Conducting polymers obtained from quiescent and stirred solutions: effects on the properties.
Conducting polymers (CPs) have been extensively studied for the last few decades because of their valuable electrical, electrochemical, and optical properties [1-3]. Among them, heterocyclic CPs derived from polypyrrole (PPy) and, especially, polythiophene (PTh) have deserved special attention due to their potential applications in different fields, such as electronics, biomedical engineering, sensing, and energetic (solar cells and batteries) [1-10]. Poly(3,4-ethylenedioxythiophene), hereafter abbreviated PEDOT (Scheme 1), is one of the most important PTh derivatives because of its high electrical conductivity (up to 500 S [cm.sup.-1]) and ability to store charge, good thermal, electrochemical and chemical stability, fast doping-dedoping processes, and excellent biocompatibility [11-17].
In recent studies we reported multilayered conducting systems that alternate PEDOT and poly(N-methylpyrrole), abbreviated PNMPy (Scheme 1), layers [18-20]. It should be mentioned that the electrochemical and electrical properties of PNMPy are clearly worse than those of PEDOT [21-24], Multilayered films made of PEDOT and PNMPy (abbreviated ml-PEDOT/PNMPy) were prepared using the layer-by-layer (LbL) electrodeposition technique, showing better electrochemical properties than each of the two individual CPs. This improvement was attributed to a synergistic effect produced by a favorable interaction between PEDOT and PNMPy layers at the corresponding interfaces [18, 19, 25], Indeed, the coupling between such two CPs was found to be, in terms of enhancement of the electrochemical properties, significantly more positive in multilayered systems than in conventional copolymers ,
Although the electrochemical and morphological properties of PEDOT [11-17], PNMPy [21-24] and mlPEDOT/PNMPy [18-20, 25] were extensively investigated, static conditions (i.e., lack of solution agitation) were used in all cases. This work is essentially focused to examine the effects of controlled dynamical conditions (i.e., stirring the generation solution with a magnetic bar) in the polymerization process and in the properties of the resulting materials. For this purpose, individual PEDOT and PNMPy films, as well as PEDOT/PNMPy/PEDOT three-layered films (i.e., the internal and external layers are made of PEDOT while the central layer is of PNMPy), have been prepared in presence and absence of solution agitation (i.e., stirred and quiescent solutions, respectively) using a minimal concentration of monomer. Thus, the increase in the transport rate produced by controlled agitation, which accelerates the electropolymerization, has allowed us to optimize the yielding by reducing significantly the concentration of monomer in the generation medium. It should be remarked that the information reported in the literature about the influence of dynamical stirring in the polymerization and properties of PTh and PPy derivatives is very scarce [26-28], This is an amazing feature since in such studies it was suggested that controlled agitation improves significantly the flow of monomer at the electrode surface and affects the organization of polymer chains in the resulting films.
[FORMULA NOT REPRODUCIBLE IN ASCII]
3,4-Ethylenedioxythiophene (EDOT) and N-methylpyrrole (NMPy) monomers of analytical reagent grade were purchased from Aldrich and used as received without further purification. Anhydrous lithium perchlorate ([greater than or equal to] 95%) of the same grade was provided by Sigma-Aldrich and kept stored in an oven at 80[degrees]C before use in the electrochemical trials. Acetonitrile solvent of HPLC gradient grade was supplied by Panreac.
Synthesis by Anodic Polymerization
Both single component (PEDOT or PNMPy) and the three-layered (PEDOT/PNMPy/PEDOT) films were prepared by chronoamperometry under a constant potential of 1.40 V with a PGSTAT101 AUTOLAB potenciostat-galvanostat connected to a PC computer and controlled through the NOVA 1.6 software. All electrochemical experiments were conducted in a three electrode one-compartment cell under nitrogen atmosphere (99.995% in purity) at room temperature. The cell was filled with 50 mL of an acetonitrile solution containing 2 mM of the corresponding monomer and 0.1 M LiC104. It should be remarked that the concentration of monomer was smaller than that used in our previous works by five-folds (i.e., 10 mM) [16-20, 24].
The working and counter electrodes consisted of steel AISI 316 sheets of area 1 [cm.sup.2]. To avoid interferences during the electrochemical analyses, the working and counter electrodes were cleaned with acetone and dried in an air-flow before each trial. The reference electrode was an Ag|AgCl electrode containing a KC1 saturated aqueous solution ([E.sup.0] = 0.222 V vs. standard hydrogen electrode at 25[degrees]C). All the potentials reported in this work are referred to the latter electrode.
Three-layered films were prepared using LbL electro-deposition technique. Initially, the electrode was immersed into a solution containing the EDOT monomer and LiCl[O.sub.4] to generate the internal PEDOT layer (i.e., that in contact with the steel electrode). After rinsing with acetonitrile, the steel substrate coated with the first layer was subsequently immersed in a NMPy-and LiCl[O.sub.4]-containing solution to produce the central PNMPy layer. The resulting two-layered system was rinsed again with acetonitrile and immersed in an EDOT-containing solution to produce the external PEDOT layer. This electrochemical LbL procedure was detailed in previous studies [18-20]. With exception of the polymerization kinetics study for which polymerization times are explicitly indicated in the text, all the studies on individual polymers were performed with films produced using two polymerization times, [theta]= 100 and 300 s. Studies on the three-layered system were carried out with films prepared using [[theta].sub.1] = 30 and 100 s per layer (i.e., the total polymerization time in three-layered films was [theta] = 3 X [[theta].sub.1]).
