Article pubs.acs.org/Macromolecules
Morphology and Optical Properties of P3HT:MEH-CN-PPV Blend Films Matthias A. Ruderer,† Cheng Wang,‡ Eric Schaible,‡ Alexander Hexemer,‡ Ting Xu,§ and Peter Müller-Buschbaum†,* †
Lehrstuhl für Funktionelle Materialien, Physik-Department, Technische Universität München, James-Franck-Strasse 1, 85748 Garching, Germany ‡ Advanced Light Source, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, California 94720, United States § Department of Materials Science and Engineering and Department of Chemistry, University of California Berkeley, Berkeley, California 94720, United States ABSTRACT: Thin photoactive polymer blend films of poly(3-hexylthiophene2,5-diyl) (P3HT) and poly(5-(2-(ethylhexyloxy)-2-methoxycyanoterephthalyliden) (MEH-CN-PPV) are investigated. The morphology is probed as a function of blend ratio (21, 28, 44, 54, and 70 wt % P3HT) and annealing using imaging techniques and soft X-ray scattering. The surface structure is detected with optical microscopy and atomic force microscopy (AFM), the inner film morphology and the near-surface structure with grazing incidence resonant soft X-ray scattering (GI-RSoXS) using different X-ray energies. Characteristic lateral structures determined with GI-RSoXS are in agreement with AFM observations and complemented with optical microscopy. The topography and the inner film morphology have the same structural length scales. Grazing incidence wide-angle X-ray scattering (GIWAXS) results confirm the crystallinity of the P3HT domains, which is increasing with annealing, and shows no indication for crystallinity in MEH-CN-PPV. In addition, GIWAXS measurements reveal a blend ratio dependent orientation of P3HT crystals. Absorption and photoluminescence measurements complement the structural investigations.
■
polymers is used as the active layer in organic solar cells.31,32 However, these so-called all-polymer systems have suffered from low efficiencies so far, perhaps due to a limited understanding. The most efficient all-polymer solar cells are based on the combination of different polyphenylenevinylenes (PPVs) and polythiophenes and show efficiencies of approximately 2%15,33−35 which is significantly below the values achieved with other organic systems.36 The low photovoltaic performance was attributed to the typical low electron mobilities and suboptimal LUMO levels in conjugated polymers. Moreover, large-scale phase separation is a significant challenge for this class of materials.29 It was stated that the synthesis of new electron-accepting polymers with high electron mobilities and low LUMO levels, which drive the charge separation, would improve the photovoltaic performance.31 However, an advantage of using polymers as electron accepting materials instead of fullerene derivatives is the higher variability in terms of adapting the chemical structure of polymers.37−41 Moreover, the mechanism of structure formation on the mesoscale are less complex than in blends with small molecules and relatively pure phases are achieved in
INTRODUCTION Semiconducting polymers have attracted strong interest due to the large variety of applications in organic electronics such as organic field effect transistors (OFETs), organic light emitting diodes (OLEDs), and organic photovoltaics (OPVs).1−8 These polymers give access to lightweight and low cost devices with a high degree of tunability, which results in applications beyond the common solid state devices. Moreover, making use of wetchemical preparation routes allows alternative thin film preparation methods, such as print technologies,9−11 which are inaccessible for devices fabricated under ultrahigh vacuum conditions. Thus new competitive technologies arise from the use of semiconducting polymers. However, due to the complexity of these polymers concerning molecular structure and physical behavior the fundamental understanding is still insufficient.12−16 The morphology of the active layer is of paramount importance in the photon to electron conversion process.8,9,17 Commonly semiconducting polymers are rod-like and semicrystalline and therefore differ significantly from the well-investigated coil-shaped, amorphous polymers such as for example polystyrene, polyisoprene, and polyacrylates.18−20 In particular, blends of two semiconducting polymers are more complex as their standard counterparts and significantly less well investigated.21−30 Besides the combination of a conjugated polymer with a fullerene derivative, the composition of two conjugated © XXXX American Chemical Society
Received: April 4, 2013 Revised: May 16, 2013
