Raman and in Situ FTIR-ATR Characterization of Polyazulene Films and Its Derivate
Beatriz Meana-Esteban,†,‡ Cecilia Lete,§ Carita Kvarnstro1m,*,† and Ari Ivaska†
Process Chemistry Centre, c/o Laboratory of Analytical Chemistry, Åbo Akademi UniVersity, FIN-20500Turku-Åbo, Finland, Graduate School of Materials Research, Åbo Akademi UniVersity, FIN-20500 Turku-Åbo,Finland, and Institute of Physical Chemistry “I. G. Murgulescu” of Romanian Academy, Bucharest, Romania
ReceiVed: May 24, 2006; In Final Form: September 5, 2006
Polymers from azulene (A) and 2-[(E)-2-azulen-1-ylvinyl]thiophene (B) electrochemically synthesized arematerials with a broad absorbance in the UV-vis spectral region. An experimental approach to correlate theRaman and in situ FTIR spectra from azulene based polymers according to the effective conjugation coordinatetheory (ECC) is presented. Film characterization was made by Raman and Fourier transform infrared attenuatedtotal reflectance, FTIR-ATR spectroscopy. Throughout the whole workA was used as a model compound.The polymers were synthesized at different polymerization potentials in order to create different structures.Polyazulene showed a divergent Raman response upon change in excitation wavelength,λexc) 514 nm andλexc) 780 nm, in comparison to common conducting polymers. The FTIR-ATR measurements were madeduring charging-discharging of the polymers. The IR spectra of the conducting state show new doping inducedinfrared active vibrations (IRAV) in the region between 1600 and 700 cm-1 and a broad electronic absorptionin the high energy range (4000-8000 cm-1). Two different structures of the polymer fromB are formed, andboth follow the trends for conducting polymers upon charging.
1. Introduction
Infrared and Raman spectroscopy can be used as comple-mentary techniques to study not only the structural but also theelectronic properties of conjugated polymers. In situ Fouriertransform infrared-attenuated total reflectance (FTIR-ATR)spectroscopy has been used for the study of changes in theconjugated polymer films during charging-discharging reactions(p- and n-doping). The infrared spectra of conjugated polymersexhibit upon doping new and very intense infrared activevibrations (IRAV) bands typically in the low energy range(1600-700 cm-1) due to the strong electron-phonon couplingwithin the molecule. The characteristics of the IRAV bands arepolymer specific; i.e., the localization of the doping inducedcharge carriers formed in the material upon charging. In addition,IRAV bands are accompanied by a broad electronic absorptionwith maximum at higher energies (1600-8000 cm-1) due toelectronic transitions involving new quasiparticles states in theband gap.1
Several theoretical models have been developed in order toexplain the strongly enhanced infrared active vibrational (IRAV)modes. Horovitz et al. presented the first model using theamplitude mode (AM) formalism to explain the IRAV bandsin doped polyacetylene.2,3 The AM theory was later reformulatedin terms of effective conjugation coordinate theory (ECC) byZerbi and co-workers.4,5 They showed the correlation of theIRAV bands with infrared activation of totally symmetric Ramanactive vibrational Ag modes which contain a contribution bythe “effective conjugation coordinate”. In this model theeffective conjugation coordinate,Я, basically describes the
geometry change of the polymer skeletal atoms in going fromthe ground (benzenoid) to the excited (quinoid) state. Theoriginally infrared inactiveЯ mode of the neutral polymer filmbecomes strongly infrared active due to lowering of the electricalsymmetry caused by the polarization of the bonds upon doping.Ehrenfreund and Vardeny6 correlated the doping inducedelectronic states within the semiconducting electronic band gapand the IRAV bands. Their model is based on a linear responsetheory developed by Soos and co-workers.7 The commonconcept in the three models is the strong link between theeffective conjugation length in the macromolecule, the delo-calization of the doping induced quasiparticules, and thesignature of the IRAV modes strongly enhanced upon doping.
Raman spectroscopy is very useful for the description andbetter understanding of the structural and vibrational propertiesand the nature of the IRAV bands in conjugated polymers.