Films of individual PEDOT or PNMPy were prepared in presence and absence of solution agitation, hereafter denoted as dynamic and static conditions, respectively. A magnetic bar and a stirrer SBS were used for the dynamical conditions, the optimum rotation speed being explicitly determined (see next section). However, the two PEDOT layers of three-layered films were obtained using static conditions in all cases, the difference between films generated dynamically and statically referring exclusively to the central PNMPy layer. Thus, three-layered films prepared under dynamic and static conditions refer to those in which the central PNMPy layer was generated in presence and absence of solution agitation, respectively.
The kinetics of the anodic polymerization processes using static and dynamic conditions were investigated considering different polymerization times. The weight of the insoluble polymer films ([W.sub.ox]) was determined as the difference between the masses of the coated and uncoated electrodes using a CPA26P Sartorius analytical microbalance with a precision of 2 x X [10.sup.-6] g. For this purpose, samples were obtained by scratching their deposits from the working steel electrode. The resulting powders were placed in a filter, rinsed with bidistilled water and ethanol, and subsequently dried under vacuum conditions for 24 h.
The thickness of the films was estimated from the mass of polymer deposited in the electrode, [m.sub.pol] using the procedure reported by Schirmeisen and Beck . The current productivity of polymers produced using static and dynamic conditions has been determined in this work, whereas the PEDOT and PNMPy densities have been taken from previous studies based on identical experimental conditions (i.e., the same potential, solvent and concentration of electrolyte) [16, 24].
The number of electrons consumed to incorporate a monomer into a polymer and to oxidize the resulting chain ([n.sub.ox]) has been determined using the following equation :
[n.sub.ox] = m x [Q.sub.pol]/F x [W.sub.ox] (1 - [W.sub.dop]) (1)
where M is the molar mass of the monomer (i.e., 142.18 and 81.12 g [mol.sup.-1] for EDOT and NMPy, respectively), F is the Faraday constant, [W.sub.ox] is the film weight (in milligram per centimeter square) and [W.sub.dop] is the mass of dopant per polymer unit of mass.
Electrochemical estimation of the doping level (dl) was carried using the following equation :
dl = (2[Q.sub.0]/[Q.sub.D] - [Q.sub.0]) x 100 (2)
where [Q.sub.D] is the total charge used for the polymer deposition and [Q.sub.0] is total charge of oxidized species in polymer films. The average number of electrons per monomer incorporated in the polymer chain ([n.sub.av]) has been obtained by discounting from [n.sub.ox] the oxidation charge used to compensate the charge of the dopant ion :
[n.sub.av] = [n.sub.ox] - dl (3)
The electroactivity, which refers to the charge storage ability, and electrochemical stability (electrostability) were determined by cyclic voltammetry using an acetonitrile solution with 0.1 M LiCl[O.sub.4]. The initial and final potentials were -0.50 V, while the reversal potential was 1.60 V. The scan rate was 100 mV/s in all cases. The electroactivity increases with the similarity between the anodic and cathodic areas of the first control voltammograms while the electrostability, which was determined as the loss of electroactivity (LEA, in %) decreases with the oxidation and reduction areas of consecutive control voltammograms. Specifically, the LEA was determined as [17-20]:
LEA = [DELTA]Q/[Q.sub.II] X 100 (4)
where [DELTA]Q is the difference of voltammetric charges (in C) between the second and the last cycle, and [Q.sub.II] is the voltammetric charge corresponding to the second cycle. In this work, measures of LEA refer to 50 consecutive oxidation-reduction cycles.
Electrochemical impedance spectroscopy (EIS) measurements were performed in potentiostatic mode at the open circuit potential using an AUTOLAB PGSTAT 30/ FRA 2 system. The amplitude of the EIS perturbation signal was 50 mV, and the studied frequency ranged from [10.sup.3] to [10.sup.-3] Hz. All experiments were carried out in an acetonitrile solution with 0.1 M LiCl[O.sub.4].
Scanning Electron Microscopy (SEM)
SEM studies were performed to examine the surface morphology of the synthesized materials. Dried samples placed in a Focussed Ion Beam Zeiss Neon 40 scanning electron microscope operating at 3 kV, equipped with an EDX spectroscopy system.
RESULTS AND DISCUSSION
Selection of the Dynamical Conditions and Preparation of the Films
Dynamical conditions were optimized by comparing the chronoamperograms collected for the oxidation of a 2 mM solution of NMPy in acetonitrile with 0.1 M LiCl[O.sub.4] considering stirring speeds of 300, 400, and 500 rpm (Fig. 1). The most stable and highest value of the current density was obtained for a stirring speed of 400 rpm (i.e., 1.82 mA [cm.sup.-2] after 60 s), the behavior observed for speeds of 300 and 500 rpm being very similar (i.e., 1.13 and 1.21 mA [cm.sup.-2], respectively, after 60 s). This should be attributed to the fact the flow of monomers for their successful incorporation into the electrode is lower at 300 rpm than at 400 rpm, whereas the flow is excessively high at 500 rpm making more difficult such incorporation. These results led us to select a stirring speed of 400 rpm for the dynamical conditions, which was used through the whole work.
