J. Phys. Chem. B 2008, 112, 2315-2318
2315
Electronic and Conformational Properties of the Conjugated Polymer MEH-PPV at a Buried Film/Solid Interface Investigated by Two-Dimensional IR-Visible Sum Frequency Generation Qifeng Li, Rui Hua, and Keng C. Chou* Department of Chemistry, UniVersity of British Columbia, VancouVer, BC V6T 1Z1, Canada ReceiVed: June 11, 2007; In Final Form: NoVember 13, 2007
We present the first measurement of the buried surface electronic states of the conjugated polymer poly[2methoxy-5-(2′-ethyl-hexyloxy)-1,4-phenylenevinylene] (MEH-PPV) using two-dimensional (2D) IR-visible sum frequency generation (SFG). SFG electronic spectra were obtained by scanning the frequencies of both incident visible and IR beams and used to study the surface electronic transitions associated with the C-C stretching of benzene rings located at the backbone of MEH-PPV. Because of the surface confinement effects, the polymer conformation, and consequently the electronic states, at the film/solid interface are different from those of the bulk film. Theoretical analysis based on an oligomer model was employed to estimate the conjugation-length distributions of MEH-PPV at interfaces. Assuming a Gaussian conjugation-length distribution, it was found that the conjugation-length distribution at the MEH-PPV/solid interface was centered at 5.8 monomer units. Similar surface effects were also observed at the air/polymer interface, with a shorter average conjugation length of 5.1 monomer units.
I. Introduction The optical and electronic properties of poly[2-methoxy -5-(2′-ethyl-hexyloxy)-1,4-phenylenevinylene] (MEH-PPV) have been studied intensively because of its broad applications in organic devices.1-6 MEH-PPV is characterized by a π-conjugated backbone, in which the π-electrons are delocalized over several monomer units along the carbon chains, forming π-bands.7,8 Because the delocalized orbitals are half-filled, the energy gap between the filled and empty bands results in semiconducting properties. The extent of delocalization of the π-electrons, the so-called conjugation length, determines the energy gap, which plays a major role in the optical and electrical properties of the materials and the performance of the organic devices they are used in. The bulk electronic and optical properties of MEH-PPV have been studied extensively by UV/ visible absorption and photoluminescence spectroscopy and have been shown to be highly dependent on the conformation of the polymer chains.9-12 Mechanisms leading to a finite conjugation length in the polymer due to abrupt flips13,14 and conformational disorder15 have been proposed. Despite enormous efforts, these organic semiconducting materials and optimization of the organic devices using them are still not well understood. Both experimental and theoretical investigations are required to meet this challenge. Compared to the bulk properties, the optical and electronic properties of conjugated polymers at a buried polymer/solid (film/solid) interface remain unexplored. In an organic device, the charge carriers, both electrons and holes, have to be injected through polymer/solid interfaces. Therefore, the band gaps of conjugated polymers at buried interfaces are important factors that affect the charge injection and overall efficiency of the organic devices. Because of the surface confinement effect at a polymer/solid interface, the surface chain conformation and the * E-mail:
[email protected].
surface band gap are expected to be different from those in the bulk. However, it has been a great challenge to measure the buried interfacial electronic states because of a lack of a suitable probing technique. Traditional techniques based on ultrahigh vacuum are not applicable to a buried interface, and absorption and emission spectroscopy do not have the necessary surface sensitivity. Recent developments in two-dimensional (2D) IR-visible sum frequency generation (SFG) spectroscopy have made it possible to study the optical properties of the conjugated polymer at a buried interface.16,17 SFG is known as a surface-sensitive tool because this second-order optical process is forbidden in centrosymmetric media, such as bulk polymers.18 As shown in Figure 1A, SFG is carried out by mixing an IR beam and a visible beam on a surface to generate a sum-frequency beam. Traditionally, IR-visible SFG vibrational spectroscopy has been carried out by tuning the incident IR frequency to obtain a surface vibrational spectrum, which reveals the surface chemical species. Recently, it has been shown that one can obtain a surface electronic spectrum by tuning the incident visible frequency.17,19 As shown in Figure 1B, SFG intensity is doubly enhanced when the IR is resonant with the vibrational state and the SFG is resonant with the surface electronic state. With the capability of tuning both the incident IR and visible frequencies, 2D SFG spectroscopy becomes a highly selective surface probe for studying the surface electronic states coupled to a specific vibrational mode. In this study, 2D SFG was used to measure the buried surface electronic states associated with the C-C stretching mode of benzene rings at the backbone of MEHPPV, as indicated in Figure 1C. Based on the measured SFG electronic spectrum, the conjugation-length distribution of MEHPPV at the buried interface was estimated using an oligomer model. The average conjugation length of MEH-PPV at the MEH-PPV/solid interface was estimated to be 5.8 monomer units. Similar surface effects were also observed at the air/
10.1021/jp0745135 CCC: $40.75 © 2008 American Chemical Society Published on Web 02/02/2008
2316 J. Phys. Chem. B, Vol. 112, No. 8, 2008
Li et al.
