Correlations between Structure and Far-Infrared Active Modes in

12662

J. Phys. Chem. B 2008, 112, 12662–12665

Correlations between Structure and Far-Infrared Active Modes in Polythiophenes P. Hermet,*,† J.-L. Bantignies,‡ R. Almairac,‡ J.-L. Sauvajol,‡ F. Serein,§ and J.-P. Le`re-Porte§ Laboratoire de Physique du Solide, Faculte´s UniVersitaires Notre-Dame de la Paix, 5000 Namur, Belgium, Laboratoire des Colloı¨des, Verres et Nanomate´riaux (UMR CNRS 5587), UniVersite´ Montpellier II, 34095 Montpellier Ce´dex 5, France, and Institut Charles Gerhardt (UMR CNRS 5253), Architectures Mole´culaires et Mate´riaux Nanostructure´s (AM2N), Ecole Nationale Supe´rieure de Chimie de Montpellier, 34096 Montpellier Ce´dex 5, France ReceiVed: May 31, 2008; ReVised Manuscript ReceiVed: July 23, 2008

We have investigated the experimental X-ray and far-infrared responses of three polythiophenes synthesized from a thiophene, R-bithiophene, and R-quaterthiophene monomer. The X-ray data show that the crystallinity of the different polythiophene samples depends on the synthesis conditions. An excellent correlation between the crystallinity of polythiophenes and their far-infrared signatures is demonstrated. In addition, the assignment of the far-infrared phonon modes in polythiophenes is given by using both an experimental filiation procedure and first-principles calculations. In particular, the ring libration inside the polymeric chain, directly involved in the electron-phonon coupling, is assigned. I. Introduction Polythiophenes are π-conjugated organic semiconductors particularly attractive for several technological applications.1,2 Their electrical and optical properties are directly connected to their molecular structure, the orientation of the macromolecular chains, and the packing of the macromolecular chains within the crystal structure.3,4 However, the correlation between such properties and the structural characteristics of these materials, which is required to improve their electronic performance, has rarely been performed.5 The investigation of the low-frequency vibration domain is particularly interesting due to its sensitivity to phonon modes and electron-phonon coupling. An accurate understanding and assignment of the experimental low-frequency signatures in polythiophenes is difficult due to defects both of conformation (conjugation length) and configuration (supramolecular order). Thus, the study of oligothiophenes, such as R-bithiophene (R-2T), R-quaterthiophene (R-4T), R-sexithiophene (R-6T), and R-octithiophene (R-8T), characterized by a well-defined structure, is especially useful to understand the properties of the polymer. In the literature, an impressive number of papers report the investigation of the lattice dynamics of polythiophenes or oligothiophenes. Among these papers, a large number of them investigate the phonon modes at high-frequency (i.e., above 300 cm-1) in polythiophenes or oligothiophenes by using several spectroscopic techniques like mid-infrared,6-16 Raman scattering,6,7,9,10,12,13,17-20 and inelastic neutron scattering (INS)10,11,21-23 and compare these experimental spectra to theoretical models obtained on isolated molecule using either empiric force field,9,20 semiempirical methods,12,24,25-27 or first-principles methods.10,11,28,29 Within these models, accurate information are limited to intramolecular phonon modes and might be obtained for low-frequency phonon modes (i.e., below 300 cm-1), where molecular (chain torsion and ring libration modes) and lattice * Corresponding author. † Faculte ´ s Universitaires Notre-Dame de la Paix. ‡ Universite ´ Montpellier II. § Ecole Nationale Supe ´ rieure de Chimie de Montpellier.

