• 沒有找到結果。

Sub-5-fs spectroscopy of a thiophene derivative with a quinoid structure

N/A
N/A
Protected

Academic year: 2021

Share "Sub-5-fs spectroscopy of a thiophene derivative with a quinoid structure"

Copied!
6
0
0

加載中.... (立即查看全文)

全文

(1)

Sub-5-fs spectroscopy of a thiophene derivative with a quinoid structure

Takayoshi Kobayashi

a,b,c,*

, Haibo Wang

a

, Zhuan Wang

a

, Tetsuo Otsubo

d

aDepartment of Physics, University of Tokyo, Hongo 7-3-1, Bunkyo, Tokyo 113-0033, Japan bInstitute of Laser Engineering, Osaka University, Ibaraki 567-0047, Japan cDepartment of Electrophysics, National Chiao Tung University, Hsinchu 300, Taiwan dDepartment of Applied Chemistry, Hiroshima University, Higashi-Hiroshima 739-8527, Japan

Received 22 February 2006; in final form 17 May 2006 Available online 3 June 2006

Abstract

Sub-5-fs spectroscopy of a thiophene derivative with an electron donative and an acceptive moieties and quinoid structure provided the experimental evidence of dynamic mode coupling. It was shown that two out-of-plane bending modes mediate the dynamic mode coupling between 1469 and 1379 cm1and between 1603 and 1469 cm1. These couplings are considered to be associated with geomet-rical relaxation relevant to the neutral bipolaron formation in all-s-trans polyacetylene.

 2006 Elsevier B.V. All rights reserved.

1. Introduction

Thiophene-based oligomers have been extensively stud-ied theoretically and experimentally for a better under-standing of physical properties of polythiophene, which has been one of most popular conjugated polymers since 1990’s[1–11]. They are also promising materials for various devices[7,12–15], such as light-emitting diodes[16–18].

Vibrational spectroscopy has shown [8] that polythi-ophenes contain all-s-trans sequences of thiophene rings linked at a- and a0-positions with a distribution of

sequential lengths, so-called the conjugation length. The distribution of conjugation lengths has been ascribed to the existence of disordered structures in the polymer chain. Since polythiophene has a nondegenerate ground state, the optical properties of doped polythiophene have been discussed in terms of polarons and bipolarons

[10,11], which are self-localized excitation in conjugated polymers. A positive polaron (a radical cation in chemi-cal terminology) is created when an electron is removed

from a polymer chain by acceptor doping. The charge +e and spin 1/2 are localized over several thiophene rings with geometrical changes. If another electron is removed from the chain, a positive bipolaron corresponding to a dication, +2e, but without spin is formed. The electronic state of polarons and bipolarons have been discussed mainly in the framework of theories incorporating elec-tron-lattice interaction [10,11]. It is of interest to study the processes following the creation of electron and hole by photoexcitation. It may generate a bound pair of the opposite charge forming exciton and relax to a self-trapped exciton or the pair may be separated into oppo-sitely charged geometrically relaxed localized excitation, namely two oppositely charged polarons. Experimental studies have been made on the vibronic coupling in thio-phene oligomers with electron-donating and accepting groups attached to the oligomer to explore the effect of charge on the branching ratio of neutral bipolaron (which is equivalent to self-trapped exciton) and two charged polarons with opposite charges separated from each other [1,11,12].

In the present study, we have performed sub-5-fs spec-troscopy of a derivative of quinoidal thiophene (Fig. 1) and found that this molecule is a good model as a repeat unit of polythiophene with all-s-trans conformation.

0009-2614/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2006.05.102

*

Corresponding author. Address: Department of Physics, University of Tokyo, Hongo 7-3-1, Bunkyo, Tokyo 113-0033, Japan. Fax: +81 3 5841 4165.

E-mail addresses: [email protected] (T. Kobayashi),

(2)

2. Experimental

The sample used in the present study is a novel syn-thesized quinoidal thiophene with push-pull charge-trans-fer groups [19,20], abbreviated as QT1, dissolved in tetrahydrofuran (THF) solution. This thiophene substi-tute has an electron-donating 1,3-dithiol-2-ylidene unit and an electron-withdrawing dicyanomethylene unit at the terminals. Because of the quinoidal structure, this molecule has an all-s-trans cis polyacetylene configura-tion, and hence it is a good prototype of a repeat-unit in polyacetylene. This is quite different from that of thi-ophene oligomer with an aromatic structure containing s-trans and s-cis configurations.