This work is an extension of our previous study concerningthe electrosynthesis, redox characterization, and in situ UV-vis characterization of films obtained from azulene (A) and2-[(E)-2-azulen-1-ylvinyl]thiophene (B) at different polymeri-zation potentials.8 Azulene and azulene compounds have in theirground state a characteristic charge distribution between thefused 7- and the 5-ring. This is influencing the charging responseusually obtained from conjugated material. In this work, weapply Raman and FTIR-ATR spectroscopy on azulene contain-ing conjugated systems in order to study the spectral responseupon charging and from charge distribution in the azulenecontaining structures.
2. Experimental Section
2.1. Chemicals.The electrochemical synthesis of azulene (A)and 2-[(E)-2-azulen-1-ylvinyl]thiophene (B) was performed in10 mM solution of the monomer material containing 0.1 Mtetrabutylammonium hexafluorophosphate (TBAPF6, 99%, Flu-ka) in acetonitrile (ACN, Lab-Scan). The monomer materials,
* To whom correspondence should be addressed. Tel.:+358-2-215-4419. Fax: +358-2-215-4479. E-mail: [email protected].
† Process Chemistry Centre, Åbo Akademi University.‡ Graduate School of Materials Research, Åbo Akademi University.§ Institute of Physical Chemistry “I. G. Murgulescu” of Romanian
azulene (A) (Aldrich) and 2-[(E)-2-azulen-1-ylvinyl]thiophene(B) were used as received. The synthesis and purificationmethods ofB are described elsewhere.9 The molecular structuresof the monomers are shown in Scheme 1.The arrows indicatethe position of highest spin density in the monomers. Theelectrolyte salt, TBAPF6, was dried at 80°C for 1 h undervacuum before use. Acetonitrile was stored over CaH2, freshlydistilled, and dried over basic alumina (Aldrich).
2.2. Instrumentation. FTIR Spectroscopy.The infraredspectra were recorded with a FTIR spectrometer (Bruker,IFS66S) using an MCT detector. A Harrick’s beam condenser4XF-BR3 served as the attachment to the FTIR spectrometer.The spectroelectrochemical measurements were carried out ina small-size ATR-spectroelectrochemical three-electrode flowcell made from Teflon. The experimental setup for the in situFTIR-ATR technique has been described earlier.10 The reflectionelement, a ZnSe crystal (size 10× 10 × 2 mm3) coated with athin evaporated layer of platinum (30 nm) served as the workingelectrode, and a Pt foil was used as the counter electrode. Theelectrochemically active area of the ZnSe reflection element was0.5 cm2. Before sputtering with Pt, the reflection element waspolished with diamond paste (1 and 0.25µm), rinsed withacetone, and cleaned in a plasma cleaner (Harrick) for at least15 min. A Ag/AgCl wire was used as a quasi-reference electrodethat was calibrated versus the ferrocene/ferrocinium couple(Eredox ) (Eox + Ered)/2 ) +0.35 V in TBAPF6-ACN). All thedata reported in this work are, if not differently stated, measuredagainst this reference electrode. The polymer films studied byFTIR spectroscopy were polymerized electrochemically directlyon the surface of the reflection element in the spectroelectro-chemical cell using Autolab PGSTAT 100 potentiostat usingthe general purpose electrochemical system (GPES) software.
Raman Spectroscopy.Raman spectra were recorded directlyfrom a Pt working electrode at room temperature with aRenishaw ramascope (system 1000 B) equipped with a LeicaDMLM microscope and connected to a CCD-camera as detector.The excitation wavelength for the Raman scattering wasprovided by a diode laser (λexc ) 780 nm) and by LaserPhysicsAr ion laser (λexc ) 514 nm). Spectra were collected at 180° tothe excitation beam. The polymer films studied by Ramanspectroscopy were polymerized electrochemically directly onthe surface of the Pt working electrode in a conventional three-electrode one-compartment cell at room temperature (T ) 23°C). The cell was connected to an Autolab PGSTAT 100potentiostat using GPES software. A Pt disk (area: 0.07 cm2)was used as the working electrode and a Pt wire was used asthe counter electrode. A Ag/AgCl wire was used as a quasi-reference electrode.
