CNTs (Suppl. Fig. 4). The enthalpy of fusion and the peak temperatures of the deconvoluted thermograms are presented in Table 1 and enthalpy values of both the peaks are almost equal in the first and second heating indicating reversible nature of the first order transition. Cooling enthalpies are somewhat low because of slow crystallization processes. Repeated experiments also yield similar results as presented in suppl.Fig. 5 and the similar DSC curves indicate that MWCNTs do not affect the crystallization and melting property of PANI-DNNDSA gel significantly. So the fibrillar network structure and reversible first order phase transition characterize the composites as thermoreversible gels [43] as in the PANI-DNNDSA systems [31–36]. The two exothermic or two endothermic peaks in the DSC thermograms may arise from lamellar structure of DNNDSA doped PANI [33–35]. The overlapping of the doped surfactant tails produce bilayers and monolayers capable of crystallization [33–35] and the two exotherms/endotherms in Fig. 3 are for monolayer and bilayer crystallization and melting, respectively. The higher melting point may be assigned for the melting of densely packed bilayer crystal of nonyl tails and the lower melting point may be that of the monolayer crystals of naphthyl heads [33–35]. The monolayer and bilayer crystals of anchored DNNDSA produce lamellar PANI structure, which supramolecularly organizes into fibrillar network responsible for thermoreversible gel formation. The monolayer and bilayer surfactant crystals are reversibly fusible and act as junction points of the gel. In G/CNTs the MWCNTs are at the core and the PANI-DNNDSA lamella/single DNNDSA doped PANI chain may constitute the shell. 3.2. 3.2.X-ray diffraction The X-ray diffractograms of MWCNTs, PANI-DNNDSA and the G/CNTs (Suppl. Fig. 5) show that there are two peaks at lower angles (2 = 3.2 and 5.4◦ ) which may be attributed to the lamellar thickness at 2.76 nm and to the thickness of pendent DNNDSA chain at 1.62 nm. So, the diffraction data supports the presence of same lamellar structure in gel composites as that in PANI-DNNDSA gel [33–35]. The d spacings and the intensity ratios (I/I0 ) of the various X-ray peaks are presented in Supplementary Table 1. The d spacings are new from those of PANI(EB) and the new low dhkl values may arise form the crystallization of pendent DNNDSA tails anchored from the PANI chains [33–35]. These X-ray results approximately tally with the energy minimized model (Fig. 4), drawn from molecular mechanics (MMX) program [47]. The I/I0 values of G/CNTs are somewhat different than those of PANI-DNNDSA gel, indicating atomic co-ordinates of unit cell is somewhat disturbed in the G/CNTs.

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

1713

Fig. 2. TEM micrographs of (a) G/CNT1 and (b) G/CNT3 (C) MWCNT.

3.3. FTIR spectra The FTIR spectra (Fig. 5) indicate that the PANI (EB) has two main peaks at 1498 and 1591 cm−1 for the stretching deformation of

benzonoid and quinonoid rings, respectively [44–47]. In the PANIDNNDSA gel 1591 peak shifts to 1579 cm−1 indicating the quinonoid electrons become delocalized due to doping by DNNDSA, facilitating the vibration of quinonoid ring at lower energy. This peak shifts to higher frequency in the G/CNTs. Probably the ␲–␲ interaction between graphene ring of MWCNTs and quinonoid ring of PANI hinder the quinonoid deformation. The vibration appears at 1607, 1595 and 1581 cm−1 for G/CNT0.5, G/CNT1 and G/CNT3, respectively. At higher MWCNT concentration its self-aggregation reduces the ␲–␲ interaction with PANI-DNNDSA causing easier vibration of Table 1 Peak temperatures and enthalpy values of PANI-DNNDSA gel and the G/CNTs obtained from DSC thermogramms. Sample

1st heating

Cooling

Peak

Peak



Fig. 3. DSC thermograms of G/CNT0.5 (a) first heating of the as prepared gel (heating rate 10 ◦ C/min), (b) cooling from 160 ◦ C at the cooling rate 5 ◦ C/min up to −20 ◦ C and (c) second heating of the gel prepared at −20 ◦ C for 15 min at the scan rate 10 ◦ C/min.