The pump and probe pulses were both produced from a noncollinear optical parametric amplifier (NOPA) seeded by a white-light continuum [21–23]. The pulse of the NOPA output was compressed with a system composed of a pair of prisms and chirp mirrors [21–23]. The pump source of this NOPA system was a regenerative amplifier (Spectra Physics, Spitfire) with central wavelength: 790 nm, pulse duration: 50 fs, repetition rate: 5 kHz, and average output power: 800 mW. A typical visible near infrared pulse was slightly shorter than 5 fs in duration and covered the spectral range of 520–750 nm, within which it carried a nearly constant spectral phase, indicating that the pulses were nearly Fourier-transform limited. The pulse energies of the pump and probe were about 35 and 5 nJ, respectively. All the measurements were performed at room temperature (295 ± 1 K).

3. Results and discussion

3.1. Fourier analysis of vibrational signals

Fig. 1 shows the absorption and fluorescence spectra, and the output spectrum of the sub-5 fs NOPA laser. The fluorescence spectra, 3–8, change with the excitation wavelength. This is ascribable to reabsorption, which reduces the fluorescence intensity in the shorter wavelength range when excited at a longer wavelength. The fluores-cence peak shifts from 500 to 580 nm when the excitation wavelength is moved from 470 to 530 nm. This is because the absorption depth for a longer wavelength excitation is deeper so that the effect of reabsorption is more serious.

Fig. 2a shows the change in the absorbance as a function of the pump–probe delay time at various probe photon energies. The signals are composed of slow dynamics and highly oscillating signals. The former and the latter are mainly assignable to the electronic and vibrational dynam-ics. However, the slow dynamics have a little complicated feature, namely, they seem to be composed of very slow oscillation with an oscillation half period as long as 1 ps. It is hard to determine the vibrational decay time, because the low vibrational mode corresponds to 16 cm1. Highly oscillatory signals exist up to 600 fs after excitation. The results of our Fourier transform (FT) analysis of these oscillatory signals are shown inFig. 2b. This FT analysis was performed after averaging over 100 fs to remove the slow decay dynamics due to those in the relevant electronic states. There are two distinct components at 1469 and

Fig. 1. Electronic spectra of QT1: (1) absorption spectrum, (2) output spectrum sub-5 fs NOPA laser, and florescence spectra excited at (3) 530, (4) 520, (5) 510, (6) 500, (7) 480, and (8) 470 nm.

(3)

911 cm1. Since the latter one also appears for the sample without the solute molecule, it can be attributed to the sol-vent molecule of THF[24]. The former one, assigned to the C–C stretching mode, is also found in the Raman spectra excited with 488 nm, near-resonant to the electronic transi-tion of S1 S0.

At probe wavelengths (photon energies) of 709 nm (1.75 eV) and 589 nm (2.10 eV), a very intense peak appears at 911 cm1. This peak also appears in the Raman spectrum at the same wavenumber, as shown inFig. 3. This corresponds to the totally symmetric breathing mode in the solvent molecule. This mode is the most intense signal in some spectral range, and there are several other modes at 1450, 1225, 618, and 590 cm1. The FT power spectra pro-vide many modes to be assigned to the QT1 solute instead of the THF solvent molecule. Since the overlap of the laser spectrum and absorption spectrum of QT1 is smaller than 1000 cm1, the vibrational modes with wavenumbers higher than this are excited only in the ground electronic state. This can also be concluded from the phase of the vibration. The vibrational modes were first assigned using Ref.[25], as summarized below.

The Raman spectrum of regioregular poly(3-decylthi-ophene) excited at 514.5 nm was studied. The band at

1518 cm1was assigned to the Ca–Cbantisymmetric mode

in the aromatic thiophene ring [26,27]. The band near 1445–1455 cm1is due to symmetric Ca–Cbstretching

defor-mations. The band at 1378 cm1is due to Cb–Cbstretching

deformations in the aromatic thiophene ring, and the peak at 726 cm1is due to C–S–C ring deformations.

From the assignment of the vibration modes observed in the Raman spectra, the peaks at 1469–1473 cm1, 1379– 1375 cm1, and 700–704 cm1in the power spectra (shown inFig. 3a–c) of the real-time data can be attributed to the Cb–Cband Ca–Cbstretching modes and C–S–C ring

defor-mation mode, respectively. The change of the assignment between Ca–Cb and Cb–Cb originates from the difference

in the p-electron distributions in the aromatic and quinoi-dal thiophene rings.

3.2. Spectrogram analysis of time dependent vibrations In order to discuss the dynamics of the interaction among the modes, the spectrograms for the real-time data were calculated with a Blackman window function with a 150 fs window width. The instantaneous wavenumbers and intensities of the peaks are plotted inFig. 4a, b against the delay time of peak of the Blackman window.

(a) (b)

Fig. 2. Real-time vibration spectra and FT power spectra of the vibrations of QT1 at five different probe wavelengths: (a) changes in the absorbance vs. the pump–probe delay time and (b) Fast FT (FFT) power spectra of real-time vibration spectra.