Figure 1. Energy diagrams of (A) IR-resonant SFG and (B) IR-visible doubly resonant SFG. |g0〉 is the ground state, |g1〉 is the first vibrational excited state, and |e0〉 is the electronic excited-state at vibrational ground state. (C) Experimental SFG setup for probing the buried interface and the structure of MEH-PPV.
polymer interface, with a slightly shorter average conjugation length of 5.1 monomer units. II. Experimental Section As shown in Figure 1C, 2D SFG involves mixing a frequency-tunable visible beam (ωvis) and a frequency-tunable infrared beam (ωir) on a surface to generate a third beam with a frequency ωSFG ) ωvis + ωir.18 The IR and visible laser beams were generated using a Nd:YAG laser at 1064 nm (30 ps, 50 mJ/pulse, and 10 Hz). The laser was used to generate harmonics at 532 and 355 nm in KTiOPO4 (KTP) crystals to pump two optical parametric generators/amplifiers (OPG/OPA) for generating frequency-tunable visible and IR beams. The tunable IR beam was produced by difference frequency mixing of the 1064 nm beam with the output of a KTP OPG/OPA pumped by the 532 nm beam. The frequency-tunable visible beam was generated in a BaB2O4 (BBO) OPG/OPA pumped by the 355 nm beam. The visible and IR beams were overlapped spatially and temporally on the sample at incident angles of 45° and 55°, respectively. Because of the large absorption coefficient of MEH-PPV in the visible region, the fluence of the visible beam was kept below 10 µJ/mm2 per pulse to avoid photodamage. The SFG intensity was detected using a photomultiplier tube after spectral filtering by a short-pass filter and a double monochromator. All SFG spectra were calibrated using a z-cut quartz crystal. MEH-PPV (molecular weight ∼55 000) was purchased from Sigma-Aldrich, Inc. The polymer films were spin-coated at ∼2000 rpm on CaF2 windows from a 2% w/v tetrahydrofuran solution. Films were annealed at 100 °C for several hours to evaporate the solvent in the film before spectroscopic measurements were taken. CaF2 was chosen because of its high transmission of IR. To ensure the measured SFG is truly generated at the buried polymer/CaF2 interface without a contribution from the air/polymer interface, the spin-casting process was repeated several times to obtain a thicker film. Because of the large absorption coefficient of MEH-PPV in the visible region, the incident visible beam and the SFG generated from the air/polymer interface were mostly absorbed by the film. Based on the absorption coefficient of MEH-PPV,20 a film thickness of ∼1 µm would guarantee that the incident visible beam and the SFG generated from the air/polymer interface in the reflected direction are at least 95% blocked for wavelengths shorter than 580 nm. However, the visible absorption spectra of MEH-PPV were measured using thin films.
Figure 2. (A) SFG vibrational spectra of MEH-PPV at the buried interface with various incident visible wavelengths. The vibrational peak located at 1593 cm-1 is the C-C stretching of benzene rings.