The chronoamperometric study for the electrogeneration of PNMPy and PEDOT films from the corresponding 2 mM monomer solution in acetonitrile with 0.1 M LiCl[O.sub.4] was performed by applying a constant potential of 1.40 V during 300 and 100 s, respectively. Figure 2 compares the chronoamperograms obtained from stirred and quiescent solutions. In all cases uniform, adherent and insoluble polymeric films grew on the steel electrodes. As it can be seen, the two profiles are significantly different evidencing that the stirring process has a drastic effect in the polymerization. The molecular diffusion of both NMPy and EDOT monomers and, therefore, the polymerization rate of PNMPy and PEDOT, respectively, are higher for the stirred solutions than for the quiescent ones. Moreover, the response of NMPy and EDOT monomers upon stirring
is drastically different. Thus, stirring at 400 rpm causes a stabilization of the current density of PEDOT at 2.24 mA [cm.sup.-2] after ~60 s, whereas the current density of PNMPy drops continuously and progressively from 2.63 to 0.40 mA [cm.sup.-2] during 300 s. A similar difference, even though much less apparent, is the also observed for the quiescent solutions. Thus, in absence of stirring the current density of PEDOT stabilizes at 0.57 mA [cm.sup.-2] after 35 s, whereas that of PNMPy decreases slowly but progressively for more than 300 s (e.g., the current density at 35, 100, 200, and 300 s is 0.46, 0.42, 0.38, and 0.34 mA [cm.sup.-2], respectively). The different shapes of the profiles should be attributed to the porous and compact surfaces of PEDOT and PNMPy, respectively (see Morphology section). The porosity of PEDOT increases upon stirring, favoring both the movement of dopant anions at the interface and the polymerization process. This results in an initial increment of the current density, which subsequently stabilizes when the polymerization rate becomes equal to the monomers arrival rate (convective flux). In contrast, the compact morphology of PNMPy makes difficult the movement of the dopant ions and the polymerization process, resulting in a decrease of the current density. These results also indicate that, independently of the stirring, the flow of EDOT becomes constant after a given time, while the flow of NMPy needs time longer than 300 s to become stable.
The kinetics for the oxidation-polymerization of EDOT and NMPy was studied by generating films under a constant potential of 1.40 V and using different polymerization times. Both static and dynamic conditions were considered. The intervals of polymerization times were selected according to the deposited masses, which in all cases were within the same order of magnitude. Thus, polymerization times used to prepare PEDOT (PNMPy) under dynamical conditions ranged from [theta] = 30--400 s ([theta] = 30-600 s) while under static conditions times ranged from [theta] = 60-1000 s (from [theta] = 60-1500 s). Reproducible film weights, [W.sub.ox] (in mg), were obtained in all cases.
Figure 3 represents the variation of [W.sub.ox] against the charge consumed in the polymerization processes, ([Q.sub.pol] (in C [cm.sup.-2]), values of [Q.sub.pol] being directly calculated on each chronoamperogram. As it can be seen, linear correlations with excellent regression coefficients (r > 0.986) were obtained in all cases, reflecting that these are Faradaic processes. The slope of each of these linear plots correspond the current productivity, which is expressed as milligrams of electrogenerated polymer per coulomb of charge consumed during the anodic polymerization. The current productivity of PEDOT obtained using static and dynamic conditions is 0.536 and 0.789 mg [C.sup.-1], respectively, while that of PNMPy is 0.414 and 0.474 mg [C.sup.-1]. These results indicate that agitation alters the polymerization process, evidencing that the increase in transport rate of reactants favors the generation of more polymer weight at equal charge consumed. Although this effect is larger for PEDOT than for PNMPy, the current productivity of the latter is lower than that of former. This feature supports the fact that the dynamical conditions were optimized for PNMPy rather than for PEDOT, trying to improve the productivity and properties of the CP with the worst behavior. Interestingly, the current productivities derived from stirred 2 mM monomer solutions are relatively close to those reported for quiescent 10 mM monomer solutions (i.e., 0.875 and 0.619 mg [C.sup.-1] for PEDOT  and PNMPy , respectively).
Table 1 lists the values of dl, [n.sub.ox], and [n.sub.av] (Eqs. 1-3) determined for PNMPy and PEDOT prepared using [theta] = 100 s and both static and dynamic conditions. These parameters have been determined considering the data derived from the electropolymerization kinetics study. The dl (Eq. 2) measures the oxidation degree of the polymer chains, while both "ox and nav (Eqs. 1 and 3, respectively) provides information about the presence of cross-links. The dl of the CPs obtained from stirred solutions is smaller than that of materials prepared using quiescent solutions, this effect being slightly more pronounced for PNMPy than for PEDOT (i.e., the reduction is 40 and 24% for PNMPy and PEDOT, respectively). Accordingly, PNMPy and PEDOT chains incorporate one perchlorate anion every ~4 and ~2.5 repeating units, respectively, when prepared from quiescent solutions, whereas the number of repeating units increase to ~6 and ~3, respectively, for the materials formed from stirred solutions.