A
dx.doi.org/10.1021/ma4006999 | Macromolecules XXXX, XXX, XXX−XXX
Macromolecules
Article
recorded under an angle of 90°. The emission monochromator (optical grating) had a wavelength range of 200−900 nm. A redsensitive Hamamatsu R928 photomultiplier detected the signal. The scan speed was set to 500 nm/min and the slits were adjusted in such a way that the photomultiplier does not saturate. Imaging Techniques. An Axiolab A microscope (Carl Zeiss) in combination with a PixeLink USB Capture BE 2.6 CCD camera and five different objectives with magnifications from 1.25× to 100× were used fort he collection of optical micrographs. Atomic force microscopy (AFM) data were recorded with an Autoprobe CP Research (Veeco Metrology Group) instrument in tapping mode. A gold covered, conical shaped silicon tip with a curvature radius of 10 nm (Ultralever canitlevers), mounted on a cantilever with a force constant of 2.1 N/m (resonance frequency 80 kHz) was used. The images were recorded with sizes from 1 × 1 μm2 up to 20 × 20 μm2. Every image consisted of 256 line scans which were collected to form a 2d image. To obtain representative information every sample was measured at several positions. For a quantitative analysis the routemean-square (rms) roughness and the power spectral density (PSD) were determined. For this purpose a Fourier transformation of the 2d images was carried out and radially averaged. The PSD curves of different scan sizes from one sample were merged to one master curve. The PSD master curve exhibited lateral, characteristic structures of the sample surface in reciprocal space and was therefore suitable to be directly compared with scattering data. Near Edge X-ray Absorption Fine Structure (NEXAFS) Spectroscopy. To obtain the wavelength dependent refractive index (n = 1 − δ + iβ) of polymers for electromagnetic radiation near the absorption edge, near edge X-ray absorption fine structure (NEXAFS) spectra of polymer films were measured at the synchrotron beamline 11.0.1.2 of the advanced light source (ALS) at the LBNL in Berkeley (USA).50,51 The sample environment is identical to the GIRSoXS setup described below. The polymer films were prepared on SiN membranes and measured in transmission geometry. The absorption part of the index of refraction, β, was obtained through Beer’s law, I = I0 exp(−4πβz/λ). Here I is the transmitted intensity collected with a photodiode detector. I0 is the incident beam intensity transmitted from a bare SiN membrane measured with the same photodiode detector. z is the film thickness, and λ is the wavelength of the X-ray. The absorption spectra were taken near the carbon K-edge from 275 to 325 eV with 0.1 eV step size. The real part of the complex index of refraction, δ, was then calculated from the measured absorption, β, using Kramer−Kronigs relations. Grazing Incidence Resonant Soft X-ray Scattering (GIRSoXS). Grazing incidence small-angle X-ray scattering measurements with soft X-rays (GI-RSoXS) were performed at the synchrotron beamline 11.0.1.2.50,51 Due to the high absorption of soft X-rays in air, the full setup, including sample and detector, was kept in high vacuum. The energy of the X-rays was altered from 280 to 320 eV which corresponds to wavelengths from 4.4 to 3.9 nm. The intensity of the primary beam had a minimum at the absorption edge of carbon (285 eV) due to carbon contaminations of the optics in the beam path. Due to the long wavelengths of soft X-rays a sample−detector distance (SDD) of 18.5 cm was sufficient to probe length scales in the range from 21 nm to more than 2 μm. An incident angle αi = 2° similar to the critical angle (αc = 2.3° and 1.5° for X-ray energies of 280 and 283 eV) of the investigated blend films was chosen. An in-vacuum CCD camera with 2048 × 2048 pixel2 (pixel size of 13 × 13 μm2) was used as detector. Since the whole setup was installed in a vacuum chamber several samples were mounted at once to avoid extensive venting and pumping of the chamber. The measuring time of one scattering pattern was on the order of 1 s or less. Grazing Incidence Wide-Angle X-ray Scattering (GIWAXS). The intermolecular structure and orientation regarding the substrate was detected via the surface-sensitive X-ray technique grazing incidence wide-angle X-ray scattering (GIWAXS). The experiments were performed at the beamline 7.3.3 of the ALS.52 A wavelength of 0.124 nm and a SDD of 196 mm were chosen there. The scattered intensity was recorded by a four quadrant Quantum CCD X-ray detector (ADSC) with 2304 × 2304 pixels2 of size 81.6 × 81.6 μm2.