modes (interchain modes) mix. Indeed, due to the lack of the intermolecular interactions in these models, a poor agreement is generally observed for both the intensities and frequency positions between the calculated vibrational bands and the experimental ones in the low-frequency range,14,30 leading the assignments of these modes extremely difficult. Only a few number of papers report the experimental low-frequency phonon signatures of oligothiophenes and compare them to theoretical spectra obtained on crystalline phase. These calculations and experiments, especially usefull to understand and assign the lowfrequency dynamics of polythiophenes, are reported using INS,21,22,30 infrared,14,15,31 and Raman spectroscopies18,19 in R-2T, and the R-4T and R-6T polymorph phases. Nevertheless, in despite of this impressive number of papers in the literature, the low-frequency phonon modes in polythiophenes has only been experimentally reported using INS spectroscopy,23 but their assignments currently still remain unknown. By contrast to Raman and INS spectroscopies, far-infrared spectroscopy is a technique very sensitive to obtain accurate information on low-frequency modes, and especially about the electron-libron coupling, in polythiophenes and oligothiophenes. This is mainly due to the fact that Raman phonon modes are not very active (nonresonant process) and the inelastic neutron scattering responses are not well resolved in this low-frequency domain.31 In this paper, we experimentally investigate the low-frequency phonon dynamics of three polythiophene samples, which only differ by the size of the monomer used before the polymerization, by using X-ray diffraction and far-infrared spectroscopies. We state a clear correlation between the crystallinity of the different polythiophene samples and their far-infrared phonon signatures. In addition, this study also demonstrates that the profile of the experimental far-infrared spectra of the different polythiophenes is very similar to that of its oligomers. Thus, a complete assignment of the far-infrared phonon modes in polythiophenes can be obtained from density functional theory calculation on the R-4T oligomer in its crystalline phase. This oligomer is long enough to reproduce in a good agreement the physical properties of polythiophenes and has a reasonable

10.1021/jp804813p CCC: $40.75  2008 American Chemical Society Published on Web 09/06/2008

Structure and Active Modes in Polythiophenes

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12663

Figure 2. X-ray diffraction (room temperature) and far-infrared (10 K) spectra of the PT-1, PT-2, and PT-4 polythiophenes.

TABLE 2: Degree of Crystallinity and Coherence Length of the Different Polythiophenes Studied Figure 1. Molecular structures of the three polythiophene samples studied (n ∈ N*).

TABLE 1: Degree of Polymerization (Number of Monomers) and Conjugation Length (Number of Coplanar Thiophene Rings) of the Different Polythiophenes Studied polythiophene samples

degree of polymerization

conjugation length

PT-1 PT-2 PT-4

50 38 17

3 5 7

number of atoms to be treated within the density functional theory framework. II. Experimental Conditions Three samples of polythiophenes have been synthesized from chemical oxidation. The experimental protocol of this synthesis can be found in ref 32. The only difference between these three samples of polythiophenes is the size of the monomer of departure used before the polymerization. In the following, we called PT-1, PT-2, and PT-4, the polythiophenes respectively synthesized from a thiophene (1T), R-2T and R-4T monomer (see Figure 1). The analysis of the Raman and the mid-infrared spectra of the different polythiophenes, respectively, give an estimate of the conjugation length and the degree of polymerization.6 These two quantities characterize the different polythiophene samples and are reported in Table 1. X-ray diffraction data on the different polythiophene samples were measured with a powder diffractometer equipped with a curved position sensitive detector INEL CPS 120 allowing to collect simultaneously the signal in a 120° 2θ range, with a Cu KR (λ ) 1.542 Å) source and a germanium monochromator. A grazing incidence was used to increase the amount of matter in the incident beam. This configuration results in some little errors on the peak positions in the diffracted data, but nevertheless, they remain insignificant regarding the broad diffraction lines in polythiophenes. Far-infrared measurements on the different polythiophenes and oligothiophenes were carried out in transmission on a Bruker IFS 113V Fourier transform spectrometer equipped with a Si-

polythiophene samples

degree of crystallinity (%)

coherence length (Å)

PT-1 PT-2 PT-4

0 14 38

0 53 90

bolometer detector cooled at 4 K, a Mylar composite beamsplitter, and an arc discharge mercury source. The samples are compressed under high pression (5 tons) to form pure isotropic pellets of a 13 mm diameter and a 45 mg weight. The spectral resolution was 4 cm-1, and 64 scans were accumulated for each spectrum in the 40-160 cm-1 range. The measurements were performed in the 10-300 K range with a temperature precision of 1 K using a liquid helium cryostat. III. Results and Discussion A. Correlation between Structure and Low-Frequency Signatures. Figure 2 reports the diffracted intensity by the different polythiophene samples (PT-1, PT-2, and PT-4) at room temperature as a function of the module of the wave vector Q. We observe that the PT-1 has no diffraction line whereas the PT-2 and PT-4 are respectively dominated by one and three diffraction lines which are the 110, 200, and 210 reflections of the crystalline fraction of the sample according to ref 33. These reflections characterize the packing of the polythiophene chains in a plane perpendicular to the chains axis. In order to quantify the degree of crystallinity of the different polythiophene samples, we have determined a parameter defined as the ratio between the intensity of the diffraction lines over the total diffracted intensity in the 0.5-3 Å-1 range. Note that this parameter differs from that of Mo et al.33 by its definition, but this choice has the merit of the simplicity. Table 2 gives the degree of crystallinity calculated according to this method for the different polythiophenes. We observe that this degree of crystallinity, initially null for the PT-1, increases significantly from PT-2 (14%) to PT-4 (38%). This latter value is in good agreement with that of 36.5% obtained by Mo et al.33 In addition, we can correlate this evolution of the structure of polythiophenes to the evolution of the coherence length, ξ, of the samples. Under the assumption that all the line broadening results from finite crystallite size, the coherence length, also reported in Table 2, was calculated from the Sherrer equation applied on the 110 line34

