Preview only show first 10 pages with watermark. For full document please download

Aging And Thermal Degradation Of Poly( N-methylaniline

Aging and thermal degradation of poly( N-methylaniline

   EMBED


Share

Transcript

  Thermochimica Acta 411 (2004) 109–123 Aging and thermal degradation of poly(  N  -methylaniline) P. Syed Abthagir a , R. Saraswathi a , ∗ , S. Sivakolunthu b a  Department of Materials Science, Maduai Kamaraj University, Madurai 625021, Tamil Nadu, India b  Department of Inorganic Chemistry, School of Chemistry, Madurai Kamaraj University, Madurai 625 021, Tamil Nadu, India Received 12 May 2003; received in revised form 23 July 2003; accepted 11 August 2003 Abstract The conductivity aging and thermal stability of poly(  N  -methylaniline) are reported. Poly(  N  -methylaniline) doped with chloride ion waselectrochemicallysynthesized.Theconductivitydataobtainedinthetemperaturerangebetween118and483KareanalysedbyArrheniusandMottmodelstoelucidatetheconductionmechanism.Thethermaldegradationofbothdopedanddedopedsamplesofpoly(  N  -methylaniline)inairandnitrogenatmospherehasbeenfollowedusingthermogravimetricanddifferentialthermalanalysistechniques.Thepolymerisheat-agedatvarioustemperaturesandtheagedsamplesareanalysedbyFT-IR,SEMandXRD.Thethermogravimetricdataarefurtheranalysedbythreedifferent methods: Horowitz and Metzger [Anal. Chem. 35 (1963) 1464], Coats and Redfern [Nature 201 (1964) 68], Chan et al. [Synth. Met.31 (1989) 95] to evaluate the energy of activation. The applicability of the three methods for the evaluation of kinetic parameters is discussed.© 2003 Elsevier B.V. All rights reserved. Keywords:  Conducting polymer; Poly(  N  -methylaniline); Aging; Thermal degradation; Activation energy 1. Introduction A knowledge of thermal stability of conducting polymersis important for their use in many practical applications.While the synthesis, structure and redox properties of theelectroactive polymers have been reported extensively, therehave been only few systematic studies on their aging andthermal degradation behaviour. There can be two strategiesin the description of the aging of conducting polymers. Theconduction models describe the electrical aging of conduct-ing polymers in a quantitative manner [1]. On the other hand, the thermogravimetric data can be used to explainstability in terms of the structural changes in the polymerduring thermal aging [2]. A combined use of both the ap- proaches will be helpful not only to assess but also to deviseways to improve the stability of the conducting polymers.A survey of literature reveals that most of the conductivityand thermal aging studies have been mainly made on poly-acetylene [3,4], polyaniline doped with various counter ions [5–18], polypyrrole and polythiophene [19]. The objective of this report is to discuss the conductivity aging and thermalstability of poly(  N  -methylaniline). Poly(  N  -methylaniline) ∗ Corresponding author. Fax:  + 91-452-2459181.  E-mail address:  [email protected] (R. Saraswathi). and its sulfonated analogue have been recently character-ized as cathode active materials in rechargeable batteriesin this laboratory [20]. This polymer, although being aderivative of polyaniline, is different from polyaniline be-cause of the block of the proton exchange sites by methylsubstituents. As a consequence, the deprotonation of theimine group occurring during the second oxidation stepof polyaniline does not occur in poly(  N  -methylaniline). Inother words, poly(  N  -methylaniline) can be prevented fromgoing to the completely oxidized pernigraniline base formthereby reducing the risk of electrochemical degradation[21,22]. Our recent study has shown that the redox stabil-ity of poly(  N  -methylaniline) is more than that of its parentpolymer, polyaniline [23]. In this context, it would be worthwhile to make a systematic report on the conductivityand thermal aging of this polymer. 2. Experimental  N  -Methylaniline (SRL) was purified by vacuum distil-lation. Poly(  N  -methylaniline) doped with chloride ion wasobtained on a platinum substrate galvanostatically by ap-plying a constant current of 2mAcm − 2 . The electrolytesolution consisted of 1M  N  -methylaniline and 1M hy-drochloric acid. Several runs of polymer preparation were 0040-6031/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.tca.2003.08.010  110  P.S. Abthagir et al./Thermochimica Acta 411 (2004) 109–123 needed to collect the polymer in bulk for the measurementsof conductivity, infra-red spectroscopy, X-ray diffraction,thermogravimetry analysis (TGA) and morphological stud-ies. The room temperature conductivity of the dry polymersample was measured to be 6 . 