1. Introduction

In recent years, conjugated polymers have attracted much interest due to their ubiquitous applications in different areas of physics, including optics, electronics, photonics and others. Conjugated polymers possess a large variety of interesting and unique properties, and they are relatively simple to fabricate and flexible in use. This makes them attractive for implementation in electro-optics, organic transistors, photovoltaics and other areas.

Polythiophenes, and Poly-3-hexylthiophene (P3HT) in particular, are the most studied conjugated polymers with semiconducting properties, and have been applied widely in semiconductor devices, solar cells, diodes and detectors [1–5 ]. An essential interest to P3HT is caused by the ability to crystallize into 1D and 2D structures through the mechanism of $\pi -\pi$ stacking and van der Waals interactions [3]. The crystallization of P3HT typically takes place during controlled evaporation of a solvent after the solution of P3HT is deposited on a substrate [1]. In this case, the orientation of nanocrystals can be parallel to the substrate, while their in-plane ordering is random [4]. The morphology and properties of P3HT crystals are highly dependent on the growth conditions and can be identified by the presence of specific peaks at 525, 555 and 610 nm in the absorption spectrum [2, 6–9 ].

High aspect ratio 1D crystals of P3HT are called nanofibers and typically have length of several micrometers and width of around 20-30 nm. Nanofibers grown on a substrate, are interconnected with each other by tie-chain molecules and have a higher charge mobility in the direction of the backbone chain and the $\pi -\pi$ stacking axis than along the side chains [1, 10, 11 ]. For mobility sensitive applications, nanofibers with $\pi -\pi$ stacking axis parallel to the substrate are more desirable than those with perpendicular orientation, where the former and the latter orientation types are called edge-on and face-on respectively [2, 3, 5, 6, 12, 13 ].

The direction of growth of the P3HT crystals and the out-of-plane anisotropy of the films can be manipulated by external electric field [11, 14–16 ]. There also exist other techniques to control the in-plane orientation of the crystals [1, 13 ]. Among well explored ones are mechanical rubbing, epitaxial solidification, spherulitic crystallization, crystallization in microstructures and others [10, 11, 17–20 ].

In this study, we report the impact of electric field on the orientation and optical properties of the P3HT nanofibers dispersed in solvent [21]. As we show, applied DC and AC bias causes nanofibers to align along the electric field, whereas AC bias does it more efficiently (referred to as AC and DC poling). According to spectrophotomectric measurements the absorption of layers formed after electrical poling is polarization dependent. Time resolved measurements show that response time of nanofibers strongly depends on the magnitude of applied electric field and varies from miliseconds to the order of seconds for different poling conditions.

The impact of AC bias on the orientation of P3HT nanofibers was previously reported [22], however the optical properties of such structure were never studied. Moreover, as we show, the dynamic manipulation of optical anisotropy with response time 20 ms at driving voltage 1.3 $V\cdot \mu m^{-1}$ is possible. Such controllability is of great importance as introduces approach toward implementation of P3HT nanofibers on a new applicational level [23,24 ].

2. Materials and experiment

The solution containing anisole and P3HT nanofibers was prepared by the mixed solvent method [21]. We used P3HT with molecular weight 45.000 (GPC). The average length of the nanofibers was ≈2 μm and width ≈20 nm.

To implement electric field assisted alignment of the nanofibers, we fabricated a poling device, which consisted of a pair of silver electrodes printed with ink-jet technology on a glass substrate as in Fig. 1 . The distance between the electrodes was 200 μm, whereas the thickness of the silver layer was about 3 μm. According to numerical simulations, the configuration of electric field $\vec{E}$ in the gap should be uniform and approximated by an expression for parallel plate capacitor as $\left|\vec{E}\right|=V\cdot L^{-1}$, where V is the applied voltage and L is the gap width. This device was used for poling and spectrophotometry measurements of the material before and after poling.

 

Fig. 1 Schematic structure of the poling device. V – voltage, EL – printed silver electrodes, GS – glass substrate, D – sample droplet, L – spacing between the electrodes, NF – nanofibers.

Download Full Size | PDF

To create an aligned P3HT structure about 1 $\mu l$ of solution with nanofibers was drop-casted on the poling device, and the poling voltage was applied to the silver electrodes during approximately 10 minutes until the solvent was completely evaporated. After the poling process with DC bias, nanofibers inside the gap migrated toward the cathode and created randomly distributed openings in the dried layer. On the other hand, AC bias (50 Hz) did not lead to directed migration of the material.

In order to qualitively estimate the morphology of the nanofibers after AC poling, SEM imaging was used. Figure 2(a) shows the orientation of the nanofibers precipitated in the gap after evaporation of the solvent without applying the bias. Figure 2(b) shows the alignment of nanofibers in the gap after AC poling.

 

Fig. 2 SEM image of a) unpoled and b) AC poled nanofibers on the substrate between the electrodes $\left|\vec{E}\right|$ vector points the direction of the poling field.

Download Full Size | PDF

3. Results and discussions

To quantitatively estimate the influence of AC and DC poling on the ordering of P3HT nanofibers, we measured transmission spectra by focusing the beam from a broadband light source on the area between the electrodes.