configuration (s-, s-, and p-polarized for SFG, visible, and IR, respectively). The vibrational band centered near 1593 cm-1 is assigned to the C-C stretching of benzene rings located at the backbone of MEH-PPV.21 The wavenumber of this mode has been reported at between 1583 and 1593 cm-1 depending on the molecular weight.22 The observed wavenumber at 1593 cm-1 is consistent with the previously reported value for MEH-PPV with a molecular weight of 6 × 104 g/mol.22 As shown in Figure 2, the electronic enhancement of the SFG vibrational spectra was observed when the visible wavelength is near 580 nm, which produces a SFG wavelength near 530 nm with an IR beam of 1593 cm-1. Because of the electronic resonance, the refraction index of MEH-PPV in the investigated region is wavelength-dependent.23 Therefore, Fresnel factors need to be considered to obtain the actual dispersion relation of the second-order nonlinear susceptibility. In the electric-dipole approximation, the SFG arising from the second-order polarization can be written as24 2 I(ωSFG ) ωIR + ωVis) ∝ |χ(2) eff :E(ωIR)E(ωVis)|
where χ(2) eff is the effective second-order nonlinear susceptibility tensor and E(ωIR) and E(ωVis) are the input fields. In the ssp configuration, the effective second-order nonlinear susceptibility χ(2) eff can be written as (2) χ(2) eff,ssp ) Lyy(ωSFG)Lyy(ωvis)Lzz(ωIR) sin(βIR)χyyz
(2)
with
Lyy(ωi) )
III. Results and Analysis Figure 2 shows the SFG vibrational spectra from the MEHPPV/solid interface with various visible wavelengths in a ssp
(1)
Lzz(ωi) )
2n1(ωi) cos(βi) n1(ωi) cos(βi) + n2(ωi) cos(γi) 2n2(ωi) cos(βi) n1(ωi) cos(γi) + n2(ωi) cos(βi)
(3)
(4)
Properties of the Conjugated Polymer MEH-PPV
J. Phys. Chem. B, Vol. 112, No. 8, 2008 2317
where χ(2) yyz is the yyz component of the second-order nonlinear susceptibility in the laboratory coordinate (defined in Figure 1C), Ljj(ωi) is the Fresnel factor, ni is the refractive index of medium i, βi is the incident angle, and γi is the refracted angle. The values of n2(ωi) for MEH-PPV reported by Tammer et al. were used for eqs 3 and 4.23 In general, there are two types of processes in IR-visible SFG. The first type starts with an electronic transition followed by a vibrational transition (vis-IR), and the second type begins with a vibrational transition followed by an electronic transition (IR-vis).19 Because the electronic relaxation times are generally much shorter than the vibrational relaxation times, the contribution of the vis-IR SFG is generally negligible.16 Therefore, only the IR-vis SFG will be considered in the following calculation. Assuming harmonic potential surfaces for the electronic states and the Born-Oppenheimer approximation, the IR-vis doubly 19 resonant χ(2) ijk can be described as
χ(2) ijk
)-
N p2
〈
µiegµjge
{
xSe-S
∂µkgg
∂ql ωIR - ωl + iΓl
∞
×
Sn
∑ n!
n)0
1
ωSFG - nωl - ωeg + iΓen,go
with Al describing the electronic resonance. For a system with a single electronic resonance ωeg, Al has the following form
-
1 ωSFG - (n + 1)ωl - ωeg + iΓen+1,go
}〉
Al ∝ (2) + χNR,ijk (5)
where N is the surface molecular density, µieg represents the i component of electronic transition moment, ql is the normal coordinate, S is the Huang-Rhys factor, n labels the vibrational state, g and e label the ground and excited electronic states, respectively, ωSFG is the SFG frequency, ωl and ωeg are the resonant vibrational and electronic frequencies, respectively, Γl and Γen,go are the damping constants, the angular brackets (2) indicate an average over molecular orientations, and χNR,ijk describes the non-resonant contributions. Equation 5 includes all the vibronic transitions series (HuangRhys series). However, experimentally the nonzero vibronic transitions have not been previously observed because the vibronic transitions have much shorter dephasing times than the zero-vibration transition.16,17 For MEH-PPV, the dephasing times of vibronic transitions are in the femtosecond region.25 Therefore, by assuming Γen,go . Γeo,go, the nonzero vibronic transitions can be neglected, and eq 5 can be simplified as
χ(2) ijk
〈
k
∂µgg xSe-S N ) - 2 µiegµjge × ∂ql ωIR - ωl + iΓl p
〉
1 (2) + χNR,ijk (6) ωSFG - ωeg + iΓe0,g0 It is worth pointing out that a significantly larger Γen,go also suppresses the aforementioned vis-IR SFG, which starts with an electronic transition followed by a vibrational transition. Based on eq 6, the vibrational spectra shown in Figure 2 can be fitted using
χ(2) ijk ∝
Figure 3. Absorption spectrum of bulk MEH-PPV film (solid line) and the surface SFG electronic spectra of MEH-PPV at MEH-PPV/ solid (2) and air/MEH-PPV interfaces (O). The dashed lines are theoretical fitting curves.
