ECL was generated both in the solution phase with DPA (9,10-diphenylanthracence) and in the polymer phase with PPV-derivative polymers. There are five major results from the present work. They are listed as follows.
In the literature on solution ECL, it has been widely reported that the emission intensity observed in the positive potential step, anodic ECL differs from that in the negative potential step, cathodic ECL (section 3.1.4). In our experiments we could observe equal intensities of anodic and cathodic ECLs by appropriately choosing the waiting period 'tw'.
|Fig. 5.1a. Typical shape of the triple potential steps including the waiting period tw in the ECL experiments.|
The longer the tw, the better is the similarity. The potential of the pulse during tw was kept at 0 V or -100 mV. The reasons for the improved performance can be attributed to the elimination of an adsorbed species on the electrode during the negative potential step. The adsorbate was speculated to be adsorbed hydrogen .
In the case of ECL with polymer-coated electrodes, symmetrical ECL has never been reported before. We did observe equal intensities from the conducting polymers MEH-PPV and DB-PPV with the proper choice of the potential step magnitude, the width and the waiting period. The absolute values of the potential steps were chosen to be larger than the absolute values of the [page 107↓]standard potentials for the reduction and oxidation of the polymer to ease the electrode reaction. But care was taken to avoid over oxidation or reduction leading to the formation of the bipolarons, which reduce the ECL intensity. Similarly in setting the width of the potential step, the time constant of the system (Rux Cdl = τ) must be considered. Since Ru is larger in the polymers, it needs a longer time (≈ 100 ms) for charging the electrode/polymer interface compared to the case of DPA. On the other hand the thickness of the polaron diffusion layer in the polymer needs to be considered in choosing the potential step. A longer potential step width may lead to the thin layer effect in which the diffusion layer exceeds the polymer thickness (section 2.1.2i). It was found that the following values of tw, tf and tr gave the most reproducible and symmetrical ECL transients. For DPA, tf = tr = 50 ms and tw = 200 ms at 0 V. For DB-PPV, tf = tr =150 ms and tw = 300 ms at 0 V. The results were also obtained over a range of a waiting period, tw ranging between 0.2 s and 1 s.
The ECLs obtained from the conducting polymers, MEH-PPV (section 3.2) and DB-PPV (section 3.3) were less stable than that from the solution phase ECL of DPA. That is, when with DPA the triple potential steps were repeated, the intensity of the ECL remained the same for a number of cycles. Whereas in the case of conducting polymers the intensities were lower than that of the DPA ECL and also decreased by approximately 50 % within 10 cycles. Dissolution of the polymer was observed from the microscopic images of the polymer coated electrode surfaces before and after the ECL experiment. Dissolution has also been reported in the literature . To avoid that, as a first approach, cross linking of the polymer DB-PPV was done by subjecting the polymer to synchrotron radiation of appropriate energy (section 3.3.5). The microscopic investigation revealed that the cross-linked polymer was quite stable under electrochemical conditions. But the observation of ECL has become more difficult than that for the un-cross linked DB-PPV. The cyclic voltammograms showed that the counterionic transport in the cross-linked film has become difficult, both for its oxidation and reduction. Thus, ECL could not be observed with heavily cross-linked film, which in our case was 500 mJ/cm2. However, it could be observed, when the cross linking was done with the radiation energy 100-300 mJ/cm2 in the case of TEABF4 as the supporting electrolyte. But still, [page 108↓]the intensity was low and decreased faster than the solution phase ECL. So the possibility of a decomposition reaction in addition to the dissolution was considered.
Spectroscopic analysis was done to understand the cause and kind of the decomposition. Ex-situ Raman spectroscopic investigation was carried out on the DB-PPV polymer first. After that the film has been casted on the electrode, then after subjecting it to a synchrotron cross-linking procedure and finally after the ECL reaction. The Raman spectra of the polymer were the same in all the cases except after the ECL reaction. This shows that the polymer retains its chemical characteristics in the film casting and the synchrotron cross-linking procedure, but it becomes affected in the experimental procedures for ECL. Further analysis was done by subjecting the polymer-coated electrode to negative potential cycling and positive cycling independently. The cyclic voltammogram (CV) showed that DB-PPV is stable in the negative potential region over a number of potential cycles. Whereas in the positive potential region, the first CV itself is less reversible than the negative potential behavior and subsequent CVs showed more irreversible behavior, and a rapid decrease in the peak maximum for oxidation (ipa). Thus, the decomposition of the polymer is found to happen while subjecting the polymer to a positive potential sweep/step. These analyses concerning the stability of the polymer are the first ones in the study of ECL from the conducting polymer, and the results are going to be communicated .
The nature and size of the counterions affect the oxidation/reduction of the conducting polymers. It is found that TEABF4 or TBABF4 are better supporting electrolytes yielding more intense and reproducible ECLs than TEAClO4 or TBAClO4. This is likely to be due to the hygroscopic nature of the perchlorates, introducing traces of water into the electrolyte.