6 × 10 − 4 Scm − 1 . Dedopedsamples of the polymer are obtained by treatment with0.1M ammonium hydroxide for about 6h.The electrochemical preparation of the polymer wasmade using a scanning potentiostat/galvanostat (PAR Model263A). The powder sample was pressed into a pellet usinga Perkin Elmer hydraulic press by applying a pressure of 6 tonnes. The resistance of the pellet was measured in thetemperature range between 118 and 483K by two probemethod using a digital multimeter (Model 195 A, KeithleyInstruments Ltd., USA). A liquid nitrogen bath cryostat(Model DP-422, Scientific Solution, Mumbai, India) wasused for low temperature resistivity measurements. Thecryostat consisted of a sample chamber surrounded by vac-uum isolation chamber. The sample holder was fitted with aplatinum temperature sensor (Pt-100) and a 25   heater. GEvarnish (which has good thermal conductivity at low tem-perature) was used for mounting the sample. A mixture of toluene–ethanol (1:1) was used as the thinner. The electricalconnections to the sample were made by conducting silverpaint. The morphology was inspected by a scanning electronmicroscope (Hitachi S 450, 4kV). The IR data were ob- Fig. 1. Temperature dependence of electrical conductivity for poly(  N  -methylaniline) chloride. tained using a FT-IR spectrophotometer (JASCO-410). Thethermograms were recorded at a heating rate of 10Kmin − 1 using a thermal analyser (NETZSCH-Geratebau GmbHSTA 409 PC). The X-ray diffraction data were obtained(JEOL JDX-8030) at a rating of 40kV, 20mA. Ni-filteredCu K   radiation ( λ = 0 . 154nm) was used. 3. Results and discussion 3.1. Temperature dependence of conductivity Fig. 1 shows the variation of conductivity of poly(  N  -methylaniline) doped with chloride ion as a function of temperature. The increase in conductivity with rise in tem-perature from 118 to 293K is typical of semiconductors.There is a very slow increase in the value between 293and 351K and thereafter the conductivity shows a gradualdecrease between 351 and 483K. This decrease in conduc-tivity is attributed to the elimination of the dopant that leadsto a decrease in the concentration of polarons. A support forthe above inference is obtained from the thermogravimetryof poly(  N  -methylaniline) which will be described in detaillater.The temperature dependent conductivity data obtained inthe range between 118 and 301K can be fitted to Arrhenius  P.S. Abthagir et al./Thermochimica Acta 411 (2004) 109–123  111Fig. 2. Arrhenius plot of conductivity for poly(  N  -methylaniline) chloride. equation of conductivity [24] σ   = σ  o  exp  − E a kT    (1)where  E  a  is the activation energy. A plot of ln( σ  ) versus  T  − 1 is a straight line (Fig. 2). The  E  a  value obtained from theslope is 0.11eV.The electrical transport mechanism can be due to Mottvariable range hopping [25–27]. In this model, the con-ductivity is treated as temperature activated hopping fromcentre to centre. The lack of ordering in amorphous con-ducting polymers is expected to produce localized electronicstates. An electron initially in a localized state can moveby thermally activated hopping to another localized stateand conduction occurs through variable range hopping of the electron between these localized states. The Mott modeldefines the relation between conductivity and temperatureas below: σ   = K o T  − 1 / 2 exp  −  T  o T   1 / 4   (2) T  o  = 16 α 3 k B N(E F ) (3) K o  = 0 . 39  N(E F )αk B  1 / 2 ν o e 2 (4)where  α − 1 is the decay length of the localized state;  ν o  ahopping attempt frequency;  N  (  E  F ) the density states at theFermi energy level;  e  the electronic charge;  k  B  the Boltz-mann constant;  T  o  the Mott characteristic temperature and K  o  the Mott characteristic conductivity parameter.The conductivity data obtained for poly(  N  -methylaniline)in the range of temperature between 118 and 301K are usedto construct the Mott plot (Fig. 3). The Mott parameters  T  o (1 . 3 × 10 8 K) and  K  o  (1 . 8 × 10 9 Scm − 1 K 1 / 2 ) are obtained,respectively, from the slope and intercept of the straight lineplot. Substituting these values in equations and assuminga value of 10 12 s − 1 for  ν o  [1], the decay length  α − 1 and  N  (  E  F )areestimatedtobe1.18Åand8 . 