Intermolecular Polaron Formation Mechanism and ... - ACS Publications

Article pubs.acs.org/JPCA

Cite This: J. Phys. Chem. A 2018, 122, 461−469

Acid Doping of Phenyl-Capped Aniline Dimer: Intermolecular Polaron Formation Mechanism and Its Applicability to Polyaniline Olga E. Bogomolova and Vladimir G. Sergeyev* Chemistry Department, Lomonosov Moscow State University, 1-3, Leninskie Gory, Moscow, 119991, Russia S Supporting Information *

ABSTRACT: The acid doping process of the mixtures of N,N′-diphenyl-1,4-phenylenediamine (DPPD) and N,N′-diphenyl-1,4-quinonediimine (DPQD), reduced and oxidized phenyl-capped aniline dimers respectively, has been studied in detail by UV−vis−NIR and ESR spectroscopies. The addition of the acid to the mixtures of DPPD and DPQD leads to the formation of one type of radicals by intermolecular interactions. Full conversion of the aniline dimer molecules to the radical form is achievable only in stoichiometric mixtures of DPPD and DPQD. Concentration of radical species may change over time according to processes involving hydrolysis. The mechanisms of the acid doping of the phenyl-capped dimer process and subsequent side processes are proposed.

■

INTRODUCTION Polyaniline is the one of most studied conductive polymers due to its ease of synthesis and high environmental stability. Conductive form of polyaniline can be obtaned using redox chemical or electrochemical processes or by interaction with protonic acids through nonredox acid doping.1 Together with high and adjustable conductivity, these properties open a wide range of potential applications of polyaniline in electronic and electrochemical devices such as sensors, batteries, light-emitting diodes, capacitors, and photovoltaic devices. A number of different redox states of polyaniline are known: leucoemeraldine (completely reduced state), pernigraniline (completely oxidized state), emeraldine (half-oxidized state), and the intermediate states. The highest conductivity of polyaniline is achieved in the doped emeraldine (emeraldine salt) state, thus this redox state draws the most attention. Traditionally, polyaniline in emeraldine state is considered to be a polymer, consisting of alternating oxidized and reduced dianiline moieties:

exchange with electron pairs at amine nitrogen atoms of reduced dianiline moieties. However, it is reasonable to describe the emeraldine redox state of polyaniline as a dynamic block copolymer, containing leucoemeraldine, pernigraniline, and emeraldine domains, where leucoemeraldine and pernigraniline blocks are considered “defects” and only emeraldine domains are expected to be involved into charge carrier formation process.7 Moreover, taking into account globular conformation of polyaniline, there should be a number of short chain segments sterically isolated from intramolecular conjugation and bipolaron splitting. Considering that some parts of the polyaniline macromolecules are excluded from the intramolecular process of charge carrier generation, it is interesting to understand whether the polaron formation can be realized by intermolecular bipolaron splitting between isolated leucoemeraldine and pernigraniline moieties. Aniline oligomers serve as a good model for polyaniline due to similar electronic properties, better solubility in common solvents, and facile synthesis and characterization.8 Such model compounds help investigate such properties, which are difficult to study at the macromolecular level.9 Thus, using aniline trimers, it was proven that bipolarons could split not only intramolecularly (if allowed by the conjugation length) but also intermolecularly.10,11 Phenyl-capped aniline dimer, N,N′-diphenyl-p-phenylenediamine (DPPD) and its oxidized form, N,N′-diphenyl-pquinonediimine (DPQD), can be considered as simple model compounds for leucoemeraldine and pernigraniline states of polyaniline, respectively.12 DPPD and DPQD molecules are

The process of charge carrier formation during acid doping of emeraldine is usually described as follows: during the first stage, the imine nitrogen atoms of oxidized dianiline moieties become protonated, forming dications, which exist in tautomeric equilibrium with dication-diradicals (bipolarons). Then, bipolarons can split into paramagnetic cation radicals (polarons) located at the nitrogen atoms. These polarons present charge carriers, which are able to migrate along the polymer chain,2−4 or hop to other chains5,6 by electron © 2017 American Chemical Society

Received: October 5, 2017 Revised: November 30, 2017 Published: December 19, 2017 461

DOI: 10.1021/acs.jpca.7b09851 J. Phys. Chem. A 2018, 122, 461−469

Article

The Journal of Physical Chemistry A

volumes of the samples were determined using a calibration plot by measuring the height of the liquid column in a capillary taking into account density of solutions determined previously. To eliminate the influence of instrumental conditions during different measurement sessions, the ESR signal of the sample was double integrated, and the resulting number was divided by the amplitude of Mn2+ signal and compared with the result for the standard carbon sample spectrum in the same conditions. The molar concentrations of the radicals in the samples were calculated using the following equation:

able to interact in acidic media forming semiquinone radicals, similar to the polarons observed during acid doping of polyaniline.13 The stoichiometric mixture (SM) of DPPD and DPQD can model the emeraldine state of polyaniline because of the equal amount of amine and imine nitrogen atoms, while nonstoichiometric mixtures (non-SMs) can model intermediate redox states of polyaniline.