Enthalpy



2nd heating Enthalpy

Peak ◦

Enthalpy

( C)

J/gm

( C)

J/gm

( C)

J/gm

PANI-DNNDSA

69 87

4.0 6.6

36 63

1.9 2.7

36 66

6.7 8.0

G/CNT0.5

69 89

3.6 5.1

36 63

1.0 2.0

36 66

3.2 3.8

G/CNT3

70 89

4.2 5.6

36 65

1.0 1.9

36 67

5.7 5.4

1714

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

Fig. 4. A schematic model showing wrapping of MWCNTs by PANI-DNNDSA gel through ␲–␲, CH–␲ and acid–base interactions. (The PANI lamella and DNNDSA doped PANI are taken from energy minimization from molecular mechanics (MMX) program [47].

quinonoid ring. The benzenoid ring vibration at 1498 cm−1 though affected on doping (shifts to 1469 cm−1 ), remains unaffected on addition of MWCNTs. A new peak at 1727 cm−1 of carbonyl group vibration appears in the G/CNTs (not present in pure MWCNTs), indicating that during gelation at 65 ◦ C in formic acid, functionalization of MWCNTs by –COOH group occurs. The increasing –OH group absorbance intensity with MWCNT concentration indicates presence of –COOH group on the MWCNT surface. Also the –OH group peak position decreases by 5 cm−1 indicating H-bonding between –COOH group and nitrogen atom of PANI. PANI (EB) has peak at 1163 cm−1 for N=Q=N vibration which is partially electronic in nature [45]. In the PANI-DNNDSA gel due to doping the 1163 cm−1 peak shifts to lower energy (1126 cm−1 ) for polaron formation. This band position, however, increase (1134 cm−1 ) with addition of MWCNTs and it is due to the ␲–␲ interaction of quinonoid ring

of (B–N+ H = Q) with graphene ring of MWCNTs. Here also increasing MWCNT concentration shifts the peak to lower energy for selfaggregation of MWCNTs. The peaks around 2852–2958 cm−1 are due to the stretching frequency of CHn of PANI-DNNDSA and also that for MWCNTs [48]. Their intensity increases with MWCNT concentration in the samples and the 2864 peak shifts to 2857 cm−1 . The red shift is probably due to the CH–␲ interaction between nonyl tails of PANI- DNNDSA and graphene ring of MWCNTs. So from the FTIR study it may be inferred that there are three types of interactions (␲–␲, CH–␲ and acid–base) between MWCNTs and PANI-DNNDSA. An approximate model considering all the above interactions between PANI-DNNDSA and MWCNTs is illustrated in Fig. 4. The PANI- DNNDSA lamella structure is drawn from energy minimization by MMX program [47]. PANI-DNNDSA lamella length is 2.8 nm (cf. X-ray results). From the average thickness of MWCNTs the diameter of MWCNTs is calculated to be 17.0 nm. Absorption of a lamella on the MWCNT surface would yield a diameter of 22.4 nm yielding a thickness value of 31.6 nm. Such thickness values are evident in the G/CNT1 (30.8 ± 8) and G/CNT3 (31.8 ± 7.5 nm) samples. If a single PANI chain doped with DNNDSA molecules is absorbed, an average diameter of 20.1 nm corresponding to 28.5 nm thickness should be observed and the average value of G/CNT0.5 (27.7 ± 5.5) is very close to that. It is apparent from the thickness data that the standard deviation values are large and this is because PANI chain doped with DNNDSA and PANI-DNNDSA lamella are randomly absorbed on MWCNT surfaces. Possibility of some unabsorbed vacant space on the MWCNT surface cannot be ruled out. So an uneven surface of wrapped MWCNTs would result, the thickness at each point being governed by the wrapping component, PANI lamella or doped PANI chain. 3.4. 3.4.Thermogravimetric analysis

Fig. 5. FTIR spectra of (a) PANI (EB), (b) MWCNTs, (c) PANI-DNNDSA, (d) G/CNT0.5, (e) G/CNT1 and (f) G/CNT3.