polymer blends. In particular, the fast diffusion of fullerene derivatives in conjugated polymers42−45 does not take place in all-polymer systems. In addition, it was shown that in the case of the combination of poly(3-hexylthiophene-2,5-diyl) (P3HT) and poly((1-methoxy)-4-(2-ethylhexyloxy)-p-phenylenevinylene) (MEH-PPV) a minimum structure size was found at the critical blend ratio calculated from the simple Flory− Huggins theory.46 Thus the simple Flory−Huggins approach might still be useful for blends of conducting polymers. Similar to polymer:fullerene systems, the morphology formation in allpolymer systems is a self-assembly process and influenced by external parameters.8 In this investigation, the optical properties and morphology of an all-polymer system is investigated as a function of the annealing temperature and blend ratio using spectroscopy, imaging and scattering techniques. The blend films contain P3HT and poly(5-(2-(ethylhexyloxy)-2-methoxycyanoterephthalyliden) (MEH-CN-PPV), two semicrystalline and conducting polymers. To what extend the simple Flory−Huggins approach for the calculation of the critical blend ratio is suited for such system, is tested. Both materials have been used separately in the most efficient all-polymer solar cells.15,33 However, no study using the combination of both polymers in one system has been published yet. In such a combination, P3HT could act as the electron donor and MEH-CN-PPV as the electron acceptor. After the characterization of the spectral properties of this system, the crystalline properties of pure P3HT and pure MEH-CN-PPV as well as of the corresponding blends are presented. In addition to the investigations of the crystalline structure, the options of grazing incidence resonant soft X-ray scattering (GI-RSoXS) in terms of tuning the X-ray energy are demonstrated. GI-RSoXS is a novel technique, which is in particular suited for complex systems such as polymer blend films. So far resonant soft X-ray scattering (RSoXS) already proved to successfully assess spatial dimensions of phase-separated domains in OPV systems.47,48 In addition GI-RSoXS allows for detecting surface and inner structures separately.
■
EXPERIMENTAL SECTION
Sample Preparation. Poly(3-hexylthiophene-2,5-diyl) (P3HT) and poly(5-(2-(ethylhexyloxy)-2-methoxycyanoterephthalyliden) (MEH-CN-PPV) were purchased from Rieke Metals Inc. and Sigma Aldrich, respectively. P3HT had a molecular weight of Mw= 50 kg/mol and MEH-CN-PPV of Mw= 25 kg/mol. Both polymers were dissolved in chloroform (CF) and mixed with different ratios (21, 28, 44, 54, and 70 wt % of P3HT). By adapting the concentration thin films with a constant thickness of about 70 nm were prepared on acidic precleaned silicon substrates49 via spin coating. The films were annealed at 200 °C for 10 min in air. Degradation due to this annealing is not expected to impact on the structure probed in this investigation but will be disadvantageous for devices. Absorption. Absorption data were taken with an UV/vis spectrometer (Lambda35, PerkinElmer) in a wavelength range from 260 to 1100 nm (scan speed of 120 nm/min and a slit width of 1 nm). The measurements were performed in transmission geometry. The data were corrected with an uncoated cleaned glass substrate and transformed into the absorbance via Lambert−Beer law. Photoluminescence Spectroscopy. The fluorescence spectrometer LS55 (PerkinElmer) was used to detect the PL signal. The light source was a Xenon discharge lamp with a pulse width at half height of