Crad(smpl) =

Due to the short conjugation length and the presence of only two nitrogen atoms in each molecule, intramolecular bipolaron splitting in aniline dimer is impossible. Thus, the study of DPPD, DPQD, and their mixtures allows us to understand the role of intermolecular bipolaron splitting in polaron generation during polyaniline acid doping. Here we describe a quantitative study of DPPD, DPQD, and their mixtures with protonic acids in organic solvents using UV−vis, FT-IR, and ESR spectroscopies and propose the mechanism of intermolecular polaron formation.

Int(smpl) AMn (smpl) Int(Cst) AMn (Cst)

×

n(Cst) V (smpl) × 6.02 × 1023

where Int(smpl) - result of double integration of the sample signal, AMn(smpl) - amplitude of Mn2+ signal average for the measurement session, Int(Cst) - result of double integration of the carbon standard sample signal, AMn(Cst) - amplitude of Mn2+ signal average for carbon standard measurement session, n(Cst) - number of the radicals in carbon standard sample, V(smpl) - volume of the sample, L All measurements were carried out at ambient temperature, in air. IR spectra were recorded with Thermo Scientific FT-IR “Nicolet IR 200” spectrometer with 2 cm−1 resolution. For the spectral measurements, dry samples were mixed with KBr and pressed into pellets, while liquid samples were dripped on a KRS-5 surface and dried. Acid Doping Procedure. Calculated volumes of concentrated solutions of DPPD, DPQD, acids, and solvent were mixed to achieve the required molar concentrations and ratios of components. The sample solutions were used for measurements immediately after preparation. The total concentration of DPPD, DPQD, or their mixtures in all samples was 0.005 M unless otherwise specified. In order to elucidate the effect of the mixing order, we compared samples prepared two ways: either by mixing the concentrated DPPD and DPQD solutions, followed by the addition of solvent and acid solution, or by mixing of DPQD with the solvent and acid solution, followed by the addition of DPPD solution. In the first case, the color of the sample changed from orange to green; in the second one the color changed from orange to red and then, after addition of DPPD, to green. The resulting spectra of both systems were identical, which confirms that the order of mixing components does not affect the reaction product (see Supporting Information 1). All samples of doped DPPD and PDQD mixtures were prepared by the first method. We have used camphorsulfonic acid (CSA) as a main doping agent because it is a strong acid easy soluble in organic solvents. The results obtained in different solvents (DMSO, DMF) with the use of different acids (CSA, DBSA) were nearly identical. In the absence of the acid, UV−vis−NIR spectra of various mixtures of DPPD and DPQD present a superposition of individual spectra of DPPD and DPQD, confirming that DPPD and DPQD do not react with each other at any ratio.

■

EXPERIMENTAL SECTION Materials. N,N′-diphenyl-p-phenylenediamine (DPPD) (Aldrich), the reduced form of phenyl-capped aniline dimer, was purified by sublimation before use. Dimethyl sulfoxide (DMSO), dimethylformamide (DMF), acetone, hydrochloric acid (Chimmed, Russia), aniline hydrochloride, ammonium peroxydisulfate, Ag2O, camphorsulfonic acid (CSA), and dodecylbenzensulfonic acid (DBSA) (Aldrich) were used without further purification, unless otherwise specified. N,N′-diphenyl-p-quinonediimine (DPQD), the oxidized form of phenyl-capped aniline dimer, was obtained by oxidation of N,N′-diphenyl-p-phenylenediamine with Ag2O in acetone:14 0.026 g of Ag2O was added to DPPD solution (0.02 g DPPD in 2 mL of acetone). The mixture was stirred for 1 h and filtered. The precipitate was washed with hot acetone to extract DPQD, and the resulting orange solution was concentrated using a rotary evaporator and dried in air at room temperature. DPQD structure was confirmed by IR11 and 13C NMR spectroscopy.15,16 Spectroscopy Methods. The UV−vis−NIR spectra of solutions were recorded with a Thermo Scientific “Helios α” spectrometer in quartz cells with an optical path of 1 cm, 0.1 or 0.01 cm, depending on the concentration of the sample. To determine the molar absorbance coefficients of DPPD and DPQD, a series of samples with molar concentrations in the range from 1 × 10−5 to 5 × 10−3 M were used. In all cases, the absorbance was found to increase linearly with the concentration. Molar absorption coefficient at a given wavelength was determined from a slope of linear fitted values of absorbance at this wavelength by the method of least-squares, passing through the origin, versus concentration. The molar absorption coefficient of DPPD at 310 nm is 30000 L/(mole· cm). The molar absorption coefficients of DPQD at 310 and 450 nm are 27420 and 7120 L/(mole·cm), respectively. ESR spectra were recorded with an RE-1307 ESR radio spectrometer using the ESR (version 4.0) software in a quartz capillary. For quantitative analysis, the carbon standard sample and Mn2+ signals in a MnO standard sample were used. The 462


Article


Figure 1. UV−vis−NIR spectra, experimental (solid lines) and calculated (dotted lines) (a), and ESR spectra (b) of solutions of stoichiometric mixtures of DPPD and DPQD mixed with CSA solution in molar ratios CSA/(DPPD+DPQD) from 0 to 150. The total concentration of DPPD and DPQD is 0.005 M, in DMF.