The thermal decomposition of PANI-DNNDSA gel is one-step process (200 ◦ C, Suppl. Fig. 6) and those of G/CNTs are two-step pro-

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

1715

cesses. The first degradation temperatures are 206, 207 and 220 ◦ C for, G/CNT0.5, G/CNT1 and G/CNT3, respectively and the second degradation temperatures are 441, 445 and 450 ◦ C for G/CNT0.5, G/CNT1 and G/CNT3, respectively. The first step is for the decomposition of DNNDSA molecule whereas the second step is for the decomposition of polymer chain [49]. Within the accuracy of the instrument (±2 ◦ C) it may be inferred that in the G/CNTs PANI becomes more thermally stable for intermolecular interaction between the components [49] and MWCNTs being good conductor of heat [50], cause dissipation of heat to the surroundings easily. Thus, thermal stability of the gel increases with increasing MWCNT concentration and it is also supported from the increased residual weight of TGA curves. 3.5. Dynamic mechanical properties The G/CNTs yield free standing films except G/CNT3 which is brittle probably for the rigidity of MWCNT network structure (Fig. 2b) and also for higher intermolecular interactions hindering segmental motion of PANI [51]. The storage modulus (G ) of G/CNTs decreases with increase in temperature from 0 to 80 ◦ C [suppl. Fig. 7(a)] and is greater than that of pure gel. The highest increase is found in G/CNT0.5 sample where 72% increase of G is observed at 0 ◦ C but at 40 ◦ C it shows the highest increase (141%) (Suppl Table2). The increased molecular mobility with rise in temperature is the probable cause of increasing synergic effect. The higher reinforcing effect of G/CNT0.5 than that of G/CNT1 may be due to better dispersion of MWCNTs. The reinforcing effect of filler arises from its large surface area which decreases in its self-aggregated state at higher MWCNT concentration. The loss modulus (G ) decreases with rise in temperature (suppl. Fig. 7b) for each system and its variation with MWCNT concentration is almost the same with that of the G . 3.6. UV–vis spectra PANI (EB) has peaks at 324 and 635 nm for ␲–␲* transition of benzonoid ring and for electronic excitation from benzonoid ring to quinonoid ring, respectively [52–55] (suppl. Fig. 8). The 635 nm peak is absent in the gel and in the G/CNTs but a new peak at ≥840 nm appears, indicating doping is complete in all the PANI-DNNDSA systems. The 840 nm peak corresponds to the ␲ band to localized polaron band transition. Another new peak at ≤444 nm corresponding to the polaron band to ␲* band transition also appears in the doped systems. The absence of any free carrier tail indicates absence of delocalized polaron [52–55]. The shift of the 840 nm peak to 890 nm indicates that the band gap decreases with addition of MWCNTs. Probably the interaction between PANI-DNNDSA system with MWCNTs stabilize the polaron of PANI chain through resonance with the graphene ring of MWCNTs. Due to the above stabilization ␲ band-polaron band and polaron band to ␲* band transitions show red and blue shifts, respectively.

Fig. 6. Photoluminescence spectra of PANI-DNNDSA gel and G/CNTs after excitation by radiation of 226 nm wavelength at 30 ◦ C.

3.8. Conductivity The dc conductivity values at 27 ◦ C are 0.76 × 10−2 , 0.83 × 10−1 , 1.05 × 10−1 and 5.7 × 10−1 S/cm for PANI-DNNDSA, G/CNT0.5, G/CNT1 and G/CNT3 respectively. So, the dc conductivity of the G/CNTs increases by one order for G/CNT0.5 and two orders for G/CNT3. A charge exchange phenomenon is necessary for the conductivity of PANI even when it is 50% doped because of the structural defects [51]. This may involve interchain or intrachain proton exchange as well as electron transport. The interchain proton exchange occurs easily with addition of MWCNTs and hence conductivity increases. MWCNTs behave as conducting bridges connecting PANI conducting domains. Hence the band gaps are lower resulting in the increase of conductivity in the G/CNTs (cf. UV–vis spectra) The higher conductivity increase of G/CNT3 than those of the others is due to the network formation of MWCNTs contributing an additional intertube conductivity through the MWCNT network junctions. The I–V characteristic curves of PANI-DNNDSA and the G/CNTs (Fig. 7) illustrate that the conductivity of PANI-DNNDSA gel is Ohmic in nature and a high current (mA order) is due to the inter and intra chain contribution of network structure. Current increases on increasing MWCNT concentration due to increasing contacts between PANI and MWCNT. The sharp increase of current with