In addition, the size of the counterions also play a major role on the electrochemistry of CPs. These counterions are needed for maintaining charge neutrality in the CPs during electrochemical oxidation and reduction processes. Thus, the larger the counterion, the more difficult becomes its movement inside [page 109↓]the CP and, hence, the electrode reaction. In the experiment with a MEH-PPV film coated electrode, the polymer could be oxidized and reduced with TEABF4, but could not be reduced at all with TBABF4 due to the larger size of the TBA+ cation compared to the TEA+ cation, which is needed for charge compensation during reduction of the polymer (section 3.2.2).
However, the TBABF4 gave better results for the DB-PPV polymer. This is due to the difference in the molecular structure between MEH-PPV and DB-PPV. DB-PPV has two butoxy groups in the adjacent position, which produces more strain in the polymer due to steric hindrance. Thus, the DB-PPV film is less uniform than the MEH-PPV film having the two alkoxy substituents at the sterically favored opposite ends (section 3.3.5).
Even in the case of the DB-PPV films the counterionic mobility depends strongly on the degree of cross-linking. The film cross-linked with 500 mJ/cm2 energy of synchrotron radiation cannot be oxidized or reduced appreciably. Whereas the less cross-linked films (when the synchrotron energy is between 100-300 mJ/cm2), can become oxidized or reduced effectively (section 3.3.2).
Since the ECL process in the conducting polymers yields more difficulty than that in the solution phase, a quantitative study has never been carried out in the literature. This task has been undertaken for the first time in the present work. The diffusion-migration mode of charge transport in the CPs is approximated by a diffusion model, which implicitly takes into account the migration process by considering the electrode/polymer interface charging process (sections 4.2.1 - 4.2.4). Digital simulation technique has been used to solve the kinetic equations and to estimate the rate parameters.
Initial evaluations of the parameters for the simulation like, concentration of the active polymer sites, uncompensated resistance of the electrochemical system, double layer capacitance, and apparent diffusion coefficients are estimated from different appropriate experiments. First the simulation has been carried out for DPA ECL, and the rate constant was found out to be on the order of 1010 l mol-1 s-1, which is consistent with earlier theoretical studies of the solution phase [page 110↓]ECL. Then the simulation was performed for the MEH-PPV polymer, yielding a value of 103 l mol-1 s-1 for the rate constant. In the case of DB-PPV polymer it is 104 l mol-1 s-1. The low value of the ECL reaction rate constant in the conducting polymers compared to the solution phase has been analyzed based on the Marcus theory (section 4.2.5). It was concluded that the small rate constants are caused by the relatively larger internuclear distance between the positive and negative polarons in the activated complex.
Though the intensity maxima of the anodic and cathodic ECLs are similar in MEH-PPV and DB-PPV, the decay kinetics are strikingly different. This feature has been found to be associated with the nature of the charge compensating counterion in the anodic and cathodic ECLs. The apparent diffusion coefficients have been found to be on the orders of 10-13 cm2 s-1 for MEH-PPV and 10-11 cm2 s-1 for DB-PPV. Such a low value cannot be attributed to the electron drift mode of the charge transport. Thus, the electron transport in these two polymers under ECL experimental condition must be coupled to the movement of the counterions. This is another important conclusion from the present theoretical studies. Parts of the results have been published in the Synthetic Metals  and in the ACS Proceedings .
The intensity of the ECL from DB-PPV is found to be 100 times higher than that from MEH-PPV. In the earlier solid-state electroluminescence (EL) studies, this has been attributed to the stabilization of the excited state in DB-PPV by the strain introduced into the polymer chain by the ortho substitution (section 3.3.5). This strain yields the loss of co planarity between the polymer backbone and the side groups. Thus, excited state deactivation by the oxygen lone pair of butoxy groups on the PPV moiety was avoided.
Hence, ECL from DB-PPV has been chosen for analysis of the ECL emission spectra. Anodic and cathodic ECLs were recorded and were found to occur at the same wavelength (485 nm), i.e., the two modes of ECL generation yield the same excited state. This again becomes a proof that this annihilation reaction between positive and negative polarons is the only cause for the observation of ECL in DB-PPV. The ECL spectrum is identical with the solution phase [page 111↓]luminescence spectrum in CHCl3. Further comparison with solid-state photoluminescence (PL) and electroluminescence (EL) showed an insignificant shift of 25 nm. In fact, the π-π* emission band in the case of a conjugated network as the present polymer is known to show a weak environmental effect .
Further analysis can be carried out in the direction of studying the orientation effect of the conjugated bonds of the conducting polymers by choosing different deposition methods like electrodeposition, Langmuir-Blodget (LB) coating, vapor deposition and spin casting. The conjugated bonds can be oriented either parallel or perpendicular to electrode surfaces , which can affect the extension of the polarons and excitons in the CP. In situ FTIR studies can be done to study the exact mechanism of chemical deactivation of the conducting polymer during ECL reaction. From the theoretical point of view, migration of charges and counterions can be involved explicitly, and its effect can be analyzed on the ECL transient.
|© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.|
|DiML DTD Version 3.0||Zertifizierter Dokumentenserver|
der Humboldt-Universität zu Berlin