57 × 10 20 eV − 1 cm − 3 ,respectively.There have been some reports on the temperature de-pendence of conductivity for polyaniline. Salaneck et al.studied the temperature dependence of conductivity in therange between 166 and 333K for undoped and salt formsof polyanilines [28] and reported an activation energy of 0.05eV for the dedoped form. The activation energy showedan increase from 0.037 to 0.493eV when the doping pH wasvaried between 0 and 6.3 [29]. The electrical conductivity of poly(  N  -methylaniline) and its copolymer with anilinewas measured by Langer over a temperature range of 200 to400K [30]. The activation energy of the conductance pro- cess was found to be a function of the relative concentration  112  P.S. Abthagir et al./Thermochimica Acta 411 (2004) 109–123 Fig. 3. Mott plot of conductivity for poly(  N  -methylaniline) chloride. of   N  -methylaniline and aniline units in a polymer chainand it changed from 0.267 to 0.08eV. Probst and Holze[31] suggested that the method of polymerization wouldbe an important factor in modeling the conduction process.Borkar and Gupta [32] measured the electrical conductivityof poly(  N  -ethylaniline) in the range of temperature between303 and 433K and the activation energy of conductivityhad been found to be 0.078eV. The results suggested a po-laron hopping conduction mechanism. The higher activationenergy (0.11eV) of poly(  N  -methylaniline) obtained in thepresent study is in agreement with its lower conductivity(6 × 10 − 4 Scm − 1 ). It is inferred that the intrinsic charge car-riers are less at room temperature in poly(  N  -methylaniline)than in polyaniline.In spite of the availability of the large volume of litera-ture on polyaniline, it is found that only very few reportsexist on the applicability of variable range hopping modelfor polyaniline. In one of the earliest reports by Salanecket al. [28], the  T  o  parameter for the salt form of polyanilinewas given as 7 × 10 6 K. Pingsheng et al. [29] have madea detailed study for polyaniline samples doped at differentpH values. The  K  o  and also the  T  o  values increased withincreasing pH. More recently Gosh et al. [17] have studiedthe transport properties of chloride doped polyaniline in therange between 1.8 and 300K. The variation of conductivitywas remarkably high at high temperature whereas it wasvery small at low temperature. From the analysis of theresults it was concluded that the high temperature conduc-tivity obeyed the variable range hopping conduction mech-anism. A cross-over from Mott ( T  − 1 / 4 ) to Efros–Shklovskii( T  − 1 / 2 ) in temperature dependent conductivity was observedat 10K.It is difficult to make a direct comparison of the resultsobtained in the present study with those of literature. Thereare uncertainties in the value used for the phonon frequency( ν o ). For example Sato et al. [33] have used 6 × 10 13 s − 1 for  ν o . Gosh et al. [17] have used  ν o  values ranging from5 × 10 8 s − 1 and 8 × 10 12 s − 1 . A value of 10 12 s − 1 has beenused in the present study [1]. Further the slope  T  o  was cal-culated from the plot of log 10  σ  T  1 / 2 versus  T  − 1 / 4 insteadof using the natural logarithmic function on the  y -axis [33].The slope obtained from log 10  plot can be 2 orders less thanthat obtained from the ln plot. The larger  T  o  and  k  o  valuesfor poly(  N  -methylaniline) chloride compared to polyanilinechloride (Table 1) imply a short decay length and low den- sity of states at the Fermi level. 3.2. Thermal stability There are a number of reports on the thermal degrada-tion of polyaniline [10–15,34–37]. In general, the neutral polymer showed a one-step weight loss while the dopedpolymer showed a two-step process. The first step was at-tributed to the loss of dopant counter anion and second step  P  . S   .A  b   t   h   a  g i   r  e t   a l    . /   T  h   er  m o c h   i    m i    c aA  c t   a4  1  1    (   2   0   0  4    )   1   0   9  –1  2   3   1   1    3    Table 1A comparison of Mott parameters of poly(  N  -methylaniline) with those of polyanilinePolymer Mott parameters Density states atFermi energy(eV − 1 cm − 3 )Decay length,   (Å)Phonon frequencyused,  ν o  (s − 1 )Reference T  o  (K)  K  o  (Scm − 1 K 1 / 2 )Polyaniline 7  ×  10 6 – – – – [28]Polyaniline chloride 3.3  ×  10 6 to 2.1  ×  10 32 6.9  ×  10 6 to 2.3  ×  10 10 9.43  ×  10 22 1.15–3.14  ×  10 − 38 – [29]Polyaniline chloride(20% dopant concentration)4529.21 – 5.26  ×  10 19 72 5  ×  10 8 [17]Polyaniline chloride(40% dopant concentration)6.46  ×  10 5 – 4.88  ×  10 17 65.5 8  ×  10 12 [17]Poly(  N   methylaniline) chloride 1.3  ×  10 8 1.81  ×  10 9 8.57  ×  10 20 1.18 1  ×  10 12 Present work