Figure 2. Dependencies of UV−vis−NIR absorbance at 395 and 720 nm (a), and radical content (b) on the molar ratio of CSA to stoichiometric mixture of DPPD and DPQD. The total concentration of DPPD and DPQD is 0.005 M, in DMF.

■

RESULTS AND DISCUSSION

Along with the changes in UV−vis−NIR spectra, the ESR signal appears and increases with the increase of the acid concentration in the system (Figure 1b). Such ESR spectrum is typical for cation radicals obtained by chemical or electrochemical single electron oxidation of DPPD.17,20,21The sevenline shape of the ESR spectrum can be explained by splitting the signal from one unpaired electron on two equivalent nitrogen nuclei and two equivalent protons, assuming that the hyperfine interaction constants are nearly identical for both groups of nuclei. It is important to note that due to the short conjugation length in aniline dimers, only the molecules with one unpaired electron could be detected by ESR spectroscopy. In the case of more than one unpaired electron (two), the spins of these electrons are likely to be opposite, which makes such molecules diamagnetic and undetectable by ESR spectroscopy. While the intensity of ESR signal depends on the molar ratio of the acid and the SM (DPPD-DPQD), the number of the lines and the correlation of their heights in the spectra remain constant in all samples (Figure 1b). This indicates that the different amounts of the same radicals are generated during the addition of CSA to the SM (DPPD-DPQD). The UV−vis−NIR absorbance and radical content (mole fraction) in the solutions of SM (DPPD-DPQD) mixed with the CSA in different molar ratios are presented in Figure 2a,b, respectively. The radical content in the samples was calculated as the ratio of the molar concentration of the radicals

Acid Doping of Stoichiometric Mixture of DPPD and DPQD (SM (DPPD-DPQD), Emeraldine Base Model). Addition of the protonic acid to the SM (DPPD-DPQD) leads to visible color change of the solution, from orange to green. In UV−vis−NIR spectra this change can be seen as decrease of the absorption bands corresponding to π − π* transition of phenylamine (310 nm) and quinoneimine (450 nm) groups of the initial DPPD and DPQD,9,12,17,18 with simultaneous appearance and growth of two new bands at 395 and 720 nm. These new bands are similar to 440 and 830 nm bands in the spectra of doped polyaniline,2,19 and correspond to polaron absorption, blue-shifted due to the shorter conjugation length in phenyl-capped aniline dimer relative to polyaniline12 (Figure 1a). The presence of isosbestic points in the spectra series (Figure 1a) indicates the quantitative transition between the pure initial and doped forms. The spectra of SMs of DPPD and DPQD doped with different molar excesses of acid present a superposition of the spectra of initial (without acid) and completely doped mixtures (see Supporting Information 2a for calculations). The similarity of calculated and experimental spectra presented in Figure 1a confirms the absence of byproducts of the acid doping process in the system. 463


Article


Figure 3. UV−vis−NIR and ESR spectra of the SM (DPPD-DPQD) doped with 50-fold CSA excess, measured right after mixing (in 1−5 min, solid lines) and after 10−15 min (dotted lines) (a), and the dependencies of the absorbance at 720 nm (solid lines) and radical content (dotted lines) on time of the SM (DPPD-DPQD), mixed with CSA at different molar ratios (b). The total concentration of DPPD and DPQD is 0.005 M, in DMF.

Figure 4. Dependencies of absorbance at 720 nm (a) and radical content (b) on the mole fraction of DPQD in the mixtures of DPPD and DPQD (i.e., on the mole fraction of stoichiometric mixture in nonstoichiometric mixtures - upper axis). The total concentration of DPPD and DPQD is 0.005 M, in DMF.

radical content is achieved right after mixing, we suggest that radical (polaron) formation occurs completely during the mixing and then decreases over time due to degradation processes that will be described below. The rate of these changes increases with the initial concentration of radicals in the samples, determined by the acid−to−SM (DPPD-DPQD) molar ratio (Figure 3b), which is in good agreement with the law of mass action (increase of the radical concentration leads to the increase of the rate of the reaction consuming this radical). Acid Doping of Nonstoichiometric DPPD and DPQD Mixtures (Non-SMs (DPPD-DPQD), Models of Intermediate Polyaniline Redox Forms). According to the results presented above (Figure 2), practically complete doping of stoichiometric mixtures of DPPD and DPQD is achieved at 50fold or higher molar excess of the acid. As we were mostly interested in the study of the doped state of nonstoichiometric mixtures of DPPD and DPQD, 50-fold acid excess was used. The absorption bands of both doped SM (DPPD-DPQD) and doped excessive component can be seen in the UV−vis− NIR spectra of non-SMs (DPPD-DPQD) (see Supporting Information, 2b). ESR spectra of the doped non-SM (DPPDDPQD) have the same 7-line shape as the spectra of the doped SM (DPPD-DPQD).