12664 J. Phys. Chem. B, Vol. 112, No. 40, 2008

ξ)

57.3 λ (B - b02)1⁄2 cos θ 2

Hermet et al.

(1)

where B is the half-width in degrees of a diffraction line (in 2θ), b0 is the instrumental resolution in degrees (in our case 0.15° in 2θ), and λ is the wavelength of the X-photons. These results show that the size of the monomer used before the polymerization is therefore a pertinent parameter which significantly influences the crystallinity of the polymer: the degree of crystallinity and the coherence length of the polymer increase when the size of the monomer increases. A possible interpretation is that a large monomer contributes to decrease the number of conjugation defects in polythiophenes, leading to a better interchain organization and larger crystalline domains. Figure 2 also reports the far-infrared absorbances of PT-1, PT-2, and PT-4 samples in the 40-160 cm-1 range at 10 K. By contrast to the PT-2 and PT-4 which have spectroscopic signatures, we observe that the PT-1 has no phonon mode in this domain. The observed oscillations in PT-1 are Fabry-Pe´rot interferences due to internal multireflection between the flat faces of the pellet. The PT-2 have wide bands between 40 and 160 cm-1 whereas PT-4 has, in the same frequency range, four well resolved bands centered at 48, 70, 103, and 134 cm-1. Since the degree of crystallinity of PT-2 is lower than that of PT-4, we can assign the broad infrared signatures of PT-2 to structural disorder, leading to a spreading of the frequency of the infrared active modes. The broadening of the bands is due to the loss of the translational symmetry leading to a breaking of the infrared selection rules. The absence of infrared signature in PT-1 is therefore associated with the amorphous character of this sample. Thus, the far-infrared spectroscopy is very sensitive to the crystallinity of the different polythiophene samples which is conditioned by the size of the monomer of departure. B. Phonon Mode Assignments in Polythiophenes. Figure 3 compares the 10 K experimental far-infrared spectra of the PT-4 polythiophene and three of its oligomers (R-4T, R-6T, and R-8T). The calculated far-infrared spectrum of the R-4T oligomer using density functional theory is also displayed in this figure. The computational details of this calculation can be found in ref 15. Figure 4 displays the temperature dependence of the far-infrared spectra of PT-4 in the 10-300 K range. As expected, the spectral signatures of the PT-1 and PT-2 show an identical behavior as a function of the temperature due to the poor crystallinity of these samples, and these spectra are therefore not shown. The experimental far-infrared spectra of the different oligothiophenes are dominated by two vibrational bands, centered around 62 and 92 cm-1 for the R-4T. These vibrational bands have the same relative intensities from R-4T to R-8T oligomers, and as expected, their frequency position upshifts when the size of the oligomer increases. The analysis of the R-4T eigendisplacements calculated by density functional theory shows that these two phonon bands are respectively assigned to an in-plane and an out-of-plane CR-S-CR bending modes.15 The profile of the experimental far-infrared spectra of PT-4 is very similar to that of its oligomers. Indeed, it is dominated by two intense and broad vibrational bands centered at 70 and 103 cm-1 which have the same relative intensities as in oligothiophenes. As expected, the vibrational bands of PT-4 are larger than those of oligothiophenes due to the lower crystallinity of polythiophenes with respect to oligothiophenes. The bandwidth could be related to a dispersion of the conjugation length. In addition, the frequency position of the phonon bands of PT-4 shows an upshift with respect to R-8T oligomer which can be

Figure 3. Comparison between the experimental far-infrared spectra of PT-4 polythiophene and its oligothiophenes (R-4T, R-6T, and R-8T). The density functional theory calculation of the far-infrared spectrum of R-4T oligomer is also displayed in the bottom of the figure.15 A linear baseline correction has been performed on the experimental spectra for the presentation.