The degree of cross-linking is typically determined using [n.sub.av], which refers to the average number of electrons consumed to incorporate a monomer into the polymer chain . The value of [n.sub.av] for a linear polymer chain obtained from a typical condensation reaction is ~2.0, indicating that two protons and two electrons are involved in the formation of the [alpha]-[alpha] bond between the terminal repeating unit of the chain and the incorporated monomer . The value of [n.sub.av] for PNMPy obtained from the quiescent solution exceeds such value in ~20% (Table 1), evidencing an appreciable degree of reticulation. However, the [n.sub.av] of the material produced under stirring shows a value very close to 2.0, which indicates that agitation favors the linear growing of the polymer chains. Accordingly, application of dynamic conditions represents a serious methodological improvement for the preparation of N-substituted pyrrole derivatives, which usually show a considerably tendency to grow forming reticulated architectures [20, 32-34],
PEDOT molecules are exclusively formed by [alpha]-[alpha] linkages, since the dioxane ring fused onto the thiophene ring occupies the [beta]-positions of the latter. As crossrlinked structures are not possible, the value of [n.sub.av] obtained for PEDOT in absence of agitation, which is slightly >3, should be attributed to the low monomer concentration in the generation medium. Specifically, the current efficiency and, therefore, the yielding of the electropolymerization are very low, only 2/3 of the monomers being involved in the formation of polymer chains. In contrast, the [n.sub.av] of PEDOT produced under agitation is very close to 2, evidencing again the benefits of appropriated application of stirring.
Thickness of the Films
Table 2 compares the thickness (l) of the PNMPy and PEDOT films produced from quiescent and stirred solutions using both [theta] = 100 and 300 s. As it can be seen, the thickness of the films generated in absence of agitation is within the nanometric (~200 nm) and submicrometric (~0.5 [micro]m) scales for [theta] = 100 and 300 s, respectively. However, the thickness increases 2-5 folds when the materials are generated under stirring, producing a change in the length scale. Thus, the thickness of the PEDOT films produced under dynamical conditions is micrometric (~1-4 [micro]m), independently of [theta], while that of PNMPy films prepared using [theta] = 100 and 300 s is submicrometric (~0.3 [micro]m) and micrometric (~1 [micro]m), respectively. Similar conclusions are reached from Table 3, which compares the thickness determined for individual layers for the three-layered films produced under dynamic and static conditions.
However, detailed analysis of the thickness measured for individual layers in three-layered films indicates that stirring affects not only to the central PNMPy layer but also to the external PEDOT layer. Thus, the thickness of the PNMPy layer increases by 4-5 folds when the generation medium is agitated with a magnetic bar. This effect is accompanied by a significant increment of the roughness, as was evidenced by SEM images (see next section). The thickness of the external PEDOT layer, which was generated from quiescent solutions in all cases, was considerably affected by such change at the surface of the PNMPy layer, being ~2 folds thicker when the central layer was produced under stirring.
Morphology and Porosity
SEM micrographs of PNMPy films obtained from the quiescent and stirred solutions using [theta] = 100 s are displayed in Fig. 4a and b, respectively. Although compact structures were obtained in both cases, agitation induces the formation of a more porous globular morphology at the surface. This is because of the tendency of N-substituted PPy derivatives to form reticulations [20, 32-34], Thus, in static electropolymerization processes, which are typically characterized by a directional growing preference (i.e., in absence of cross-linking polymer chains grow perpendicularly to the substrate), crosslinking of PNMPy chains favours the formation of compact structures. In contrast, branching reduces significantly when the generation medium is stirred, which results in the formation of clusters at the surface. These clusters, which correspond to aggregations of linear segments of the polymer chains, increase the superficial roughness and the porosity. Indeed, the morphology displayed in Fig. 4b is reminiscent of that typically found for linear CPs , even though the porosity is significantly higher in the latter. The same effects are observed when SEM micrographs of PNMPy films produced using [theta] = 300 s are compared (not shown).
Figure 5 displays SEM micrographs of PEDOT films generated using [theta] = 100 s. As it can be seen, the film obtained without agitation is significantly less porous than that obtained using static conditions. Thus, the cavernous morphology displayed in Fig. 5a transforms into a spongy morphology upon stirring (Fig. 5b), superficial pores and caverns being wider and deeper in the material obtained using dynamic conditions. However, PEDOT is more porous than PNMPy in all cases. This well-known feature [17, 20] is consequence of the linear and reticulated architecture of PEDOT and PNMPy chains, respectively, as discussed above. The structural study presented in this section suggests that the mobility of the dopant ions is higher in the materials obtained from stirred solutions than in those produced using quiescent solutions, which is fully consistent with the chronoamperograms displayed in Fig. 2. Accordingly, both the electroactivity and electrostability are expected to be higher in the former than in the latter ones (see next sections).