(calculated from the ESR spectra as described in the Experimental Section) to the sum of the initial molar concentrations of DPPD and DPQD in solution (0.005 M). It is clear that with the increase of acid concentration, the mole fraction of the radicals in the system increases to almost 1 (Figure 2b). The highest radical content is obtained in the samples containing 50-fold or higher acid excess. In these samples the absorbance peaks at 310 and 450 nm (corresponding to the initial DPPD and DPQD) disappear, while the intensity of the peaks at 395 and 720 nm reach their maxima (Figure 2a). This confirms that bands with maxima at 395 and 720 nm correspond to the radical (polaron) absorbance. Based on the obtained results, we can conclude that the stoichiometric mixture of DPPD and DPQD interacts with the protonic acid, forming only one type of radical. In the large excess of the acid, aniline dimer molecules convert fully to the radical state. This result is in a good agreement with the highest contents of charge carries in emeraldine in comparison with other redox states of polyaniline, which is the key to its conductivity.22 Both UV−vis−NIR and ESR spectra of the acid-doped SM (DPPD-DPQD) undergo changes within time: the intensity of ESR signal and the absorbance at 395 and 720 nm decreases simultaneously (Figure 3a). Based on the fact that the highest 464


Article


Figure 5. Dependencies of the absorbance at 720 nm (a) and the radical content (b) on time in mixtures of DPPD and DPQD. The total concentration of DPPD and DPQD is 0.005 M; the concentration of CSA is 0.25 M, in DMF.

Figure 6. UV−vis−NIR spectra of DPPD and DPPD-CSA mixtures (a), and the dependencies of the wavelength of maximum (squares) and absorbance at this wavelength (circles) on the molar ratio CSA/DPPD (b). The concentration of DPPD is 0.005 M, in DMSO.

Dependencies of absorbance at 720 nm (corresponding to the polaron absorption) and the radical content in doped nonSMs (DPPD-DPQD) on the mole fraction of DPQD (or on the mole fraction of stoichiometric mixture in nonstoichiometric mixtures - upper axis) are presented in (Figure 4). Both of them increase with the change of DPQD mole fraction from 0 to 0.5 and decrease with the change of DPQD mole fraction from 0.5 to 1. The maxima of both characteristics are observed at the mole fraction of DPQD about 0.5, i.e. in SM. Thus, the maximum content of the radicals can be achieved only in stoichiometric mixtures of DPPD and DPQD, while in nonstoichiometric mixtures the radical content corresponds to the content of SM (DPPD-DPQD). Considering non-SMs (DPPD-DPQD) as a SM plus the excess of one of the components, the absorption at 720 nm and the radical content are expected to increase and decrease with the mole fraction of SM (DPPD-DPQD) linearly. Deviations from the linearity observed in Figure 4 dealt with the processes which occur in doped non-SMs (DPPD-DPQD) with the rate less than that of radical formation. The dependencies of absorbance at 720 nm (corresponding to the polaron absorbance) and radical content in non-SMs (DPPD-DPQD) on time are presented in Figure 5. It is important to notice that only the intensity of the ESR signal changes with time, while the hyperfine structure of it remains the same. This fact suggests that only the

concentration of the radical species is changing, not their type, which allows one to calculate the radical content. In the mixtures containing an excess of DPPD, the timerelated changes are similar to ones observed in SMs (DPPDDPQD): the intensity of absorbance at 720 nm and the radical content decrease simultaneously, but the decrease is less pronounced (Figure 5a,b). Thus, we can say that only the SM (DPPD-DPQD) undergoes changes, and, according to the law of mass action in the systems with lower concentration of it (non-SMs with excess of DPPD), these changes occur with less rate. It is worth mentioning that, with the high excesses of DPPD (molar ratio DPQD/DPPD = 20/80 and less), the rate of the changes is lower than expected, which indicates that the excess of DPPD prevents the process leading to the decrease of the radical content in the system. In mixtures containing an excess of DPQD, the pattern of time-related changes is more complex: both intensity of absorbance at 720 nm and the radical content increase and then decrease. The duration of increase of both characteristics correlates with the excess of DPQD: in the mixtures with molar ratio DPQD/DPPD = 60/40, it is negligible, while in the mixtures with DPQD/DPPD = 80/20, it is quite pronounced (Figure 5a,b). Thus, we can suggest that the two distinct processes are competing in the excess DPQD. One process leads to the decrease of radical concentration, and another process leads to increase of radical concentration and is likely related to DPQD. 465


Article


Figure 7. UV−vis−NIR spectra of DPQD and DPQD-CSA mixtures with the molar ratio CSA/DPQD from 1 to 150 (a), and the dependencies of the absorbance of initial (at λ = 310, 450 nm) and doped (at λ = 395, 502, 530, 720 nm) forms of DPQD on the molar ratio CSA/DPQD (b). The concentration of DPQD is 0.005 M, in DMF; all spectra were measured immediately after sample preparation.

Figure 8. UV−vis−NIR spectra of DPQD doped with 50-fold CSA excess measured at various times after sample preparation (a), and the dependencies of the absorbance at λ = 395, 450, and 720 nm on time after sample preparation (b). The concentration of DPQD is 0.005 M, in DMSO.