3.7. Photoluminescence (PL) spectra PANI (EB) does not have any PL property but PANI-DNNDSA gel exhibits PL emission at 356 nm after excitation by 226 nm radiation (absorbance of naphthalene moiety) (Fig. 6) [47]. In the G/CNTs the emission peak is red shifted and significant enhancement of PL emission takes place. The red shift, which increases with increasing MWCNT concentration, may be due to the resonance stabilization of excitons on MWCNT surface. The stabilization of excitons on the MWCNT surface causes the increase in the PL emission.

Fig. 7. I–V characteristic curves of PANI-DNNDSA gel and G/CNTs at 30 ◦ C.

1716

A. Garai, A.K. Nandi / Synthetic Metals 159 (2009) 1710–1716

voltage in G/CNT3 is due to the network structure of MWCNTs contributing additional intertube transfer of charges. 4. Conclusions PANI-DNNDSA system has the characteristics of thermoreversible gel when prepared from formic acid medium. The gel structure remains intact on addition of MWCNTs producing gel composite. TEM and FTIR study indicates uneven wrapping of MWCNT by PANI-DNNDSA lamella or by doped PANI chain from ␲–␲, CH–␲ and acid–base interactions. The thermal stability of the polymer increases with increasing MWCNT concentration into the gel. The G of G/CNTs has increased dramatically (maximum 141%) by the addition of 0.5% (w/w) MWCNTs at 40 ◦ C. The ␲-band to polaron band transition peak shows red shift, so dc conductivity increases in the G/CNTs. PL enhancement occur more in the G/CNT3 than that in the other G/CNTs. Compared to other PANI/MWCNT composites these G/CNTs are easily processible due to its thermoreversible nature. So an easily processible, thermally stable, high modulus and highly conducting PANI/MWCNT composites with interesting PL property, has been successfully developed by ex situ method suitable for sensor application and inkjet printing. Supporting information SEM, TEM, TGA, DMA & DSC thermogramms, X-ray patterns, Uvvis & Raman spectra, crystal spacing and storage modulus data of PANI-DNNDSA gel and G/CNTs. Acknowledgments We gratefully acknowledge council of scientific and industrial research (Grant No-01- (2224)/08/EMR-II) and Department of Science and Technology, New Delhi for financial support. The help extended by King Industries, Norwalk, CT, USA for gift sample of DNNDSA is also gratefully acknowledged. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.synthmet.2009.05.011. References [1] P.M. Ajayan, O. Stephan, C. Colliex, D. Trauth, Science 265 (1994) 1212. [2] (a) C.N.R. Rao, A.K. Cheetham, J. Mater. Chem. 11 (2001) 2887; (b) S.R.C. Vivekchand, L. Sudheendra, M. Sandeep, A. Govindaraj, C.N.R. Rao, J. Nanosci. Nanotechnol. 2 (2002) 631–635. [3] M. Moniruzzaman, K.I. Winey, Macromolecules 39 (2006) 5194. [4] (a) A.B. Dalton, S. Collins, E. Munoz, J.M. Razal, V.H. Ebron, J.P. Ferraris, J.N. Coleman, B.G. Kim, R.H. Baughman, Nature 423 (2003) 703; (b) M. Cadek, J.N. Coleman, V. Barron, K. Hedicke, W.J. Blau, Appl. Phys. Lett. 80 (2002) 2767. [5] M.J. Biercuk, M.C. Llaguno, M. Radosavljevic, J.K. Hyum, J.E. Fischer, A.T. Johnson, Appl. Phys. Lett. 80 (2002) 2767. [6] S. Manna, A.K. Nandi, J. Phys. Chem. C 111 (2007) 14670. [7] (a) C. Downs, J. Nugent, P.M. Ajayan, D.J. Duquette, K.S.V. Santhanam, Adv. Mater. 11 (1999) 1028; (b) G.B. Blanchet, C.R. Fincher, F. Gao, Appl. Phys. Lett. 82 (2003) 1290; (c) S.A. Curran, P.M. Ajayan, W.J. Blau, D.L. Carroll, J.N. Coleman, A.B. Dalton, A.P. Davey, A. Drury, B. McCarthy, S. Maier, A. Strevens, Adv. Mater. 10 (1998) 1091.