2.3. Procedure.For Raman and in situ FTIR experimentsthe polymerization of azulene (A) and 2-[(E)-2-azulen-1-ylvinyl]-
thiophene (B) was performed in ACN containing 0.1 M TBAPF6
and 10 mM of the respective monomer by potentiodynamiccycling, in the potential ranges between-0.6 and 1.2 V for 24times (procedure I) and between-0.6 and 1.8 V for 12 times(procedure II), at 50 mV/s scan rate. The amount of charged(mC) using during polymerization was approximately the same.A more detailed description of the electrosynthesis has beenreported earlier.8 In the in situ FTIR-ATR measurements, thespectroelectrochemical cell was rinsed with monomer-freesolution after electropolymerization. The doping process of thefilm was studied by recording the FTIR spectra in situ duringslow potential scanning, 5 mV/s. For each spectrum, 32interferograms were co-added covering a range of about 80 mVin the redox experiments at 5 mV/s; the resolution was 4 cm-1.The spectra are related to a reference spectrum, recorded atpotentials just before the studied redox process takes place andwhere no Faradaic reaction occurs. In all ATR-FTIR experi-ments, the solutions were deareated with dry nitrogen prior tomeasurements after which the cell was sealed with paraffin.
3. Results and Discussion
3.1. Raman Characterization.Figure 1a shows the Ramanspectra of dry azulene films after polymerization (A) accordingto procedure I and (B) according to procedure II and measured
SCHEME 1
Figure 1. Raman spectra of dry azulene films polymerized accordingto procedure I (A) and II (B) and measured with (a) 514 nm and (b)780 nm laser excitation wavelengths. The insert shows the UV-visabsorption spectra of the corresponding films on TO glass electrode.
23344 J. Phys. Chem. B, Vol. 110, No. 46, 2006 Meana-Esteban et al.
with the excitation lineλexc ) 514 nm. The Raman spectra ofthe same films but measured with the excitation lineλexc )780 nm are shown in Figure 1b. The insert shows the UV-visabsorption spectra of the corresponding films in their neutralstate (Figure 1a) and in their doped state (Figure 1b). The arrowsindicate the excitation line of the laser used for the Ramanmeasurements. As can be seen in both figures, the Ramanspectra of films electrosynthesized at the two different polym-erization potentials do not show remarkable structural differ-ences, indicating that the different polymerization conditionsdo not have any significant effect on the material formed. Bothfilms, however, are very sensitive to the excitation wavelength,and big differences can be seen between spectra measured usingλexc ) 514 nm andλexc ) 780 nm. The reason is that thevibrations originating from either quinoid or neutral units willbe differently resonance enhanced by the applied excitation lines.Excitation with theλexc ) 514 nm enhances the resonance effectof the Raman lines originating from the neutral segments ofthe film, while excitation atλexc ) 780 nm provides informationof vibrations associated with the quinoid units of the film. Thisis in good agreement with the UV-vis results.8 The Ramanspectrum measured withλexc ) 514 nm of the azulene filmsconsists of two broad bands at 1575 and 1363 cm-1 (Figure1a). With this excitation wavelength the main contributionshould be from the neutral form of polyazulene. Comparisonwith the UV-vis spectrum, however, shows that a slightexcitation of the material might take place already at thiswavelength. According to Zerbi et al.,11 in general, a planarstructure of sp2 carbon atoms (poly(azulene) falls into thiscategory) has two strong Raman modes in their neutral state.In addition, some similarities have been found between this classof molecules and disorder graphite material which has twoRaman peaks at 1580 and 1330 cm-1. These peaks can becorrelated to the so-called G and D bands of disorderedgraphite.11
On the other hand, it can be expected that the vibrationsoriginating from the quinoid units of the films should beresonance enhanced by theλexc ) 780 nm. The spectra of thepoly(azulene) films (λexc ) 780 nm) (Figure 1b) show a morecomplicated pattern with bands localized at 1494, 1343, 1177,1075 1033, 825, and 725 cm-1. The fact that the polyazulenefilms measured withλexc ) 780 nm show a more resolvedspectrum (has more bands) than the same films measured withλexc ) 514 nm should be pointed out. According to the theoryof Zerbi et al.,4,5,12 the change in the excitation wavelengthshould cause dispersion in the frequency and intensity of theRaman spectra. In general, this would cause the broadening and/or the vanishing of some bands in the Raman spectra. In thecase of polyazulene, the contribution from the quinoidic partsin the structure strongly dominates the spectrum. A possibleexplanation could be that in the excited form of polyazulenethe intralayer interactions decreases, resulting in phonon disper-sion.13 This gives rise to the more complicated pattern observedin the spectra of polyazulene films measured withλexc ) 780nm.