(4)

Fig. 5shows the Fourier power spectra of the molecular vibration real-time signal together with those of wavenum-ber and intensity modulations of the 1469 cm1component in the spectrogram at 624 nm below 300 cm1. In the wave-number and intensity modulation spectra probed at 624 nm, there are two peaks located at 90 and 182 cm1, and 130 and 260 cm1, respectively. The difference in the wavenumbers of 90 and 130 cm1 is very close to that between the peaks at 1469 and 1379 cm1, and 1603 and 1469 cm1, respectively. One could argue that the two spec-tral components within the FT of this window can couple with each other through an artificial interference. There-fore, the observed modulation with frequencies of 90 and 130 cm1would in principle be ascribed to this artifact.

However, this possibility can be ruled out for the follow-ing two reasons:

In the first place, the second harmonic of the modula-tion wavenumbers at 182 cm1 (90 cm1· 2) and 260 cm1(=130 cm1· 2) are observed with high inten-sities. Their ratios, [I(182)/I(90) = 0.28 and I(260)/ I(130) = 0.50], cannot be expected for ordinary overtones.

The second reason is that the peaks at 90 and 182 cm1 appear only in the intensity modulation whereas those at

130 and 260 cm1 only in the frequency modulation. If they were artificial interferences, these modulations should have appeared simultaneously.

The artifact being ruled out, this behavior can be explained in terms of dynamic mode coupling [28,29]. Experimentally, the two sets of components, (1469 and 1379 cm1) and (1603 and 1469 cm1), appear alternatively in the frequency or intensity modulation, as reported above.

It is difficult to assign the modes of the 90 and 130 cm1 vibrations. The presence of the strong overtone indicates that they are out-of-plane bending modes, because the mir-ror symmetry of the p-electron distribution with respect to the molecular plane and deformation in the out-of-plane mode has a planar configuration in one oscillation period. The two modes of 90 and 130 cm1can be accounted for in the following way from the above observation. The former mode is probably such that modifies the intensity of 1469 cm1 (the symmetric Cb–Cb stretching mode), while

the latter modifies the bond order of the Cb–Cb bond.

The former mode corresponds to the processes in which the symmetric Cb–Cb stretching mode couples with the

Ca–Cb stretching deformation through the out-of-plane

bending mode of 90 cm1. The latter effect may be

(a) (d)

(b) (e)

(c) (f)

Fig. 3. FFT power spectra and Raman spectra of QT1. FFT spectra probed at (a) 1.99, (b) 2.03, and (c) 2.2 eV; in arbitrary units but mutually comparable, (d) Raman spectra excited at 488 nm, (e) and (f) expanded Raman spectra.

(5)

rephrased as that the vibration energy is transformed between the symmetric Cb–Cb stretching mode of

1469 cm1 and the other mode of 1603 cm1, which is

probably the antisymmetric Cb–Cb stretching mode,

through the 130 cm1out-of-plane bending mode. In both cases, the three modes, x1, x2, and x3, involved in the

couplings satisfy the Fermi resonance condition, x1+

x2= x3. A similar effect has been reported for

polydiacet-ylene[1].

In summary, we conclude that the two out-of-plane bending modes mediate the dynamic mode coupling between 1469 and 1379 cm1 and that between 1603 and 1469 cm1. These couplings are considered to be a model associated with the geometrical relaxation relevant to the formation of nonlinear excitations such as neutral bipola-ron (self-trapped exciton, exciton bipolabipola-ron) and charged bipolarons in all-s-trans polyacetylene.

Acknowledgements

The authors are grateful to Mr. Akira Ozawa for his help in pump–probe experiment. This research is partly supported by Grant-in-Aid for Specially Promoted Re-search (#14002003) and partly supported by the program for the Promotion of Leading Researches in Special Coor-dination Funds for Promoting Science and Technology

(a)

(b)

Fig. 4. Contour maps of the spectrograms for the real-time data calculated at 610 nm (a) and 624 nm (b). Thick and thin lines denote intensity and frequency modulations, respectively.

Fig. 5. Fourier power spectra. Thick and thin solid lines denote FT spectra of intensity and frequency modulations, respectively, and dashed line the real-time spectra. All lines were analyzed at the probe wavelength of 624 nm.

(6)

from the Ministry of Education, Culture, Sports, Science and Technology.

References

[1] T. Kobayashi, M. Yoshizawa, U. Stamm, M. Taiji, M. Hasegawa, J. Opt. Soc. Am. B 7 (1990) 1558.

[2] U. Stamm, M. Taiji, M. Yoshizawa, K. Yoshino, T. Kobayashi, Mol. Cryst. Liq. Cryst. A 182 (1990) 147.

[3] R.N. Marks et al., Chem. Phys. 227 (1998) 49.

