Elsevier

Electrochimica Acta

Volume 46, Issue 17, 15 May 2001, Pages 2721-2732
Electrochimica Acta

p- and n-doping processes in polythiophene with reduced bandgap. An electrochemical impedance spectroscopy study

https://doi.org/10.1016/S0013-4686(01)00485-6Get rights and content

Abstract

Electrochemical impedance spectroscopy has been used for the characterisation of electrodes modified with different polythiophenes, namely poly[4,4′-bis(butylsulfanyl)-2,2′-bithiophene], poly[4,4′-bis(methylsulfanyl)-2,2′-bithiophene] and poly(3-methylthiophene), at different applied potentials, using different supporting electrolytes. By comparison of the results obtained under experimental conditions in which n-doping is prevented and those obtained from tests where it does occur, some general features have been deduced, all of them being coherently described by a recently proposed ‘generalised transmission line circuit’ model: impedance plots at different applied potentials exhibit progressive changes which are well accounted for by the ‘evolving’ model. The results obtained on the n-doping process of S-alkyl substituted polymers suggest a behaviour interestingly similar to that exhibited in the p-doping; this supports a symmetry that was also found by us in a previous work, with respect to the incorporation and release of counterions during the n- and p-charge–discharge processes.

Introduction

The electrochemical switching of a conducting polymer (CP) between the doped and undoped states involves both electron and ion injection into or extraction from the polymer, concomitant with the transport of electronic and ionic charges within the CP. Consequently, the charge transport processes inside the CP bulk, as well as across the CP's interfaces, constitute crucial points in many applications and have been the object of extensive researches. In particular, it has been reported that in most cases ion transport is the slow process, i.e. the step limiting the switching rates of, for example, displays and electronic devices based on similar materials [1].

When applied to CP modified electrodes, electrochemical impedance spectroscopy (EIS) has proved to be quite a useful tool for studying the charge transport inside the polymer coating and the charge transfer at the metal/polymer and polymer/solution interfaces [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]. However, several problems concern both the correct execution of the measurements and the interpretation of the data obtained, due to the high complexity of the system. As to the measurements, a first, very serious though often underestimated, problem arises from the changes that the CP deposit undergoes when working on it for the long times required by the execution of a complete series of impedance measurements [1], [16]. Unfortunately, this ‘drawback’ is particularly heavy for the CPs that, in principle, would deserve the greatest attention, i.e. those with the simplest formulation [17], [18], [19].

We synthesised a series of polymers, e.g. poly[4,4′-bis(butylsulfanyl)-2,2′-bithiophene] (PBST), poly[4,4′-bis(octylsulfanyl)-2,2′-bithiophene] [20] and poly[4,4′-bis(2-methylbutylsulfanyl)-2,2′-bithiophene] [21]. The series of homologous alkylsulfanyl derivatives is prolonged by the poly[4,4′-bis(methylsulfanyl)-2,2′-bithiophene] (PMST), whose electrochemical behaviour was first studied by Bäuerle and co-workers [24]. Hence, these polymers attracted our attention, on the one hand, due to their very interesting properties, reported in the cited relevant literature and, on the other hand, specifically due to (i) the reliability of the repeatable data obtained on polymers which are particularly stable; (ii) the good definition and resolution of the processes attributable to polarons and bipolarons formation, respectively; and (iii) the relatively easy reduction leading to n-doping and the consequent reduced bandgap [22], [23]. All these characteristics suggested to us an exhaustive study of the properties of this class of compounds by looking at the results of different instrumental approaches as a whole.

Numerous models have been proposed for the elaboration of EIS data reported in this article, both of mathematical nature and based on a physical view of the system, i.e. on equivalent electrical circuits. However, none of the models proposed has achieved general acceptance, apart from the cases of quite simple systems. Basically, two different views have been proposed, making reference to the structural properties of the CP, i.e. the so-called ‘uniform’ and ‘porous’ medium [3], [4], [5], [6]. The mathematical approaches [3], [6] are often based on the former model while equivalent electrical circuits [4], [5], [10], [11], [12], [13], [14], [15], [16] are mainly based on the latter, introducing suitable ‘transmission line circuits’ to account for charge percolation inside the polymer. One of the aims of our work is just that of finding a correspondence between the impedance plots obtained and a given equivalent circuit, thus suggesting a rationale to the physics of the system and, at the same time, also allowing a choice among different possible hypotheses about the characteristics of the system under the specific conditions. In a recent publication we analysed the impedance behaviour of the PBST-coated electrode in 0.1 M LiClO4 supporting electrolyte, MeCN solvent [16]. The use of LiClO4 as supporting electrolyte simplifies the system since only p-doping is possible in such a medium [16], [22], [25], so that the neutral polymer can be studied over a wide potential range and, due to the nature of the ions involved, accumulation of both cations and anions inside the polymer by repetitively cycling the potential is unfeasible. By using the ‘generalised transmission line circuit model’ recently proposed by Buck and Mundt [9], we could coherently give account for the results obtained and, applying reasonable simplifications to the model, an approximate evaluation of the relevant parameters was also possible. The model has clear physical meaning and general validity since it does not require any a priori assumptions about the status of the polymer, i.e. if it has a porous or a uniform structure. Evolution of the general circuit to simpler formulations accounts for specific situations. In the work described in the present paper we also used a different supporting electrolyte, i.e. TBAPF6, which allows the study to be extended to the n-doping process, so collecting a complete picture of the impedance behaviour of the coating, in the whole potential range of stability of the polymer–solution system. Once more, conclusions have been drawn out on the basis of the generalised transmission line circuit model. In order to verify if these conclusions are not just limited to the studied polymer, we also tested two other polythiophenes, viz. poly(3-methylthiophene) (PMT) and PMST, using both LiClO4 and TBAPF6 supporting electrolytes.