Acid Doping of DPPD (Reduced Aniline Dimer, Leucoemeraldine Base Model). The UV−vis−NIR spectra of DPPD and its mixtures with CSA in different molar ratios are presented in (Figure 6a). Only one absorbance band observed in UV−vis−NIR spectrum of DPPD corresponds to the π−π* transition of the phenylamine group9,12,17,18 Addition of CSA to DPPD does not lead to the appearance of any new bands. However, with a significant excess of CSA, at molar ratios of CSA/DPPD > 50, a small (less than 10 nm) blue shift and decrease of intensity of the absorbance band are observed (Figure 6a,b). In order to eliminate the effect of overlapping CSA absorbance (∼290 nm), the CSA spectra were subtracted from the spectra presented in Figure 6a. The similar spectral changes, observed during the addition of the acid to N,N,N′,N′-tetramethyl-p-phenylenediamine23 and leucoemeraldine19 base originate from the protonation of amine nitrogen atoms. No ESR signal was observed in solutions of DPPD mixed with the acid. Thus, we can conclude that DPPD interacts with the acid only in case of a large excess of it (at molar ratios CSA/DPPD > 50) without any radical species formation. Acid Doping of DPQD (Oxidized Aniline Dimer, Pernigraniline Base Model). The UV−vis−NIR spectra of DPQD and its mixtures with CSA in different molar ratios are presented in Figure 7a.

The spectrum of DPQD exhibits two absorbance bands: one with a maximum at ∼310 nm corresponding to the π−π* transition of the phenylamine group (the same as observed in DPPD spectrum), and another one at 450 nm, corresponding to the absorbance of quinoneimine group9,12,17,18(molecular exciton24). The addition of CSA to DPQD solution results in the color change from orange to red and remarkable changes in UV− vis−NIR spectra. The intensity of the initial absorption bands decreases with the simultaneous appearance and increase in the intensity of the new bands. The absorbance with a maximum at about 500−530 nm (seen as a shoulder because of overlapping with the initial band) is associated with protonated DPQD.25,26 The intensity of this band rises along with the acid concentration and reaches a plateau at more than 50-fold acid excess (Figure 7b). We found that some processes occur in doped DPQD over time: the color of solutions of CSA-DPQD mixtures changes from red to green in several minutes after preparation and then changes to gray in 24−48 h. The absorbance bands with the peaks at about 395 and 720 nm appear, increase, and then decrease (Figure 8) simultaneously with the weak ESR signal during same period. The rate of these processes rises with the molar ratio CSA/DPQD. We suggest that two types of processes are taking place in the system: the faster processes occur during mixing, and slower 466


Article

The Journal of Physical Chemistry A processes occur over time after the mixing is complete, and discuss them in next paragraph. Acid Doping: The Scheme of the Process. We observed the generation of radical species only in the systems containing DPQD. The basicity of DPPD appeared to be lower than that of DPQD. At 50-fold molar excess of CSA, the DPQD was completely doped, while DPPD did not interact with acid at all (Figures 6 and 7). Here we propose the mechanism of processes taking place in solutions of DPPD, DPQD, and their mixtures with the 50-fold molar excess of CSA. The spectrum of DPQD doped with CSA (CSA/DPQD = 50) and measured right after mixing can be modeled as a superposition of Gauss peaks with maxima at 304, 390, 501, and 743 nm (see Supporting Information 3). The absorbance at about 501 nm corresponds to diprotonated DPQD 25 (DPQDH2+ 2 ). We propose that addition of acid to DPQD solution lead to sequential protonation of both imine nitrogen atoms. The second protonation step seems to be easier than the first one and occurs immediately after it, yielding the dication molecule DPQDH2+ 2 . This dication molecule can undergo electron redistribution with restoration of the aromaticity of the central ring and formation of the dication−diradical (bipolaron) DPQDH2+• 2 molecule. Dication and bipolaron forms of doped DPQD can exist in tautomeric equilibrium (Scheme 1). Both protonation of DPQD and redistribution of electrons with bipolaron formation are fast processes and occur during mixing.

Scheme 2. Interaction of Bipolaron Localized on Protonated DPQD with DPPD

species achievable right after sample preparation is twice as high as the quantity of the deficit component molecules (Figure 4a). In the case of DPQD excess, as in the case of pure DPQD, the increase of radical concentration in time, followed by the decrease of it is observed (Figure 5 and Figure 8). We propose the following explanation of this fact. Protonated quinone imines are known to be unstable and subjected to degradation by hydrolysis27−29 in the acidic media. As DMSO and DMF solvents are highly hygroscopic, they are likely to contain traces of water. Thus, the changes observed in doped DPQD solutions over time (Figure 8) could be caused by the hydrolysis of protonated imine groups with water absorbed by the solvent from the air and further reactions of the products. We suggest that the hydrolysis of diprotonated DPQD leads to aniline and quinone formation (Scheme 3), which is in a good agreement with the schemes proposed for other pquinonediimines30−33

Scheme 1. Protonation of DPQD with Further Electron Redistribution Leading to Bipolaron Formation

Scheme 3. Hydrolysis of DPQDH2+ 2

The resulting aniline can interact with diprotonated DPQD the same way as DPPD (see Scheme 2) and reduce it to the semiquinone radical. This is confirmed by the appearance of the absorbance peak with maxima at about 395 and 720 nm and the ESR signal, which is characteristic to semiquinone radical, in doped DPQD over time (Figure 8). The second product of this interaction, the aniline cation radical, can interact with pbenzoquinone resulting from the hydrolysis as described in ref 34. In order to confirm the hypothesis about the role of the hydrolysis, we studied the process of acid doping of DPQD with increased water content and with reduced oxygen content. Addition of water to the system leads to the increased rates of radical formation and degradation, which confirms the hydrolytic nature of the process (see Supporting Information, 5). Reducing oxygen concentration does not affect the result considerably. We also tested the effect of aniline addition to DPQD in acidic media and found that it leads to an increase of the radical