[4] M. Muccini, E. Lunedei, A. Bree, G. Horowitz, F. Garnier, C. Taliani, J. Chem. Phys. 108 (1998) 7327.

[5] F. Garnier, R. Hajlaoui, A. Yassar, P. Srivastava, Science 265 (1994) 1684.

[6] L.L. Miller, Y. Yu, E. Gunic, R. Duan, Adv. Mater. 7 (1995) 547. [7] D. Fichou, J.-M. Nunzi, F. Charra, N. Pfeffer, Adv. Mater. 6 (1994)

64.

[8] Y. Furukawa, M. Akimoto, I. Harada, Synth. Met. 18 (1988) 151. [9] A. Yang, S. Hughes, M. Kuroda, Y. Shiraishi, T. Kobayashi, Chem.

Phys. Lett. 280 (1997) 475.

[10] A.J. Heeger, S. Kivelson, J.R. Schrieffer, W.-P. Su, Rev. Mod. Phys. 60 (1988) 781.

[11] H. Kiess (Ed.), Conjugated Conducting Polymers, Springer-Verlag, Berlin, 1992.

[12] G. Horowitz, D. Fichou, X.Z. Peng, Z.G. Xu, F. Garnier, Solid State Commun. 72 (1997) 381.

[13] A. Dodabalapur, L. Torsi, H.E. Katz, Science 268 (1995) 270. [14] N. Noma, T. Tsuzuki, Y. Shirota, Adv. Mater. 7 (1995) 647. [15] R. Hajlaoui, G. Horowitz, F. Garnier, Adv. Mater. 9 (1997) 389. [16] F. Geiger, M. Stoldt, H. Schweizer, P. Bauerle, E. Umbach, Adv.

Mater. 5 (1993) 922.

[17] T. Noda, H. Ogawa, N. Noma, Y. Shirota, Adv. Mater. 9 (1997) 720. [18] K. Uchiyama, K. Akimichi, S. Hotta, H. Noge, H. Sakaki, Synth.

Met. 63 (1994) 57.

[19] T. Takahashi, K. Takimiya, T. Otsubo, Y. Aso, Organic Lett. 7 (2005) 4313.

[20] T. Takahashi, K. Matsuoka, K. Takimiya, T. Otsubo, Y. Aso, J. Am. Chem. Soc. 127 (2005) 8928.

[21] A. Shirakawa, I. Sakane, T. Kobayashi, Opt. Lett. 23 (1998) 1292. [22] A. Shirakawa, I. Sakane, M. Takasaka, T. Kobayashi, Appl. Phys.

Lett. 74 (1999) 2268.

[23] A. Baltuska, T. Kobayashi, Appl. Phys. B: Lasers Opt. 75 (2002) 427.

[24] A. Ozawa, T. Kobayashi, J. Appl. Phys., submitted for publication. [25] G. Louarn, M. Trznadel, J.P. Buisson, J. Laska, A. Pron, M.

Lapkowski, S. Lefrant, J. Phys. Chem. 100 (1996) 12532.

[26] G. Louarn, J.Y. Mevellec, J.P. Eckhardt, S. Lefrant, J. Chim. Phys. Phys.-Chim. Biol. 89 (1992) 897.

[27] Y. Furukawa, M. Akimoto, I. Harada, Synth. Met 18 (1987) 151. [28] T. Kobayashi, M. Yasuda, S. Okada, H. Matsuda, H. Nakanishi,

Chem. Phys. Lett. 267 (1997) 472.

[29] A. Vierheilig, T. Chen, P. Waltner, W. Kiefer, A. Materny, A.H. Zewail, Chem. Phys. Lett. 312 (1999) 349.

數據

Fig. 1 shows the absorption and fluorescence spectra, and the output spectrum of the sub-5 fs NOPA laser
Fig. 2. Real-time vibration spectra and FT power spectra of the vibrations of QT1 at five different probe wavelengths: (a) changes in the absorbance vs
Fig. 5 shows the Fourier power spectra of the molecular vibration real-time signal together with those of  wavenum-ber and intensity modulations of the 1469 cm 1 component in the spectrogram at 624 nm below 300 cm 1
Fig. 4. Contour maps of the spectrograms for the real-time data calculated at 610 nm (a) and 624 nm (b)

參考文獻

相關文件

Reading Task 6: Genre Structure and Language Features. • Now let’s look at how language features (e.g. sentence patterns) are connected to the structure

The Secondary Education Curriculum Guide (SECG) is prepared by the Curriculum Development Council (CDC) to advise secondary schools on how to sustain the Learning to

Step 1: With reference to the purpose and the rhetorical structure of the review genre (Stage 3), design a graphic organiser for the major sections and sub-sections of your

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

We explicitly saw the dimensional reason for the occurrence of the magnetic catalysis on the basis of the scaling argument. However, the precise form of gap depends

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement.. 