Section snippets

Experimental

The dimeric starting compounds 4,4′-bis(butylsulfanyl)-2,2′-bithiophene (BST) and 4,4′-bis(methylsulfanyl)-2,2′-bithiophene (MST) were obtained as described in Refs. [26] and [27], respectively; 3-methylthiophene (MT) was obtained from Merck, 98% pure. Electropolymerisation and electrochemical characterisations were performed using the computerised Autolab Mod. PGSTAT 30 electrochemical instrument, equipped with FRA module (Ecochemie, Utrecht, The Netherlands). All electrochemical tests were

Results and discussion

Fig. 1a reports cyclic voltammetric curves recorded on a PBST-coated Pt electrode in a 0.1 M TBAPF6 MeCN solution. The continuous line trace refers to the steady-state i/E curve relative to a potential scan between the limits 0.00 and +1.10 V; the broken line trace shows the steady-state i/E curve between −1.90 and +1.10 V. Steady-state conditions are reached from the third cycle onwards. In addition to the polaron and bipolaron formation peaks, coupled with the relevant undoping responses, a

Conclusions

EIS has proved to be quite a helpful tool in studying some properties of CPs, evidencing aspects that can be hardly picked up or only hypothesised on the basis of the measures from other techniques. The possibility of working on CPs with a reduced bandgap that, consequently, undergo both p- and n-doping and that, depending on the nature of the supporting electrolyte, exhibit different behaviour as to the p- and (mainly) the n-doping processes, allowed us to first evidence an interesting

Acknowledgements

MURST (Rome) (Ricerche di Interesse Nazionale) is acknowledged for financial support. H.D. gratefully acknowledges the financial support of the University of Modena and Reggio Emilia for his one-year stay in Modena.

References (32)

  • M.M. Musiani

    Electrochim. Acta

    (1990)
  • B.W. Johnson et al.

    J. Electroanal. Chem.

    (1994)
  • R.P. Buck et al.

    J. Electroanal. Chem.

    (1993)
  • R.P. Buck et al.

    Electrochim. Acta

    (1999)
  • X. Ren et al.

    J. Electroanal. Chem.

    (1997)
  • X. Ren et al.

    Electrochim. Acta

    (1996)
  • K. Juttner et al.

    Electrochim. Acta

    (1999)
  • G. Zotti et al.

    Synth. Met.

    (1989)
  • C. Visy et al.

    J. Electroanal. Chem.

    (1998)
  • M. Zhou et al.

    Electrochim. Acta

    (1999)
  • B. Ballarin et al.

    Electrochim. Acta

    (2001)
  • A. Smie et al.

    J. Electroanal. Chem.

    (1998)
  • M. Mastragostino et al.

    Electrochim. Acta

    (1990)
  • J.R. Macdonald

    Electrochim. Acta

    (1990)
  • G. Lang et al.

    Electrochim. Acta

    (1999)
  • X. Ren et al.

    J. Phys. Chem.

    (1993)
  • Cited by (49)

    • Effects of doping methods and dopant sizes on the performance of solar cells constructed with anchor-guided photoelectrochemical polymerization of thiophene

      2020, Electrochimica Acta
      Citation Excerpt :

      In ex-situ doping, currents decreased from cycle 1 to cycle 3 and reached a steady state in later cycles. This is the behavior expected from the ejection of anions (ClO4−) in the preformed PT films and the incorporation (doping) of cations in the vacancies [18]. The steady states are indicative of the completion of the doping process.

    • In situ investigations of electrogenerated polybithiophene film growth on indium tin oxide substrate using optical fixed-angle reflectometry

      2012, Thin Solid Films
      Citation Excerpt :

      Furthermore during the oxidation stage, the film was globally positively charged due to radical BT+ cations located along the backbone of the polymer. Anions were thus needed to ensure the electro neutrality of the film, which corresponds to the doping phenomenon [3,39,40,43]. Consequently when PF6− concentration increased near the surface, so did ΔS/S0.

    • The effect of chain length on the dielectric and optical properties of oligothiophenes

      2011, Synthetic Metals
      Citation Excerpt :

      Therefore it is preferred that the planar molecules are closely packed together in a layered fashion [2,19,20] as this gives a higher mobility. This is also true for polythiophene [14–19,21–23]. To get significant conduction the intermolecular movement of charge must be perpendicular [1,2] to the molecular axis and will be mediated by the weak van der Waal forces of the crystal bonds.

    • Polypyrrole film modified with electroless metal deposition

      2011, Progress in Organic Coatings
      Citation Excerpt :

      This behaviour was thought to be helpful to evaluate polymer coated sample behaviour. The coated sample prepared with single step route (Fig 17B) exhibited a negative slope between −0.40 and −0.10 V, which is attributed to p-type semiconductor behaviour [27,34,35]. However, at potentials higher than −0.10 V, the measured interfacial capacitance was found to decrease continuously.

    View all citing articles on Scopus
    1

    ISE member.

    2

    On leave from University of Science & Technology of China, Hefei, People's Republic of China.

    View full text