[8] H. Zengin, W. Zhou, J. Jin, R. Czrew Jr., D.W. Smith, L. Echegoyen, D.L. Carroll, S.H. Foulger, J. Ballato, Adv. Mater. 14 (2002) 1480. [9] Z. Wei, M. Wan, T. Lin, L. Dai, Adv. Mater. 15 (2003) 136. [10] R. Saniz, A.M. Benito, M.T. Martinez, J.F. Galindo, J. Sotres, A.M. Baro, B. Corraze, O. Chauvet, W.K. Maser, Adv. Mater. 17 (2005) 278. [11] X. Zhang, J. Zhang, R. Wang, Z. Liu, Carbon 42 (2004) 1455. [12] T. Jeevananda, N. Siddaramaiah, H. Kim, S.-B. Heo, J.H. Lee, Polym. Adv. Technol. 19 (2008) 1754. [13] W. Feng, X.D. Bai, Y.Q. Lian, J. Liang, X.G. Wang, K. Yoshino, Carbon 41 (2003) 1551. [14] J. Wang, J. Dai, T. Yarlagadda, Langmuir 21 (2005) 9. [15] B.K. Kuila, S. Malik, S.K. Batabyal, A.K. Nandi, Macromolecules 40 (2007) 278. [16] S. Bhattacharyya, E. Kymakis, G.A.J. Amaratunga, Chem. Mater. 16 (2004) 4819. [17] (a) E. Kymakis, G.A.J. Amaratunga, Synth. Met. 142 (2004) 161; (b) T.-M. Wu, Y.-W. Lin, Polymer 47 (2006) 3576. [18] A. Star, J.F. Stoddart, D. Steuerman, M. Diehl, A. Boukai, E.W. Wong, X. Yang, S.-W. Chung, H. Choi, J.R. Heath, Angew. Chem. Int. Ed. 40 (2001) 1721. [19] Y. Kanai, J.C. Grossman, Nano Lett. 8 (2008) 908. [20] R.P. Kingsborough, T.M. Swager, Adv. Mater. 10 (1998) 1100. [21] W.W. Focke, G.E. Wnek, Y. Wei, J. Phys. Chem. 91 (1987) 5813. [22] Y. Cao, A. Andreatta, A.J. Heeger, P. Smith, Polymer 30 (1989) 2305. [23] R. Menon, C.O. Yoon, D. Moses, A.J. Heeger, in Handbook of Conducting Polymers, 2nd ed., T.A. Skotheim, R.L. Elsebaumer, J.R. Reynolds, Eds. Marcell Dekker: N.Y. 1998. [24] A.G. MacDiarmid, J.C. Chiang, M. Halpern, W.S. Huang, S.L. Mu, N.L.D. Somasiri, W. Wu, S.I. Yaniger, Mol. Cryst. Liq. Cryst. 121 (1985) 173. [25] C.L. Gettinger, A.J. Heeger, D.J. Pine, Y. Cao, Synth. Met. 74 (1995) 81. [26] T. Vikki, L.O. Pietila, H. Osterholm, L. Ahjopalo, A. Takala, A. Toivo, K. Levon, P. Passiniemi, O. Ikkala, Macromolecules 29 (1996) 2945. [27] P.J. Kinlen, J. Liu, Y. Ding, C.R. Graham, E.E. Remsen, Macromolecules 31 (1998) 1735. [28] S. Kababya, M. Appel, Y. Haba, G.I. Titelman, A. Schmidt, Macromolecules 32 (1999) 5357. [29] R.K. Paul, C.K.S. Pillai, Synth. Met. 114 (2000) 27. [30] T.E. Olinga, J. Fraysse, J.P. Travers, A. Dufresne, A. Pron, Macromolecules 33 (2000) 2107. [31] T. Vikki, J. Ruokolainen, O.T. Ikkala, P. Passiniemi, H. Isotalo, M. Torkkeli, R. Serimaa, Macromolecules 30 (1997) 4064. [32] T. Vikki, H. Isotalo, J. Ruokolainen, P. Passiniemi, O. Ikkala, Synth. Met. 101 (1999) 742. [33] T. Jana, A.K. Nandi, Langmuir 16 (2000) 3141. [34] T. Jana, A.K. Nandi, Langmuir 17 (2001) 5768. [35] T. Jana, J. Chatterjee, A.K. Nandi, Langmuir 18 (2002) 5720. [36] T. Jana, S. Roy, A.K. Nandi, Synth. Met. 132 (2003) 257. [37] A. Andreatta, Y. Cao, J.C. Chiang, A.J. Heeger, P. Smith, Synth. Met. 26 (1988) 383. [38] D.G. Baird, J.K. Smith, J. Polym. Sci., Polym. Chem. Ed. 16 (1978) 61. [39] R.J. Nemanich, S.A. Solin, Phys. Rev. B 20 (1979) 392. [40] (a) Y. Bin, M. Kitanaka, D. Zhu, M. Matsuo, Macromolecules 36 (2003) 6213; (b) Q. Chen, Y. Bin, M. Matsuo, Macromolecules 39 (2006) 6528. [41] N.B. Clothup, L.H. Daly, S.E. Wiberley, Introduction to Infrared & Raman Spectroscopy;, Academic Press, New York, 1964. [42] J.E. Frommer, R.R. Chance, in: H.F. Mark, N.M. Bikales, C.G. Overberger, G. Menges (Eds.), Encyclopedia of Polymer Science Engineering, 5, 2nd ed., John Wily & Sons Inc., New York, 1986, p. 473. [43] C. Daniel, C. Dammer, J.M. Guenet, Polymer 35 (1994) 4243. [44] J. Tang, X. Jing, B. Wang, F. Wang, Synth. Met. 24 (1988) 231. [45] (a) W.R. Salaneck, B. Liedberg, O. Inganas, R. Erlandsson, I. Lundstrom, A.G. MacDiarmid, M. Halpern, N.L.D. Somasiri, Mol. Cryst. Liq. Cryst. 121 (1985) 191; (b) S. Palaniappan, V. Nivasu, New. J. Chem. 26 (2002) 1490. [46] J. Stejskal, M. Trchova, J. Prokes, I. Sapurina, Chem. Mater. 13 (2001) 4083. [47] A. Garai, B.K. Kuila, A.K. Nandi, Macromolecules 39 (2006) 5410. [48] H.S. Kim, H. Lee, K.S. Han, J.H. Kim, M.S. Song, M.S. Park, J.Y. Lee, J.K. Kang, J. Phys. Chem. B 109 (2005) 8983. [49] M. Ginic-Markovic, J.G. Matisons, R. Cervini, G.P. Simon, P.M. Fredericks, Chem. Mater. 18 (2006) 6258. [50] H. Xie, A. Cai, X. Wang, Phys. Lett. A 369 (2007) 120. [51] G. G Wallace, G. M Spinks, L.A.P. Kane-Maguire, P.R Teasdale, Conductive Electroactive Polymers: Intelligent Material Systems, 2nd ed., 2003 (1-st Chapter) CRC Press, Boca Raton, FL, USA. [52] J. Stejskal, P. Kratochvil, N. Radhakrishnan, Synth. Met. 61 (1993) 225. [53] W.S. Huang, A.G. MacDiarmid, Polymer 34 (1993) 1833. [54] Y. Xia, J.M. Wiesinger, A.G. MacDiarmid, A.J. Epstein, Chem. Mater. 7 (1995) 443. [55] J. Ruokolainen, H. Eerikainen, M. Torkkeli, R. Serimaa, M. Jussila, O. Ikkala, Macromolecules 33 (2000) 9272.