In the case of the presence of DPPD in solution, an interaction between the cation-radicals localized on DPQDH2+• 2 molecule and the lone pairs of electrons on the amine nitrogen atoms of DPPD takes place through π-complex formation. One electron transfer from DPPD to protonated DPQDH2+• leads 2 to the formation of two molecules, each bearing cation radical (semiquinone radical, polaron) (Scheme 2). These molecules (DPPD+•) are identical, so the acid doping of DPPD and DPQD mixtures leads to the formation of only one type of charge carrier. This process occurs during sample preparation. It is worth noting that DPPD and DPQD interacts with each other only in the presence of acid, i.e., there is no complex formation before protonation (see Supporting Information, 1). In the case of SM (DPPD-DPQD), the process described above leads to almost complete conversion of aniline dimers to the radical form right after sample preparation (Figure 2) but then the concentration of radical species decreases with the kinetics of the first order (see Supporting Information, 4). In nonstoichiometric mixtures, the maximum quantity of radical 467


Article


emeraldine-to-pernigraniline states (see Supporting Information, 10). We have found that the addition of the acid to the polyaniline-DPPD mixtures leads to the DPPD+• semiquinone radical appearance, while in the absence of the acid there is no interaction between polyaniline and DPPD. This indicates that DPPD−polyaniline interaction in the acidic media is similar to DPPD−DPQD interaction. It includes the stage of intermolecular electron transfer from DPPD to protonated pernigraniline domain of polyaniline and leads to the increase of the concentration of the radicals in the system. This kind of interaction opens a possibility to influence the oxidation state of polyaniline during acid doping by the addition of DPPD or DPQD, increase the charge density and the associated conductivity in the system. We have found that the reduction of oxidized polyaniline by the interaction with DPPD really takes place, but the effect is not very pronounced (see Supporting Information, 10), presumably due to the supramolecular polyaniline structure, steric hindrance, and effective electron delocalization in protonated pernigraniline domains of polyaniline.

concentration, confirming that aniline can interact with diprotonated DPQD (see Supporting Information, 6). The UV−vis−NIR spectra of the solution of aniline, pbenzoquinone, and CSA mixed in concentrations and ratios that are expected as a result of hydrolysis, and doped DPQD solution stored for 24 h, correspond with the results described in ref 34 and prove the proposed mechanism (see Supporting Information, 7). Described hydrolytic and associated processes are slower than the processes of protonation and electronic redistribution, and lead to the formation of semiquinone radicals minutes after sample preparation in doped DPQD and non-SMs (DPPDDPQD) with excess DPQD. The further decrease of the semiquinone radical concentration until their complete depletion, seen in doped DPQD and non-SMs (DPPD-DPQD) with DPQD excess as the decrease of characteristic UV−vis−NIR absorbance and ESR signal, can be explained as follows. Basing on ref 13, we can propose that the semiquinone radical exists in equilibrium with diprotonated DPQD and DPPD. DPQDH 22 + + DPPD ⇔ 2DPPD+•

■

(1)

CONCLUSIONS The acid doping process of phenyl-capped aniline dimer has been studied in detail. The addition of the acid to the mixtures of DPPD and DPQD leads to the formation of one type of radicals by intermolecular interactions. Full conversion of the aniline dimer molecules to the radical form is achievable only in stoichiometric mixtures of DPPD and DPQD. Concentration of radical species may change in time according to processes dealt with hydrolysis. The mechanisms of the acid doping of the phenyl-capped dimer process and subsequent side processes are proposed. It was proven that in a system consisting of reduced and oxidized molecules, intermolecular interactions are uniquely responsible for the formation of the number of charge carriers, which is equal to the number of molecules, i.e., the mixture of oxidized and reduced phenyl-capped aniline dimers behave as polyaniline or longer oligomers in the emeraldine base state. Based on the obtained results, and taking into account the concept that polyaniline is a dynamic block copolymer, we conclude that the oxidized and reduced domains of polyaniline can be involved in the process of charge carrier formation during acid doping and thus contribute to total conductivity by interdomain (inter- or intramolecular) interactions.

This equilibrium is strongly shifted to semiquinone radical (DPPD+•) formation (see Supporting Information, 8), but the concentration of DPQDH2+ 2 does not equal zero. We propose that the semiquinone radical DPPD+• originally generated from the interaction of DPQDH2+ 2 with aniline also exists in equilibrium with DPQDH2+ 2 and DPPD: DPQDH 22 + + aniline ⇒ DPPD+• ⇔ DPQDH 22 + + DPPD