Figures 3(a) and 3(b) display the absorption spectra of the unpoled and AC poled samples under polarized and unpolarized light, where $\perp$ - polarization is the one perpendicular to the electric field and, therefore, to aligned nanofibers. In the case of unpoled sample both polarizations are equally absorbed while for the poled one the polarization perpendicular to the nanofibers is absorbed stronger in the region 500 – 620 nm.

 

Fig. 3 Absorption spectrum of the unpoled (a) and AC poled samples (b).

Download Full Size | PDF

The ratio between absorption for $\perp$- and $\parallel$-polarized light (defined as dichroic ratio) at 550 and 610 nm is 1.44 and 1.51, respectively. The dichroic ratio of DC poled samples at 550 and 610 nm is lower and equal to 1.19 and 1.28, respectively, which indicates a weaker absorption due to poor layer homogeneity and a lower degree of anisotropy. The poling conditions giving a higher dichroic ratio and, thus, degree of anisotropy can be found by continious adjustment of the poling voltage, solvent viscosity, concentration and lateral size of nanofibers. However, absorption measurements are not best suited for such routine of optimization and alternative methods should be introduced instead.

The response of nanofibers to sine-shape AC bias of 50 Hz was evaluated by time resolved measurements. The transmission of 532 nm light passing through the droplet of solution containing P3HT nanofibers was monitored at the moment the poling voltage was turned on. To prevent the droplet from rapid drying another glass piece was placed on top using 100-150 μm thick spacers. The results show that the light polarized perpendicular to the nanofibers experiences a drop of transmission, while the parallel polarization increases as shown in Figs. 4(a) and 4(b) . The value of response time was taken at the stabilization point of the signal and depended strongly on the magnitude of applied field as shown in Fig. 5 . Suchwise, the response time for poling field 0.65 $V\cdot \mu m^{-1}$ was 110-150 ms, while for 0.25 $V\cdot \mu m^{-1}$ it was 2.9 s. The amplitude of the increase or drop was also proportional to poling field.

 

Fig. 4 The response time of P3HT nanofibers in anisole to 0.65 V/μm poling field. a) light polarized perpendicular to poling field, b) light polarized parallel to poling field and thus nanofibers.

Download Full Size | PDF

 

Fig. 5 The response time of P3HT nanofibers in anisole at various poling field.

Download Full Size | PDF

The overall phenomenon of poling can be reasonably explained by the redistribution of charge carriers inside the nanofibers. As P3HT nanofibers are p-type semiconductors, external electric field forces positive charges to move along the backbone chain and $\pi -\pi$ stacking axis. Since nanofibers are separated from each other and the solvent is nonconductive, holes are trapped inside the crystal domain and accumulated at either end of the nanofiber. In the case of DC bias, accumulated positive charges become dragged together with nanofibers toward the cathode. Such migration produces inhomogeneities in the layer of partially aligned nanofibers. In the case of AC bias, charge motion is limited along the $\pi -\pi$ stacking axis from one end of the nanofiber to another. Thus, the electric force acting on the nanofibers is continually changing direction and the resultant linear displacement becomes negligible. During evaporation of the solvent, nanofibers hold their orientation and precipitate on the substrate in an ordered manner.

The migration of supramolecules and carbon nanotubes under DC bias was reported previously and was explained by defects during fabrication step [25,26 ]. As P3HT nanofibers migrate under DC bias only towards cathode this phenomenon is unlikely to be attributed to dielectrophoretic origin. In turn, we assume that such migration happens due to additional positive charge acquired by P3HT nanofibers through solvent-P3HT interaction. It is known that P3HT in anisole can be oxidized under electric potential and become positively charged, while the solvent is reduced [1]. It is also reasonable to assume that charging plays an important role and can intensify the alignment at low voltages.

The response time measurement showed that the rotational relaxation time of nanofibers depends on the magnitude of applied AC field. This observation supports our hypothesis about electrical polarization and redistribution of charge carriers inside the nanofibers. Torque N rotating a symmetrical dipole with charge $q$ and length $l$ under electric field E is expressed by a classical expression:

From Eq. (1), it is clear that the dynamical response of the system should be faster with the factor of E . However, it is still unclear whether any repulsive interaction between polarized nanofibers is present, and what is the charge distribution along them. More sophisticated research is required to verify this model.

4. Conclusions

The influence of external electric field on the orientation of P3HT nanofibers and their optical properties was studied. Our research demonstrates that nanofibers in anisole get positively charged and respond to electric field by aligning along the field lines. Spectrophotometric measurements show that aligned nanofibers are optically anisotropic and AC bias leads to a higher ordering compared to DC. The response time of nanofibers in solution to AC bias decreases significantly with the magnitude of the voltage applied.

P3HT nanofibers, with their anisotropic properties, can be combined with LC as it is traditionally being done for creating optical modulating systems. However, we assume that nanofibers can be used individually without the host if reversible switching is organized by a pair of orthogonal electrodes. According to our observations the response of nanofibers in solution to 1.3 $V\cdot \mu m^{-1}$ poling field increases the transmission signal by 3 dB for the optical path of 100-150 μm. With further development P3HT nanofibers can be configured into a rapid and reversible component in an amplitude modulation device.