The fact that the resonance enhancements of the quinoidvibrations in the polymer films are seenλexc ) 780 nm is ingood agreement with the UV-vis absorbance for the sameelectrochemically doped film showing an increasing absorptionband starting from 750 nm.8
Figure 2a shows the Raman spectra of dry films frommonomerB after polymerization (A) according to procedure Iand (B) according to procedure II and measured with theexcitation lineλexc ) 514 nm. The insert shows the correspond-
ing films in their neutral state. The arrow indicates the excitationline of the laser used for Raman measurements. According tothe UV-vis spectra8 the use of this excitation line will result,as in the case for polyazulene, in an enhancement of the Ramanmodes that originated from the neutral segments of the polymer.Bands originating from the different substituents can clearly beidentified in both spectra. Bands at 1600 and 1220 cm-1 canbe assigned to the symmetric stretching of CdC and symmetricbending mode of the C-H bonds in the vinylene groups.14,15
Generally, three characteristic ring stretching bands can beexpected for thiophenes.16 In the spectra of the films in Figure2a, thiophene related bands can be seen at 1538 and 1460 cm-1
from the asymmetric and symmetric stretching modes of CdCbonds in thiophene rings. In addition, the band associated tothe intra-ring C-C stretching has been reported to be between1450-1365 cm-1. In our spectra the bands originating fromthis stretching can be found either at 1450 or 1370 cm-1. Thesymmetric bending of the C-H groups inâ-position in thethiophene ring can be seen as a doublet at 1089 and 1046 cm-1.The four characteristic Raman bands described above forthiophenes are termed by Zerbi et al.4,17 as lines A, B, C, D,respectively. Line B, found in the spectra of the films at 1460cm-1, is always the strongest in the spectrum, even for shortconjugation length.17 This line B together with the line D can
Figure 2. Raman spectra of dry 2-[(E)-2-azulen-1-ylvinyl]thiophenefilms polymerized according to procedure I (A) and II (B) and measuredwith (a) 514 nm and (b) 780 nm laser excitation wavelengths. Theinsert shows the UV-vis absorption spectra of the corresponding filmson TO electrode.
FTIR-ATR Characterization of Polyazulene Films J. Phys. Chem. B, Vol. 110, No. 46, 200623345
be associated with the collective totally symmetric vibrationdescribed by theЯ mode in the ECC theory. Moreover,vibrations from C-H deformation vibration and C-S stretchingvibration can be observed in the spectra of the films at 940cm-1 and at 690 cm-1, respectively. Band characteristics fromthe azulene unit can be found in the Raman spectra of bothfilms at 1565 and 1301 cm-1. These bands are slightly shiftedto lower wavenumbers in comparison to neutral azulene films(Figure 1a) which might be due to the extensiveπ-electronconjugation between the azulene and vinylene and/or thiophenegroups.
Even though no remarkable structural differences can befound between films synthesized at different potentials, it isworth pointing out that the internal relative intensity of the bandfrom the vinylene group at 1600 cm-1 and the band from theazulene part at 1565 cm-1 is different in the two spectra. Theobserved change together with the fact that no bands related tothiophene moiety have shifted indicates that polymerizationaccording to procedure II results in a slightly different structure.A possible structure might be that the thiophene group in onemonomer unit couples with the azulene group of a second unit,since azulene is more reactive (lower oxidation potential) thanthiophene. In this way, the vinylene group would lose its doublebond character due to the formation of the new structure andlocalization of charges in it.