Figure 4. Temperature dependence of the far-infrared spectra of PT-4 polythiophene in the 10-300 K range.

interpreted as a consequency of an average size of the PT-4 chain greater than eight thiophenes. An accurate assignment of the far-infrared phonon modes in the PT-4 requires the calculation of the dynamical matrix and the Born effective charges in crystalline phase.15 Unfortunately, a such calculation is not possible because (i) the important number of atoms in the polymer and (ii) the crystallographic data of this polymer are incomplete due to the difficulty to have a polythiophene monocrystal without defects. However, since the profile of the far-infrared spectra between the PT-4 polythiophene and its oligothiophenes are very similar, we propose

Structure and Active Modes in Polythiophenes an assignment of the phonon modes of PT-4 by a direct comparison of its far-infrared spectrum with that of R-4T oligomer. The phonon bands of PT-4 centered at 70 and 103 cm-1 are therefore identified to the corresponding phonon bands centered at 62 and 92 cm-1 in R-4T and are respectively assigned to an in-plane and an out-of-plane CR-S-CR bending modes. These two experimental phonon bands in PT-4 have a quasi-harmonic dependence related to strengthening of the effective force constants when the temperature decreases (see Figure 4). This behavior is also consistent with intramolecular vibration assignments for these two modes. Note that the outof-plane bending mode at 103 cm-1 is, as expected, the most affected by the decreasing of the crystallinity of the sample. The phonon band centered at 48 cm-1 in PT-4 is small and very broad and only visible to low-temperature. A similar band centered at 46 cm-1 in the calculated spectrum of R-4T, and assigned to the ring libration,31 suggests the same assignment for the PT-4. We emphasize that the ring libration in the polythiophenes is assigned for the first time. Finally, the phonon band centered at 134 cm-1 in PT-4 has a quite harmonic behavior within the resolution of our experiment. In addition, there is no phonon band around this frequency range in the calculated spectrum of R-4T. Thus, the band centered at 134 cm-1 in the PT-4 could be assigned to a conjugation defect mode. IV. Conclusions In conclusion, three polythiophene samples, which only differ by the size of the monomer used before the polymerization, have been experimentally studied by X-ray diffraction and farinfrared spectroscopies. The X-ray results show that the crystallinity of the different polythiophene samples depends on the synthesis conditions. In particular, the amorphous character of the polythiophene synthesized from the shortest monomer (PT1) is demonstrated. By contrast, the polythiophenes synthesized from R-2T and R-4T monomers show a crystalline and an amorphous fraction. A concomitant increase of the crystallinity of the polymers and the size of the monomer is observed. The far-infrared spectroscopy is very sensitive to lowfrequency phonon modes in polythiophenes. Four far-infrared phonon modes, centered at 48, 70, 103, and 134 cm-1, are identified in PT-4. The comparison between the X-ray and farinfrared data show an excellent correlation between the crystallinity of polythiophenes and their far-infrared signatures. In particular, there is no far-infrared active mode when the sample is amorphous (PT-1) whereas the intensity of the far-infrared bands increases and are better resolved when the crystallinity of the samples increases (PT-2 and PT-4). The comparison between the experimental far-infrared spectrum of the PT-4 and the R-4T calculated one allows the assignment of the four phonon bands observed experimentally in PT-4. The band centered at 48 cm-1 is assigned to the ring libration whereas the bands centered at 70 and 103 cm-1, especially sensitive to the crystallinity of the polythiophenes, are respectively assigned to an in-plane and an out-of-plane intramolecular CR-S-CR bending modes. The band centered at 134 cm-1 could be a defect mode.