Quantitative differences in the porosity of the films derived from static and quiescent solution were obtained using a previously described procedure . Specifically, the thicknesses of the films were estimated by cross-sectional SEM before and after apply 200 oxidation-reduction cycles ([l.sub.0] and [l.sub.200], respectively). This electrochemical treatment provokes important structural changes because of the reduction of both the mean pore size and the number of pores. As a consequence, the reduction in the porosity can be quantified through the parameter [DELTA], which is defined as follows:
[DELTA] = [l.sub.0] - [l.sub.200]/ [l.sub.0] X 100 (5)
The value of A obtained for PNMPy and PEDOT films prepared in absence of stirring and using [theta] = 100 s is 13 and 46%, respectively, these values increasing to 21 and 68%, respectively, for films derived from stirred solutions.
It is well known that the electroactivity of PEDOT is intrinsically higher than that of PNMPy  and, therefore, this section is exclusively focused in the effect of agitation. Figure 6a compares the voltammograms collected for PNMPy films generated under static and dynamic conditions, using [theta] = 300 s in both cases. The charge stored per unit of PNMPy mass is 245 and 211 C [g.sup.-1] for the polymer generated without and with (400 rpm) agitation, respectively. However, comparison of the electroactivities determined for the masses of PNMPy deposited on the electrode using identical polymerization time indicates higher ability to store charge for the material obtained from the stirred solution (400 rpm) than for the polymer prepared without agitation. However, the agitation-induced benefit decreases with the polymerization time (i.e., the mass deposited on the electrode), the difference between the electroactivities of the films generated using [theta] = 100 s (data not shown) decreasing to ~29%.
The improvement in the electroactivity is noticeably higher for PEDOT films. Thus, the charge stored per unit of PEDOT mass is 305 and 189 C [g.sup.-1] for systems produced without and with agitation, respectively. However, the total charges (i.e., without normalization per unit of mass) derived from the voltammograms collected for PEDOT films generated using [theta] = 300 s (Fig. 6b) reflects that agitation increases the electroactivity by 278%. Although this improvement decreases with [theta], as was also found for PNMPy, it is still very remarkable for films generated using [theta] = 100 s (not shown). In such case, the electroactivity of the film produced under dynamic conditions was 174% higher than that of the material generated without agitation.
The influence of dynamic conditions in the electroactivity of three-layered films was studied by considering the PEDOT/PNMPy/PEDOT system, agitation being only applied during the electrogeneration of the central PNMPy layer. This particular arrangement was selected because of the following two reasons: (i) PNMPy is the component with lowest intrinsic electroactivity; and (ii) the improvement produced by agitation in individual PEDOT is so high that the benefits produced by the multilayered system become relatively unimportant. Figure 6c compares the control voltammograms recorded for the three-layered films prepared under dynamic and static conditions using [theta] = 100 s per layer. Agitation increases significantly the electroactivity of the three-layered film (i.e., about two-folds), this effect being even higher for the films prepared using [theta] = 30 s per layer (data not shown).
Comparison of the cyclic voltammograms collected for the PNMPy and PEDOT films (Fig. 6a and b) with those of the three-layered films (Fig. 6c), indicates that, as expected, the ability to store charge is significantly higher for the latter than for the two individual CPs. This improvement was attributed to the synergistic effect produced by a favorable interaction between the PEDOT and PNMPy layers at the interfaces [17, 18]. Moreover, application of dynamic conditions in the preparation of the central layer results in an extra benefit, the electroactivity of the resulting three-layered film ([theta]= 100 s per layer) being 400% higher than that of an individual PEDOT ([theta] = 300 s) produced statically.
Figure 7a and b represents the variation of the LES for PNMPy and PEDOT films, respectively, prepared using both [theta] = 100 and 300 s against the number of consecutive oxidation-reduction cycles ([n.sub.redox]). As it can be seen, the electrochemical stability of films prepared under stirring is higher than that of films obtained from quiescent solutions. The electroactivity of PNMPy films produced under dynamic and static conditions using [theta] = 300 s decays 28 and 46% after [n.sub.redox] = 50 cycles, respectively, electrochemical degradation being considerably higher in films obtained using [theta] = 100 s (i.e., the electroactivity decreases 41 and 78%, respectively, after 50 cycles). Thus, the electrochemical stability increases with the polymerization time and, therefore, the thickness of the films. PEDOT films show a very similar behavior (Fig. 7b), even though the electrochemical stability of this CP is higher than that of PNMPy (Fig. 7a) in all cases (i.e., independently of 9 and the generation conditions). This is an expected result since the chronoamperograms discussed above (Fig. 2) showed that the current density of PEDOT stabilizes at a value considerably higher than that of PNMPy, independently of the generation conditions.
The loss of electrochemical stability is consequence of a reduction in the mobility of perchlorate counteranions upon oxidation and reduction processes (i.e., access to the polymer matrix and escape from the polymer matrix, respectively). This reduction in the mobility of the anions is in turn due to the formation of dense three-dimensional networks, which result from the cross-links induced by the anodic scans. Results displayed in Fig. 7 indicate that the importance of such effect decreases with increasing polymerization time. This is particularly relevant in the case PEDOT, which grows forming linear polymer chains since agitation makes the polymerization reaction significantly faster, increasing the thickness (Table 2). For PNMPy films, networks are less dense in films generated from stirred solutions since, as was discussed above, the intrinsic tendency of this CP to form reticulated structures is minimized.