Figure 2b shows the Raman spectra of dry films frommonomerB after polymerization (A) according to procedure Iand (B) according to procedure II and measured with theexcitation lineλexc ) 780 nm. The insert shows the UV-visabsorption spectra of the corresponding films in their dopedstate. The arrow indicates the excitation line of the laser usedfor Raman measurements. From UV-vis measurements,8 theuse of λexc ) 780 nm will result in an enhancement of thevibrations associated with the quinoid units in the film. As canbe seen the spectra in Figure 2b obtained withλexc ) 780 nmshow broadening, vanishing, and shift of some bands comparedwith the spectra of the neutral film obtained withλexc ) 514nm (Figure 2a). The Raman bands at 1600 and 1220 cm-1
characteristic for the vinyl group and found in the spectra ofthe pristine form of the film (Figure 2a) totally disappear in thespectra of the films measured atλexc ) 780 nm (Figure 2b).This might be due to the structural reorganization of the vinylenegroup.14,15The Raman band at 1518 cm-1 can be correlated tothe band from the azulene moiety that can be seen in the spectraof the pristine film at 1538 cm-1 (Figure 2a). The downshift inthe frequency of this band in the spectra of the excited film canbe due to the formation of a quinoid structure. Furthermore,this band at 1518 cm-1 can also contain a contribution of theasymmetric stretching of CdC in the thiophene rings. The bandat 1442 cm-1 is attributed to the symmetrical stretching modeof CdC in thiophene rings of the quinoid units in the polymerchain.18 The latter band is correlated with the Raman band at1460 cm-1 in the spectra of the pristine film (Figure 2a) andcan be assigned to the so-called band B. The band at 1272 cm-1
from the intra-ring C-C stretching in the thiophene ring shiftstoward lower wavenumbers in comparison to the spectra of thepristine form of the film due to the increasing of the quinoidcharacter in the excited film. The Raman band seen at 1188cm-1 can also be found in the spectra of azulene films measuredwith λexc) 780 nm (Figure 1b). Finally, the band correlated tothe D line due to symmetric bending of the C-H groups inâ-position in the thiophene ring and found in the spectra of thepristine film (Figure 2a) cannot be clearly distinguished in thespectra of the excited film. Downshift of the different bands in
the spectra of the excited film may be explained in terms of thedistortion of the polymer chain, leading partially to a quinoidstructure.
3.2. In Situ FTIR-ATR Characterization. The infraredspectra are obtained in situ during p-doping in monomer-freesolution of the electrosynthesized films. In this way, structuraland electronic changes in the films can be studied as theelectrochemical reaction proceeds.
The redox response of azulene films electrosynthesized attwo different polymerization potentials (I and II) has beenstudied by in situ FTIR-ATR spectroscopy in 0.1 M TBAPF6-ACN. The redox behavior of those films has been describedearlier.8 The azulene films have been electropolymerizedaccording to procedure I. Figure 3a shows the FTIR-ATRspectra of a poly(azulene) measuring during p-doping. The cyclicvoltammogram is shown in the inset in Figure 3a. The arrowsindicate where the different spectra have been measured.Enlargement of the spectra in the wavenumber region 1600-700 cm-1 is shown in Figure 3b. The spectra are scaledindividually for a better comparison. In Figure 3a the spectraare dominated by a broad absorption at high energy thatcontinuously increases and shifts to lower wavenumbers ()lower energies) upon doping of the material. Two absorbancemaxima can be seen at approximately 3700 cm-1 and at 6200cm-1. Figure 3b shows new infrared induced bands, so-calledIRAV bands, that grow during p-doping of the film at 1514,
Figure 3. In situ FTIR-ATR difference spectra recorded duringoxidation (p-doping) of electrosynthesized azulene film (a) in thewavenumber region 7500-900 cm-1 (the numbers indicate the potentialvalues where each spectrum was recorded and refer to the cyclicvoltammogram in the inset) and (b) in the wavenumber region 1600-700 cm-1. The film was made according to procedure I.
23346 J. Phys. Chem. B, Vol. 110, No. 46, 2006 Meana-Esteban et al.
1493, 1423, 1367, 1308, 1269, 1228, 1196, 1065, 951, 916, 879,856, 754, and 704 cm-1. The pattern of the IRAV bands growingupon doping of the film is very complex. Intensity of theelectronic absorption and the IRAV bands increases drasticallywith increasing doping level of the film. The changes are morepronounced until potentials of approximately 800 mV where amaximum doping level is achieved. Upon further doping thechanges are less pronounced but a continuous increase in theintensity is still observed in the spectra. The changes in theabsorbance are highly reversible upon reduction of the film,and the absorbance decreases until it reaches its initial values.
The in situ FTIR-ATR difference spectra during oxidation(p-doping) of azulene film polymerized according to theprocedure II are shown in Figure 4a. The voltammogram isshown in the inset in Figure 4a. At higher wavenumbers (higherenergies) two electronic absorption maximums can be observedat approximately 3800 and 5200 cm-1, which increase upondoping of the material. In the lower wavenumber region (Figure4b) the IRAV bands are shown in detail. A complicated patternis observed with some IRAV bands that shift during oxidationof the azulene film. The main IRAV bands during p-dopingappear at 1510, 1460, 1419, 1342, 1296, 1261, 1230, 1176, 1063,and 1001 cm-1. The increase of both the electronic absorptionand the IRAV bands can be observed during the entire dopingreaction. Furthermore, these changes are fully reversible uponreduction.