The decrease in concentration of DPQDH2+ 2 due to hydrolysis, while DPPD+• is stable,32 shifts the equilibrium (eq 1) to the left and leads to semiquinone radical decomposition in doped SM (DPPD-DPQD), doped non-SMs (DPPD-DPQD) with DPQD excess, and doped DPQD. In the case of doped non-SMs (DPPD-DPQD) with excess DPPD, the maximum radical concentration is achieved right after preparation, and only slow decrease of radical concentration is observed over time. The decreased rate of radical degradation process in these mixtures compared to SM (DPPD-DPQD) (Figure 5) can be explained by the additional shift of the equilibrium (1) toward the semiquinone radical formation (to the right), due to the law of mass action. The necessity of a big molar excess of the acid (≥50-fold) for complete doping of DPQD and its mixtures with DPPD in comparison with polyaniline can be explained by decreased basicity of DPPD and DPQD compared to polyaniline segments and other oligomers. It can be caused by short conjugation length and phenyl-capping of both sides of the molecules. An investigation of the acid doping process of SM (DPPD-DPQD) in different concentration ranges and calculation the equilibrium constant are presented in Supporting Information, 9. Interchain Interactions in Polyaniline Doping Process. Considering the mixtures of DPPD and DPQD as models for polyaniline as a dynamic block copolymer, we suppose that pernigraniline and leucoemeraldine domains could be involved in polaron formation during acid doping by interchain interaction. As the next step of the model, we have studied acid doping of the mixtures of DPPD with polyaniline in intermediate

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.7b09851. Additional results and calculations mentioned in the article (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Phone: +7 495 9393124. ORCID

Olga E. Bogomolova: 0000-0003-3608-6566 Vladimir G. Sergeyev: 0000-0002-8200-5993 Notes

The authors declare no competing financial interest. 468


Article


■

(19) Huang, W.; MacDiarmid, A. Optical Properties of Polyaniline. Polymer 1993, 34, 1833−1845. (20) Grossmann, B.; Heinze, J.; Moll, T.; Palivan, C.; Ivan, S.; Gescheidt, G. Electron Delocalization in One-Electron Oxidized Aniline Oligomers, Paradigms for Polyaniline. A Study by Paramagnetic Resonance in Fluid Solution. J. Phys. Chem. B 2004, 108, 4669−4672. (21) Rapta, P.; Vargová, A.; Polovková, J.; Gatial, A.; Omelka, L.; Majzlík, P.; Breza, M. A Variety of Oxidation Products of Antioxidants Based on N,N′-Substituted p-Phenylenediamines. Polym. Degrad. Stab. 2009, 94, 1457−1466. (22) Focke, W.; Wnek, G.; Wei, Y. Influence of Oxidation State, pH, and Counterion on the Conductivity of Polyaniline. J. Phys. Chem. 1987, 91, 5813−5818. (23) Steigman, J.; Cronkright, W. Acid-Base Reactions of N,N,N′,N′Tetramethyl-p-Phenylenediamine with Carboxylic Acids in Benzene at 25. Deg. J. Am. Chem. Soc. 1970, 92, 6729−6736. (24) Duke, C.; Conwell, E.; Paton, A. Localized Molecular Excitons in Polyaniline. Chem. Phys. Lett. 1986, 131, 82−86. (25) Cauquis, G.; Delhomme, H.; Serve, D. L’Oxidation Anodique de Quelques N-phenyl p. Phenylenediamines en Milieu Acetonitrile. Tetrahedron Lett. 1972, 13, 1965−1968. (26) De Santana, H.; Quillard, S.; Fayad, E.; Louarn, G. In Situ UVvis and Raman Spectroscopic Studies of the Electrochemical Behavior of N,N′-Diphenyl-1,4-Phenylenediamine. Synth. Met. 2006, 156, 81− 85. (27) Michaelis, L.; Schubert, M. P.; Granick, S. The Free Radicals of the Type of Wurster’s Salts. J. Am. Chem. Soc. 1939, 61, 1981−1992. (28) Carey, F. A., Sundberg, R. J. Advanced Organic Chemistry Part A: Structure and Mechanisms., 5th ed.; Springer Science+Business Media, LLC: Berlin, 2007; pp 1−1212. (29) Leedy, D.; Adams, R. Anodic Oxidation Pathways of Aromatic Amines. IV. Diphenylamine Systems in Aqueous Acid Solution. J. Am. Chem. Soc. 1970, 92, 1646−1650. (30) Tong, L. K. J. Kinetics of Deamination of Oxidized N,NDisubstituted p-Phenylenediamines. J. Phys. Chem. 1954, 58, 1090− 1097. (31) Corbett, J. F. Benzoquinone Imines. Part II. Hydrolysis of pBenzoquinone Monoimine and p-Benzoquinone Di-Imine. J. Chem. Soc. B 1969, 0, 213−216. (32) Nickel, U.; Kemnitz, K.; Jaenicke, W. Kinetics and Mechanism of the Acid Deamination of N-Substituted Quinone Di-Imines Measured With a Multi-mixing, Stopped-Flow Technique. J. Chem. Soc., Perkin Trans. 2 1978, 0, 1188−1193. (33) Nickel, U.; Jaenicke, W. Activation Parameters and Mechanism of the Deamination of N-Substituted Quinone Monoimines and DiImines. J. Chem. Soc., Perkin Trans. 2 1980, 0, 1601−1605. (34) Stejskal, J.; Bober, P.; Trchová, M.; Horský, J.; Pilař, J.; Walterová, Z. The Oxidation of Aniline with p-Benzoquinone and Its Impact on the Preparation of the Conducting Polymer, Polyaniline. Synth. Met. 2014, 192, 66−73.