The variation of the LES against [n.sub.redox] for the three-layered films prepared using [theta] = 30 and 100 s per layer is represented in Fig. 7c. Amazingly, the electrochemical stability of the three-layered films produced under dynamic conditions with the lowest 0 conditions is lower than that of the film obtained in absence of agitation. Specifically, the electroactivity of the three-layered films ([theta] = 30 s per layer) prepared using stirred and quiescent PNMPy solutions for the central layer decreases 84 and 62%, respectively, after [n.sub.redox] = 50 cycles. This should be attributed to the nanometric thickness of the central PNMPy layer. Thus, PNMPy chains formed after [theta] = 30 s onto the internal PEDOT layer are not large enough to avoid the formation of cross-links. Specifically, the roughness of the first PEDOT layer promotes the multidirectional growing of PNMPy chains, facilitating the formation of extensive cross-linking upon consecutive anodic scanning. This explains not only the poor electrochemical stability of the three-layered film produced in absence of agitation, which was previously reported for nanometric mutilayered films , but also the very low stability of the three-layered film obtained under stirring. Thus, application of dynamic conditions for only 30 s does not permit to overcome the unfavorable situation induced by the internal PEDOT layer.
In contrast, the behavior of three-layered films produced using [theta] = 100 s per layer is in fully concordance with the expected results. Thus, films generated under dynamic and static conditions present a significant electrochemical stability, even though that of the former is ~3I% higher than that of the latter. In these systems, the thickness of the central PNMPy is large enough to reduce the templating effect of the internal PEDOT layer, which leads to reduce the extension of the cross-linking effects upon consecutive anodic scans. Moreover, the latter effect also decreases significantly when the length of the PNMPy chains increases by the stirring. The electroactivity of the three-layered films ([theta] = 100 s per layer) electrogenerated under dynamic and static conditions decreases 44 and 69%, respectively, after [n.sub.redox] = 50 cycles.
The Nyquist plots for the EIS spectra recorded PEDOT ([theta] = 300), PNMPy ([theta] = 300 s), and three-layered ([theta] = 100 s per layer) films are displayed in Fig. 8a-c, respectively. As it can be seen, all systems show three well-defined regions. A capacitive semicircle related to the polymer-electrolyte interface, which is better defined for the PEDOT and three-layered films than for the PNMPy ones, is obtained at the high frequencies. At intermediate frequencies, a linear region with a slope around 45[degrees] is observed, reflecting Warburg behavior. Finally, a nearly vertical line is found at low frequencies, which is related with the faradic pseudo-capacitance of the films. In all cases such behavior, which increases with the slope of the line, is more pronounced in films obtained using dynamic conditions than in static films, especially for PEDOT and three-layered systems.
The semicircle at the high frequency region defines the electrolyte resistance ([R.sub.s]) and the charge-transfer resistance at the polymer-electrolyte interface ([R.sub.ct]). More specifically, [R.sub.s] and [R.sub.ct] correspond to the intercept of the semicircle with the real axis (Z') at high frequencies and the diameter of the semicircle along the Z', respectively. Table 4, which lists the conductivity ([[sigma].sub.ct] = l/[R.sub.ct]) of PEDOT, PNMPy, and three-layered films at the interface, reflects the improvement produced by agitation. Thus, the [[sigma].sub.ct], of samples obtained dynamically is one and, even in the case of PNMPy, two orders of magnitude higher than that of samples produced with stirring. This behavior should be attributed to the fact that, as was discussed above, films produced under agitation are more porous, which facilitates the mobility of ions between the polymer and the electrolyte.
PEDOT, PNMPy, and three-layered PEDOT/PNMPy/ PEDOT films have been prepared using static and dynamic conditions, and considering different polymerization times. Dynamic conditions consisted on controlled agitation of the generation solution with a magnetic bar at a stirring speed of 400 rpm. Chronoamperograms recorded for the oxidation of 2 mM monomer solutions in acetonitrile with 0.1 M LiCl[O.sub.4] indicated that the flow of EDOT becomes constant after a given time while the flow of NMPy decreases gradually with increasing polymerization time, such different behavior being observed in both stirred and quiescent solutions. Agitation has been found to increase the transport rate of reactants, favoring the generation of more polymer weight at equal charge consumed. In the case of PNMPy, agitation reduces considerably the cross-linking, improving this typically reticulated material. Moreover, dynamic conditions result in a better utilization of the monomers during the polymerization of PEDOT from dilute solutions.
Structural properties, like the thickness and the roughness, are drastically affected by stirring, increasing significantly with respect to materials obtained from quiescent solutions. The effect of dynamic conditions on the three-layered films is particularly noticeable. Thus, although the central PNMPy layer is the only produced under stirring, the increment in its roughness affects drastically to the thickness of the external PEDOT layer that is directly deposited onto it. However, agitation transforms the compact morphology of PNMPy into globular, and the cavernous morphology of PEDOT into spongy.