Figure 5 compares the spectra at 1.1 V during p-doping(cyclic voltammograms are shown in the inset in Figure 5) ofelectrosynthesized azulene films polymerized according toprocedures I and II. Both spectra are related to a referencespectrum chosen at∼800 mV where in both cases the maximumdoping level is achieved. In this way, the difference spectra willshow only the changes occurring during the last part of thedoping reaction. In the case of the azulene film polymerizedaccording to procedure I the pattern of IRAV bands show fourmain IRAV bands located at 1430 as a doublet, 1352, 1172 asa doublet, and 1076 cm-1. As we can see in Figure 5, apartfrom the shift in the IRAV bands to lower energy of the azulenefilm polymerized according to procedure II, all the spectralfeatures are present in both spectra. The differences mightoriginate from a more cross-linked structure in the filmsynthesized at higher potentials that would result in a moreplanar structure where the electrons can move more freely. Thiscould be the reason both the electronic and the IRAV bands ofthe azulene film synthesized according to procedure II are shiftedto lower energies. On the other hand, when the azulene film iselectrosynthesized according to procedure I it might result in amore linear polymer. In the latter case according to the literaturethe polymerization of azulene would take place at 1 and 3positions.19-21
The charge consumed during electropolymerization is almostthe same (films of same thickness) for the films made in thetwo potential ranges. However, the charge injected duringdoping is higher (39 mC) in the film synthesized according toprocedure II than in the case of azulene films synthesizedaccording to procedure I where the charge consumed duringp-doping is 30 mC. This is in agreement with the shift in thebands to lower energies observed in the FTIR-ATR spectra ofthe azulene films made according to procedure II in comparisonto films made in I.
A direct correlation of the main IRAV vibrations with theRaman frequencies of the film measured by theλexc ) 514 nmcannot according to the obtained experimental results be madein polymers of more complex structure.
The redox response of electrosynthesized 2-[(E)-2-azulen-1-ylvinyl]thiophene (B) films has been described earlier.8
Figure 6a shows the in situ FTIR-ATR spectra duringp-doping of films electrosynthesized from monomerB accordingto procedure I. The cyclic voltammogram is shown in the insetin Figure 6a. The infrared spectra measured during p-dopingare characterized by the intense IRAV bands in the range from
Figure 4. In situ FTIR-ATR difference spectra recorded duringoxidation (p-doping) of electrosynthesized azulene film (a) in thewavenumber region 7500-900 cm-1 (the numbers indicate the potentialvalues where each spectrum was recorded and refer to the cyclicvoltammogram in the inset) and (b) in the wavenumber region 1600-700 cm-1. The film was made according to procedure II.
Figure 5. (s) IRAV spectrum of p-doped azulene film electrosyn-thesized in I and (---) IRAV spectrum of p-doped azulene filmelectrosynthesized according to procedure II.
FTIR-ATR Characterization of Polyazulene Films J. Phys. Chem. B, Vol. 110, No. 46, 200623347
700 to 1600 cm-1 and by an electronic absorption band above3000 cm-1 due to the formation of free charge carriers in thefilm. The high energy part of the spectra (high wavenumbers)shows a broad absorption with a maximum at approximately4000 cm-1 that coincides well with the maximum at 800 mVof the p-doping current seen in the cyclic voltammogram. Uponfurther doping, this maximum is further shifted toward higherwavenumbers () higher energies) and the absorbance increasescontinuously. An isosbestic point can be seen at 4500 cm-1.This behavior may be explained in terms of how the chargecarriers formed during p-doping of the film are differentlydelocalized along the different parts of the formed film. As hasbeen discussed earlier, the coupling position of azulene in theelectrochemical polymerization according to procedure I is at1- and 3-carbons. However, since the 1-position in monomerBis blocked by the substituent, in this particular case, theelectrochemical synthesis according to procedure I would resultin a dimeric species where the azulene units are coupled via3-,3- positions.22,23 Thus, at higher doping levels, the chargecarriers can be delocalized along the whole molecule and notonly in the azulene moiety. This might cause the shift of theelectronic absorption to higher energies since the thiophene parthas a higher oxidation potential.