Al (2) + χNR,ijk ωIR - ωl + iΓl
(7)
1 ωSFG - ωeg + iΓe0,g0
(8)
Equation 1-4 and 7 were used to fit the SFG vibrational (2) spectra in Figure 2 with Al, ωl, Γl, and χNR,ijk as the adjustable parameters. The fitted values of |Al| for various visible wavelengths are plotted in Figure 3. The 2D SFG electronic spectrum is red-shifted with respected to an absorption spectrum commonly reported for a film, which centered near 500 nm. For MEH-PPV, multielectronic resonances should be considered. The band gap of a conjugated polymer is closely related to the conjugation length.7-9 Conjugated polymer chains consist of a series of connected segments, each of which has a different extent of π-electron delocalization. The extent of the conjugation is limited by the twists in the polymer backbone. The longer the segment is, the smaller the band gap. The theoretical methodology for describing the properties of conjugated polymers remains an active research area.26-30 Most theoretical work on predicting the optical properties of a conjugated polymer use an oligomer approach,31 in which the properties of oligomers of various chain lengths are calculated and then treated as separated subunits. Although the oligomer model does not have a full description of the material properties, such as medium effects and oligomer interactions,32,33 it has produced reasonably good predictions for the optical properties of ΜΕΗ-PPV34 and will be adopted in the following analysis. To estimate the corresponding conjugation-length distribution at the interface based on the SFG electronic spectra in Figure 3, it is assumed that the surface polymer chains consist of oligomers with various conjugation lengths. For a conjugationlength distribution function D(N), the SFG electronic spectrum is a sum of contributions from oligomers of different lengths, and the amplitude Al in eq 8 can be modified as
Al ∝
∑N ω
D(N) SFG
- ωe(N) + iΓe0
(9)
where ωe(N) ) EN/p is the electronic transition frequency associated with an oligomer of N monomer units. For MEH-
2318 J. Phys. Chem. B, Vol. 112, No. 8, 2008
Li et al. SFG. Surface SFG electronic spectra obtained at the MEH-PPV/ solid and air/MEH-PPV interface show that the surface band gaps and conjugation-length distributions of MEH-PPV are sensitive to surface effects. Based on the measured SFG electronic spectra and an oligomer model, the oligomers at the buried interface were estimated to have an average conjugation length of 5.8 monomer units and a distribution width of 0.9 monomer unit. Similar surface effects were also observed at the air/polymer interface with a shorter average conjugation length of 5.1 monomer units and a broader width of 1.3 monomer units. Acknowledgment. This work was financially supported by the Natural Sciences and Engineering Research Council of Canada and the University of British Columbia. References and Notes
Figure 4. Calculated conjugation-length distributions of MEH-PPV at (A) MEH-PPV/solid interface and (B) air/MEH-PPV interface.
PPV, Chang et al. have shown that the energy levels for oligomers of N monomer units can be described by34
EN ) E0 + 2β cos
(N +π 1)
(10)
with E0 ) 4.3 eV and β ) -1.1 eV.29 In this expression, the energy levels with N e 2 are located in the UV region. Therefore, only segments with N g 3 will be considered in the fitting. Assume a Gaussian conjugation-length distribution function
D(N) ) exp[-(N - N0)2/σ2]
(11)