Agitation of the generation medium improves considerably both the electroactivity and electrochemical stability of PNMPy, PEDOT, and three-layered films. This behavior is fully consistent with the increment in the porosity of the samples derived from stirred solutions with respect to those obtained from quiescent solutions. The only exception to this behavior was found for the three-layered film prepared using [theta] = 30 s per layer, in which reticulation was not avoided because of the ultra-thin nature of the PNMPy layer. In summary, agitation of the generation solution is a very efficient procedure to prepare monolayered and multilayered films of PEDOT and PNMPy in electrolytic cells. Furthermore, this method allows the use of very low monomer concentrations optimizing the yielding of the polymerization reaction and improving the properties of the polymers.
Authors are indebted to the Centre de Supercomputacio de Catalunya (CESCA) for the computational resources provided.
[1.] T.A. Skotheim and J.R. Reynolds, "Handbook of Conducting Polymers," 3rd ed., in Conjugated Polymers, Vol. 1: Theory, synthesis, properties, and characterization. Vol. 2: Processing and Applications, CRC Press, Boca Raton (2007).
[2.] G.G. Wallace, G.M. Spinks, L.A.P. Kane-Maguire and P.R. Teasdale, Conductive Electroactive Polymers, CRC Press, Boca Raton (2009).
[3.] M.S. Freund and B. Deore, Self-Doped Conducting Polymers, Wiley, West Sussex, England (2007).
[4.] P. Beaujuge and J.R. Reynolds, Chem. Rev., 110, 268 (2010).
[5.] S.S. Zade, N. Zamoshchik, and M. Bendikov, Acc. Chem. Res., 44, 14 (2011).
[6.] I. Osaka and R.D. McCullough, Acc. Chem. Res., 41, 1202 (2008) .
[7.] F. Gamier, Acc. Chem. Res., 32, 209 (1999).
[8.] N.K. Guimard, N. Gomez, and C.E. Schmidt, Prog. Polym. Sci., 32, 876 (2007).
[9.] J.D. Stenger-Smith, Prog. Polym. Sci., 23, 57 (1998).
[10.] R.S. Bodade, J. Polym. Eng. 31, 209 (2011).
[11.] K. Lota, V. Khomenko, and E. Frackowiak, J. Phys. Chem. Solids., 65, 295 (2004).
[12.] K.S. Ryu, Y.G. Lee, Y.S. Hong, Y.J. Park, X.L. Wu, K.M. Kim, M.G. Kang, N.G. Park, and S.H. Chang, Electrochim. Acta, 50, 843 (2004).
[13.] S. Ghosh and O. Inganas, Electrochem. Solid State Lett., 3, 213 (2000).
[14.] L.J. del Valle, D. Aradilla, R. Oliver, F. Sepulcre, A. Gamez, E. Armelin, C. Aleman, and F. Estrany, Eur. Polym. J., 43, 2342 (2007).
[15.] E. Tamburri, S. Orlanducci, F. Toschi, M.L. Terranova, and D. Passeri, Synth. Met., 159, 406 (2009).
[16.] C. Ocampo, R. Oliver, E. Armelin, C. Aleman, and F. Estrany, J. Polym. Res., 13, 193 (2006).
[17.] D. Aradilla, F. Estrany, and C. Aleman, J. Phys. Chem. C., 115, 8430 (2011).
[18.] F. Estrany, D. Aradilla, R. Oliver, and C. Aleman, Eur. Polym. J., 43, 1876 (2007).
[19.] F. Estrany, D. Aradilla, R. Oliver, E. Armelin, and C. Aleman, Eur. Polym. J., 44, 1323 (2008).
[20.] D. Aradilla, F. Estrany, E. Armelin, and C. Aleman, Thin Solid Films, 518, 4203 (2010).
[21.] A. Cambra, M.I. Redondo, and MJ. Gonzalez-Tejero, Synth. Met., 139, 21 (2003).
[22.] E. Sanchez de la Blanca, M.I. Redondo, M.V. Garcia, M.V. Raso, M.A. Tortajada, and M.J. Gonzalez-Tejera, Synth. Met., 139, 145 (2003).
[23.] B. Teixeria-Dias, C. Aleman, F. Estrany, D.S. Azambuja, and E. Armelin, Electrochim. Acta, 56, 5836 (2011).
[24.] R. Oliver, A. Munoz, C. Ocampo, C. Aleman, and F. Estrany, Chem. Phys., 328, 299 (2006).
[25.] D. Aradilla, F. Estrany, R. Oliver, and C. Aleman, Eur. Polym. J., 44, 2222 (2010).
[26.] E. Brillas, J. Carrasco, R. Oliver, F. Estrany, J. Vilar, and J.M. Morlans, Electrochim. Acta, 45, 4049 (2000).
[27.] F. Estrany, R. Oliver, E. Garcia, E. Gualba, P.L. Cabot, and E. Brillas, Collect. Czech. Chem. Commun., 68, 1326 (2003).
[28.] J. Tietje-Girault, C. Ponce de Leon, and F.C. Walsh, Surf. Coat. Technol., 201, 6025 (2007).