Figure 6b shows the IRAV range in detail. The growth ofthe IRAV bands starts at a potential that coincides well with
the increase in the faradaic current in the cyclic voltammogramshown in the inset in Figure 6a and increases smoothly withincreasing doping levels. The main IRAV bands during p-dopingappear at 1579, 1547, 1504, 1433, 1379, 1302, 1219, 1153, 1105,1043, 951, 918, and 862 cm-1. Furthermore, some of the IRAVbands shift to lower energy at high doping level. The changesobserved in the spectra upon doping are fully reversible.
Comparison of the spectra at 800 mV (highest currentresponse in the cyclic voltammograms) of the azulene film andthe 2-[(E)-2-azulen-1-ylvinyl]thiophene films both electrosyn-thesized according to procedure I exhibit many similarities (notshown here). First of all, the electronic absorption band at higherenergies appears almost at the same wavenumbers. Furthermore,the IRAV bands in both cases have almost the same pattern. Inthe region between 1600 and 1100 cm-1 the IRAV bands ofazulene film are shifted to lower wavenumbers () lower energy)(Figure 3b) in comparison to those found in the spectra of 2-[(E)-2-azulen-1-ylvinyl]thiophene film (Figure 6b). This can beregarded as an indication of a higher delocalization (highereffective conjugation length) in the azulene films. However, thefact that some bands show frequency dispersion has beenexplained by Zerbi et al. in the framework of the ECC theory4
in terms of how the vibrational force constant fЯ, a parameterrelated to theЯ coordinate, used for the calculation of thevibrational frequency decreases due to the increase in delocal-ization of electrons in the films upon doping.
Figure 6. In situ FTIR-ATR difference spectra recorded duringoxidation (p-doping) of electrosynthesized 2-[(E)-2-azulen-1-ylvinyl]-thiophene film (a) in the wavenumber region 7500-900 cm-1 (thenumbers indicate the potential values where each spectrum was recordedand refer to the cyclic voltammogram in the inset) and (b) in thewavenumber region 1600-800 cm-1. The film was made according toprocedure I.
Figure 7. In situ FTIR-ATR difference spectra recorded duringoxidation (p-doping) of electrosynthesized 2-[(E)-2-azulen-1-ylvinyl]-thiophene film (a) in the wavenumber region 7000-900 cm-1 (thenumbers indicate the potential values where each spectrum was recordedand refer to the cyclic voltammogram in the inset) and (b) in thewavenumber region 1700-800 cm-1. The film was made according toprocedure II.
23348 J. Phys. Chem. B, Vol. 110, No. 46, 2006 Meana-Esteban et al.
Figure 7a shows the different spectra during electrochemicaloxidation of films electrosynthesized fromB according toprocedure II. The cyclic voltammogram during doping is shownin the inset figure. The maximum of the p-doping current is at650 mV. The spectra are dominated by two well-developedbroad absorption bands at high energies, correlated to theformation of free charge carriers in the film upon doping, withthe maxima around 4000 and 5700 cm-1. Both absorption bandsshift slightly toward lower wavenumbers () lower energies)upon p-doping of the film. The electronic absorption bands aremore defined in this case than in the film made according toprocedure I (Figure 6a). In addition to that, the redox responseis slightly different. This might be the first indication that twodifferent structures are formed in these two different electropo-lymerization procedures at different switching potentials. InFigure 7b the IRAV band range is shown in detail. The growthof the IRAV bands starts at potentials that coincide well withthe onset of the oxidation current in the cyclic voltammogramshown in the inset in Figure 7a, and the intensity of the bandsincreases continuously with increasing applied potential. Themain IRAV bands during p-doping appear at 1588, 1552, 1522,1450, 1381, 1283, 1221, 1043, 917, and 875 cm-1. Also in thiscase the infrared active vibration bands are fully reversible. Dueto the nature of the IRAV bands an exact interpretation of everyband is very difficult. Nevertheless, due to the nature of the
monomerB it could be possible that at high polymerizationpotentials a coupling between the thiophene rings in the 2- and5-positions might take place, resulting in a bithiophene derivatefilm with the vinylene and azulene moieties as the endinggroups. However, IRAV bands do not show any similaritieswith those found in bithiophene derivates24 which have threestrong IRAV bands at 1323, 1130, 1055 cm-1. Anotherpossibility might be that at high potentials the coupling takesplace between the thiophene ring in its freeR-position and theazulene moiety as has been discussed earlier in this paper. Inconclusion, the structures resulting at the two different polym-erization potentials are different but not remarkably. Thus, theIRAV band patterns do not differ very much. In both cases,bands between 1600 and 1500 cm-1 are due to end-ringvibration which is an indication of short chain length. Further-more, bands around 900 and 860 cm-1 are due to vibration inthe C-S ring.