The center conjugation length N0 and the distribution width σ for MEH-PPV at the interface can be derived by using eqs 9-11 to fit the SFG electronic spectrum in Figure 3. The corresponding fitting curve is shown in Figure 3. For the MEH-PPV/solid interface, the best fit was obtained with N0 ) 5.8 ( 0.2, σ ) 0.9 ( 0.1, and Γe0 ) 640 ( 40 cm-1. The conjugation-length distribution curve is shown in Figure 4A. For comparison, 2D SFG measurements were also carried out at the air/MEH-PPV interface. As shown in Figure 3, the SFG electronic spectrum at the air/MEH-PPV interface is slightly broader than that at the MEH-PPV/solid interface. The best fit was obtained for the air/MEH-PPV interface with N0 ) 5.1 ( 0.2 and σ ) 1.3 ( 0.1. A theoretical estimate for the conjugation-length distribution for a bulk film is not available for direct comparison with the current fitting results. However, these fitting results are within a reasonable range of N0 ) 5 and σ ) 1.8, as Chang et al. obtained for MEH-PPV in chloroform solution. A comparison of the polymer/solid and air/polymer interfaces indicates that a rigid surface confinement at the solid surface produces a longer average conjugation length with a narrower distribution width. A nonrigid surface confinement at the air/polymer interface shows a shorter average conjugation length with a larger distribution width. IV. Summary The optical, electronic and conformational properties of MEH-PPV located at interfaces were studied by 2D IR-visible
(1) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539. (2) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of Conducting Polymers, 2nd ed.; Marcel Dekker: New York, 1998. (3) Swager, T. M. Acc. Chem. Res. 1998, 31, 201. (4) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science 1995, 270, 1789. (5) Nguyen, T. Q.; Martini, I. B.; Liu, J.; Schwartz, B. J. J. Phys. Chem. B 2000, 104, 237. (6) Coakley, K. M.; McGehee, M. D. Chem. Mater. 2004, 16, 4533. (7) Holzer, W.; Penzkofer, A.; Gong, S. H.; Bradley, D. D. C.; Long, X.; Bleyer, A. Chem. Phys. 1997, 224, 315. (8) Bredas, J. L.; Cornil, J.; Beljonne, D.; dos Santos, D.; Shuai, Z. G. Acc. Chem. Res. 1999, 32, 267. (9) Schwartz, B. J. Annu. ReV. Phys. Chem. 2003, 54, 141. (10) Pope, M.; Swenberg, C. E. Electronic Processes in Organic Crystals and Polymers; Oxford University Press: New York, 1999. (11) Shi, Y.; Liu, J.; Yang, Y. J. Appl. Phys. 2000, 87, 4254. (12) Collison, C. J.; Rothberg, L. J.; Treemaneekarn, V.; Li, Y. Macromolecules 2001, 34, 2346. (13) Kohler, B. E.; Samuel, I. D. W. J. Chem. Phys. 1995, 103, 6248. (14) Yaliraki, S. N.; Silbey, R. J. J. Chem. Phys. 1996, 104, 1245. (15) Rossi, G.; Chance, R. R.; Silbey, R. J. Chem. Phys. 1989, 90, 7594. (16) Raschke, M. B.; Hayashi, M.; Lin, S. H.; Shen, Y. R. Chem. Phys. Lett. 2002, 359, 367. (17) Chou, K. C.; Westerberg, S.; Shen, Y. R.; Ross, P. N.; Somorjai, G. A. Phys. ReV. B 2004, 69. (18) Shen, Y. R. Nature 1989, 337, 519. (19) Hayashi, M.; Lin, S. H.; Raschke, M. B.; Shen, Y. R. J. Phys. Chem. A 2002, 106, 2271. (20) Koynov, K.; Bahtiar, A.; Ahn, T.; Bubeck, C.; Horhold, H. H. Appl. Phys. Lett. 2004, 84, 3792. (21) Wu, X. F.; Shi, G. Q.; Qu, L. T.; Zhang, J. X.; Chen, F. E. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 449. (22) Wu, X. F.; Shi, G. Q.; Chen, F. E.; Han, S. H.; Peng, J. B. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 3049. (23) Tammer, M.; Monkman, A. P. AdV. Mater. 2002, 14, 210. (24) Zhuang, X.; Miranda, P. B.; Kim, D.; Shen, Y. R. Phys. ReV. B 1999, 59, 12632. (25) Yang, X. J.; Dykstra, T. E.; Scholes, G. D. Phys. ReV. B 2005, 71. (26) Friend, R. H.; Bradley, D. D. C.; Townsend, P. D. J. Phys. D: Appl. Phys. 1987, 20, 1367. (27) Hagler, T. W.; Pakbaz, K.; Heeger, A. J. Phys. ReV. B 1994, 49, 10968. (28) Oberski, J. M.; Greiner, A.; Bassler, H. Chem. Phys. Lett. 1991, 184, 391. (29) Yu, J. W.; Fann, W. S.; Kao, F. J.; Yang, D. Y.; Lin, S. H. Synth. Met. 1994, 66, 143. (30) Gierschner, J.; Cornil, J.; Egelhaaf, H. J. AdV. Mater. 2007, 19, 173. (31) Electronic Materials: The Oligomer Approach; Mu¨llen, K., Wegner, G., Eds.; Wiley-VCH: Weinheim, Germany, 1998. (32) Salzner, U. Curr. Org. Chem. 2004, 8, 569. (33) Scholes, G. D.; Larsen, D. S.; Fleming, G. R.; Rumbles, G.; Burn, P. L. Phys. ReV. B 2000, 61, 13670. (34) Chang, R.; Hsu, J. H.; Fann, W. S.; Liang, K. K.; Chiang, C. H.; Hayashi, M.; Yu, J.; Lin, S. H.; Chang, E. C.; Chuang, K. R.; Chen, S. A. Chem. Phys. Lett. 2000, 317, 142.