[29.] M. Schirmeisen, and F.J. Beck, J. Appl. Electrochem., 19, 401 (1989).
[30.] E. Brillas, R. Oliver, F. Estrany, E. Rodriguez,and S. Tejero, Electrochim. Acta, 47, 1623 (2002).
[31.] D. Aradilla, D. Azambuja, F. Estrany, M.T. Casas, C.A. Ferreira, and C. Aleman, J. Mater. Chem., 22, 13110 (2012).
[32.] C. Aleman, J. Casanovas, J. Torras, O. Bertran, E. Armelin, R. Oliver, and F. Estrany, Polymer 49, 1066 (2008).
[33.] G. Fabregat, C. Aleman, M.T. Casas, and E. Armelin, J. Phys. Chem. B, 116, 5064 (2012).
[34.] D. Aradilla, F. Estrany, E. Armelin, R. Oliver, J.I. Iribarren, and C. Aleman, Macromol. Chem. Phys., 211, 1663 (2010).
Margarita Sanchez-Jimenez, (1) Carlos Aleman, (2,3) Francesc Estrany (1,3)
(1) Departament d'Enginyeria Quimica, Escola Universitaria d'Enginyeria Tecnica Industrial de Barcelona, Universitat Politecnica de Catalunya, Barcelona 08036, Spain
(2) Departament d'Enginyeria Quimica, E. T. S. d'Enginyers Industrials, Universitat Politecnica de Catalunya, Barcelona 08028, Spain
(3) Center for Research in Nano-Engineering, Universitat Politecnica de Catalunya, Campus Sud, Edifici C', C/ Pasqual i Vila s/n, Barcelona E-08028, Spain
Correspondence to: Carlos Aleman; e-mail: email@example.com Francesc Estrany; e-mail: firstname.lastname@example.org Contract grant sponsor: MICINN and FEDER; contract grant number: MAT2012-34498; contract grant sponsor: Generalitat de Catalunya (2009 SGR 925); contract grant sponsor: ICREA Academia; contract grant sponsor: Generalitat de Catalunya.
Published online in Wiley Online Library (wileyonlinelibrary.com).
TABLE 1. Number of electrons consumed to incorporate a monomer into the polymer and to oxidize the resulting chain ([n.sub.ox]; Eqn 1), doping level (dl Eqn 2), and difference between the two Parameters ([n.sub.av]; Eqn 3). Polymer [n.sub.ox] dl [n.sub.av] (conditions) PNMPy (static) 2.660 0.253 2.407 PNMPy (dynamic) 2.108 0.154 1.954 PEDOT (static) 3.579 0.433 3.146 PEDOT (dynamic) 2.299 0.330 1.969 TABLE 2. Thickness (l) of PNMPy and PEDOT films electrogenerated under dynamic (stirring the generation solution with a magnetic bar at 400 rpm) and static conditions considering different polymerization times ([theta]). Polymer [theta] Conditions l PNMPy 100 s Static 218 nm Dynamic 0.32 [micro]m 300 s Static 0.42 [micro]m Dynamic 0.98 [micro]m PEDOT 100 s Static 198 nm Dynamic 1.0 [micro]m 300 s Static 0.60 [micro]m Dynamic 3.7 [micro]m TABLE 3. Thickness (t) of 3-layered films electrogenerated under dynamic (stirring the generation solution of the central layer with a magnetic bar at 400 rpm) and static conditions considering different polymerization times (0). 0 Conditions Internal layer 30 s per layer Static 74 nm Dynamicb 74 nm 100 s per layer Static 132 nm Dynamicb 132 nm 0 Central External layer Total layer 30 s per layer 81 nm 86 nm 241 nm (a) 0.45 [micro]m 201 nm 0.72 pm 100 s per layer 236 nm 262 nm 0.70 pm 0.91 pm 394 nm 1.4 pm (a) The thickness of a PEDOT film generated under static conditions using [theta]- 90 s is l = 150 nm. (a) 3-Layered films prepared under dynamic conditions refer to those in which only the central PNMPy layer was generated under dynamic conditions. TABLE 4. Charge transfer conductivity ([[sigma].sub.ct], in x [10.sup.3] x S/cm) obtained by electrochemical impedance spectroscopy for PEDOT ([theta] = 300 s), PNMPy (0 = 300 s) and 3-layered films ([theta] = 100 s per layer) electrogenerated under dynamic and static conditions. Static Dynamic PEDOT 2.6 x 10-3 2.7 x 10-2 PNMPy 6.6 x 10-6 6.3 x 10-4 3-lavered (a) 9.5 x 10-4 2.1 x 10-3 (a) 3-Layered films prepared under dynamic conditions refer to those in which only the central PNMPy layer was generated under dynamic conditions.
|Printer friendly Cite/link Email Feedback|
|Author:||Sanchez-Jimenez, Margarita; Aleman, Carlos; Estrany, Francesc|
|Publication:||Polymer Engineering and Science|
|Date:||Sep 1, 2014|
|Previous Article:||A highly effective reactive liquid crystal for the improved [beta]-nucleation of isotactic polypropylene.|
|Next Article:||Impact of aliphatic amine comonomers on DGEBA epoxy network properties.|