As summary, in Table 1 the IRAV bands at high doping levels(Figure 5) are listed. In Table 2 the Raman bands ofA andBelectrosynthesized according to procedures I and II are listed.
4. Conclusions
Raman spectroscopy with two different excitation lines,λex
) 514 nm andλex ) 780 nm, has been used to study thestructure of different films electrosynthesized from azulene (A)and 2-[(E)-2-azulen-1-ylvinyl]thiophene (B) by potential scan-ning in different potential ranges. In both cases the use ofλex
) 514 nm will result in an enhancement of the Raman modesassociated with the neutral form of the film whereas withλex
) 780 nm the doped form of the film will be enhanced. A cleardifference in the Raman response of the polymer fromA uponchange in excitation wavelength was observed. This wasexplained by influence of the intralayer interactions, known tobe strong for polyazulene, on the vibrational modes from theneutral form. However, Raman spectra of dry azulene filmselectrosynthesized at the two different potential ranges do notshow any remarkable structural differences. On the other hand,in the case of 2-[(E)-2-azulen-1-ylvinyl]thiophene films, twodifferent structures have been synthesized in the two electro-chemical potentials as can be deduced from the change in therelative intensity of the band associated to the vinyl group and
TABLE 1: IRAV Modes of Electrosynthesized Films from Aand B in I and II
FTIR-ATR Characterization of Polyazulene Films J. Phys. Chem. B, Vol. 110, No. 46, 200623349
the one to the azulene molecule in the spectra of the pristinefilm synthesized at high polymerization potentials.
In situ FTIR-ATR has been used to study the redox reactionsof the different films. Upon charging, the different films exhibitan increase in the electronic absorption as well as growth ofnew infrared active vibration bands. From the experimentalresults obtained in the in situ FTIR-ATR measurements in termsof electronic absorption and IRAV bands, azulene films elec-trosynthesized at different polymerization potentials may havethe same structure, but in films electrosynthesized at highpolymerization potential the cross-linking is higher, resultingin films with higher effective conjugation length due to theplanarization of the structure. On the other hand, films elec-trosynthesized from 2-[(E)-2-azulen-1-ylvinyl]thiophene at dif-ferent polymerization potentials show IRAV bands that indicatedifferent coupling between the monomers at the low and highpolymerization potential.
Acknowledgment. Financial support from the GraduateSchool of Materials Research is gratefully acknowledged(B.M.E.). The authors thank C. Nitu and A. C. Alexandru forthe synthesis of monomerB. This work forms part of theactivities being pursued at the Åbo Akademi Process ChemistryCenter, within the Finnish Centre of Excellent Program (2000-2011) of the Academy of Finland.
References and Notes
(1) Kvarnstrom, C.; Ivaska, A.; Neugebauer H. InAdVanced FunctionalMolecules and Polymers; Nalwa, H. S., Ed.; Taylor and Francis: New York,2001; Chapter 6.
(2) Horovitz, B.Solid State Commun.1982, 41, 729.(3) Ehrenfreund, E.; Vardeny, Z. V.; Brafman, O.; Horovitz B.Phys.
ReV. B 1987, 36, 1535.(4) Del Zoppo, M.; Castiglioni, C.; Zuliani P.; Zerbi, G. InHandbook
of Conducting Polymers,2nd ed.; Skotheim, T. A., Elsenbaumer; R. L.,
Reynolds, J. R., Eds.; Marcel Dekker: New York, 1998; Chapter 28.(5) Zerbi, G.; Castiglioni, C.; Del Zoppo, M. InElectronic Materials:
The Oligomeric Approach; Mullen, K., Wegner, G., Eds.; Wiley-VCH: NewYork, 1998; pp 345-402.