ACKNOWLEDGMENTS This study is largely based on the ideas that appeared during a discussion with our colleague Vladimir B. Golubev. V.G.S. acknowledges financial support from the Russian Science Foundation (Project No 17-73-30006).

■

REFERENCES

(1) MacDiarmid, A.; Yang, L.; Huang, W.; Humphrey, B. Polyaniline: Electrochemistry and Application to Rechargeable Batteries. Synth. Met. 1987, 18, 393−398. (2) Stafström, S.; Bredas, J.; Epstein, A.; Woo, H. S.; Tanner, D. B.; Huang, W. S.; MacDiarmid, A. G. Polaron Lattice in Highly Conducting Polyaniline: Theoretical and Optical Studies. Phys. Rev. Lett. 1987, 59, 1464−1467. (3) Epstein, A.; Ginder, J.; Zuo, F.; Woo, H.; Tanner, D.; Richter, A. F.; Angelopoulos, M.; Huang, W.-S.; MacDiarmid, A. G. Insulator-toMetal Transition in Polyaniline: Effect of Protonation in Emeraldine. Synth. Met. 1987, 21, 63−70. (4) Furukawa, Y. Electronic Absorption and Vibrational Spectroscopies of Conjugated Conducting Polymers. J. Phys. Chem. 1996, 100, 15644−15653. (5) Wudl, F.; Angus, R. J.; Lu, F.; Allemand, P.; Vachon, D.; Nowak, M.; Liu, Z.; Schaffer, H.; Heeger, A. Poly-(p-Phenyleneamineimine): Synthesis and Comparison to Polyaniline. J. Am. Chem. Soc. 1987, 109, 3677−3684. (6) Krinichnyi, V. I.; Chemerisov, S. D.; Lebedev, Y. S. EPR and Charge-Transport Studies of Polyaniline. Phys. Rev. B: Condens. Matter Mater. Phys. 1997, 55, 16233−16244. (7) Shimano, J. Y.; MacDiarmid, A. G. Polyaniline, a Dynamic Block Copolymer: Key to Attaining Its Intrinsic Conductivity? Synth. Met. 2001, 123, 251−262. (8) MacDiarmid, A.; Zhou, Y.; Feng, J. Oligomers and Isomers: New Horizons in Poly-Anilines. Synth. Met. 1999, 100, 131−140. (9) Boyer, M.; Quillard, S.; Rebourt, E.; Louarn, G.; Buisson, J. P.; Monkman, A.; Lefrant, S. Vibrational Analysis of Polyaniline: A Model Compound Approach. J. Phys. Chem. B 1998, 102, 7382−7392. (10) Lokshin, N.; Pyshkina, O.; Golubev, V. B.; Sergeyev, V. G.; Zezin, A. B.; Kabanov, V. A.; Levon, K.; Piankijsakul, S. Intermolecular Electron Transfer in Low-Molecular-Weight Polyaniline Models Associating on Protonation by Amphiphilic Acid in Organic Solvent. Macromolecules 2001, 34, 5480−5486. (11) Zhang, Q.; Khajo, A.; Sai, T.; de Albuquerque, I.; Magliozzo, R. S.; Levon, K. Intramolecular Transport of Charge Carriers in Trimeric Aniline upon a Three-Step Acid Doping Process. J. Phys. Chem. A 2012, 116, 7629−7635. (12) Boyer, M. I.; Quillard, S.; Louarn, G.; Froyer, G.; Lefrant, S. Vibrational Study of the FeCl3-Doped Dimer of Polyaniline; A Good Model Compound of Emeraldine Salt. J. Phys. Chem. B 2000, 104, 8952−8961. (13) Linschitz, H.; Rennert, J.; Korn, T. Symmetrical Semiquinone Formation by Reversible Photoöxidation and Photoreduction. J. Am. Chem. Soc. 1954, 76, 5839−5842. (14) Honzl, J.; Metalova, M. Polyaniline Compounds-III Reaction of Phenyl- and Methyllithium with N,N′-Diphenyl-p-Quinone Diimine. Tetrahedron 1969, 25, 3641−3652. (15) Hagiwara, T.; Yamaura, M.; Iwata, K. Structural Analysis of Deprotonated Polyaniline by Solid-state 13C N.M.R. Synth. Met. 1988, 26, 195−201. (16) Han, C.-C.; Balakumar, R.; Thirumalai, D.; Chung, M.-T. The Different Electronic Natures Displayed by the Alkylthio Groups in Simple and Higher Conjugated Aniline Systems. Org. Biomol. Chem. 2006, 4, 3511−3516. (17) Glarum, S.; Marshall, J. Electron Delocalization in Poly(aniline). J. Phys. Chem. 1988, 92, 4210−4217. (18) Linschitz, H.; Ottolenghi, M.; Bensasson, R. The One-Electron Oxidation of Triplet Diphenyl-p-Phenylenediamine by the Diimine. J. Am. Chem. Soc. 1967, 89, 4592−4599. 469


Intermolecular Polaron Formation Mechanism and ... - ACS Publications

Recommend Documents