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12665 Acknowledgment. We are grateful to Denis Fichou for the synthesis of the R-8T oligomer. Computations were performed at the CINES (Montpellier, France). References and Notes (1) Gigli, G.; Barbarella, G.; Favaretto, L.; Cacialli, F.; Cingolani, R. Appl. Phys. Lett. 1999, 75, 439. (2) Dodabalapur, A.; Torsi, L.; Katz, H. E. Science 1995, 268, 270. (3) Kim, J.; Swager, T. M. Nature 2001, 411, 1030. (4) Loi, M. A.; Mura, A.; Bongiovanni, G.; Cai, Q.; Martin, C.; Chandrasekhar, H. R.; Chandrasekhar, M.; Graupner, W.; Garnier, F. Phys. ReV. Lett. 2001, 86, 732. (5) Yu, L.; Bao, Z.; Cai, R. Angew. Chem., Int. Ed. Engl. 1993, 32, 1345. (6) Furukawa, Y.; Akimoto, M.; Harada, I. Synth. Met. 1987, 18, 151. (7) Akimoto, M.; Furukawa, Y.; Takeuchi, H.; Harada, I.; Soma, Y.; Soma, M. Synth. Met. 1986, 15, 353. (8) Ehrendorfer, Ch.; Neugebauer, H.; Neckel, A.; Ba¨uerle, P Synth. Met. 1993, 55-57, 493. (9) Louarn, G.; Buisson, J. P.; Lefrant, S.; Fichou, D. J. Phys. Chem. 1995, 99, 11399. (10) Degli Esposti, A.; Moze, O.; Taliani, C.; Tomkinson, J. T.; Zamboni, R.; Zerbetto, F. J. Chem. Phys. 1996, 104, 9704. (11) Degli Esposti, A.; Zerbetto, F. J. Phys. Chem. A 1997, 101, 7283. (12) Zerbi, G.; Chierichetti, B.; Inga¨nas, O. J. Chem. Phys. 1991, 94, 4637. (13) Degli Esposti, A.; Fanti, M.; Muccini, M.; Taliani, C.; Ruani, G. J. Chem. Phys. 2000, 112, 5957. (14) Hermet, P.; Bantignies, J. L.; Rahmani, A.; Sauvajol, J. L.; Johnson, M. R.; Serein, F. J. Phys. Chem. A 2005, 109, 1684. (15) Hermet, P.; Bantignies, J. L.; Sauvajol, J. L.; Johnson, M. R. Synth. Met. 2006, 156, 519. (16) Gök, A.; Omastova´, M.; Gu¨l Yavuz, A. Synth. Met. 2007, 157, 23. (17) Brillante, A.; Bilotti, I.; Albonetti, C.; Moulin, J. F.; Stoliar, P.; Biscarini, F.; De Leeuw, D. M. AdV. Funct. Mater. 2007, 17, 3119. (18) Brillante, A.; Bilotti, I.; Biscarini, F.; Guido Della Valle, R.; Venuti, E. Chem. Phys. 2006, 328, 125. (19) Hermet, P.; Izard, N.; Rahmani, A.; Ghosez, Ph. J. Phys. Chem. B 2006, 110, 24869. (20) Louarn, G.; Mevellec, J. Y.; Buisson, J. P.; Lefrant, S. Synth. Met. 1993, 55-57, 587. (21) Hermet, P.; Bantignies, J. L.; Rahmani, A.; Sauvajol, J. L.; Johnson, M. R. J. Phys.: Condens. Matter 2004, 16, 7385. (22) Hermet, P.; Bantignies, J. L.; Rahmani, A.; Sauvajol, J. L.; Johnson, M. R. J. Phys. Chem. A 2005, 109, 4202. (23) Sauvajol, J. L.; Bormann, D.; Palpacuer, M.; Lere-Porte, J. P.; Moreau, J. J. E.; Dianoux, A. J. Synth. Met. 1997, 84, 569. (24) Lo´pez Navarrete, J. T.; Zerbi, G. J. Chem. Phys. 1991, 94, 957. (25) Ramirez, F. J.; Herna´ndez, V.; Lo´pez Navarrete, J. T. J. Comput. Chem. 1994, 15, 405. (26) Herna´ndez, V.; Ramirez, F. J.; Lo´pez Navarrete, J. T. J. Mol. Struct. 1993, 294, 37. (27) Bre´das, J. L.; Silbey, R.; Boudreaux, D. S.; Chance, R. R. J. Am. Chem. Soc. 1983, 105, 6555. (28) Cui, C. X.; Kertesz, M.; Eckhardt, H. Synth. Met. 1991, 41-43, 3491. (29) Kofranek, M.; Kovar, T.; Lischka, H.; Karpfen, A. J. Mol. Struct. (Theochem) 1992, 259, 181. (30) Van Eijck, L.; Johnson, M. R.; Kearley, G. J. Phys. Chem. A 2003, 107, 8980. (31) Hermet, P.; Bantignies, J. L.; Maurin, D.; Sauvajol, J. L. Chem. Phys. Lett. 2007, 445, 47. (32) Lere-Porte, J. P.; Moreau, J. J. E.; Torreilles, C. Eur. J. Org. Chem. 2001, 7, 1249. (33) Mo, Z.; Lee, K. B.; Moon, Y. B.; Kobayashi, M.; Heeger, A. J.; Wudl, F. Macromolecules 1985, 18, 1972. (34) Alexander, L. E. X-Ray Diffraction Methods in Polymer Science; Wiley-Interscience: New-York, 1976.

JP804813P