1.

Introduction

Conducting polymers have received a great deal of attention owing to their advantages, including flexibility, ease of fabrication, and low production cost. Organic light emitting diodes, organic thin film transistors (OTFTs), and organic photovoltaic (OPV) cells are the applications of those polymers.^{1, 2, 3} OPV cell probably enjoys the most attractive application nowadays because it can convert solar energy into electricity. Nowadays, the power conversion efficiency (PCE) of polymer-based OPV cells has already reached ∼7 to 8%.^{4, 5} In order to further enhance the PCE, understanding of carrier transport mechanisms within the OPV cells is essential.

There exist different techniques to study carrier transport properties. Examples are time-of-flight (TOF), OTFT, dark-injection space-charge-limited current (DI-SCLC), charge extraction by linearly increasing voltage (CELIV), and admittance spectroscopy (AS).⁶ However, not all the techniques are suitable for the organic thin film measurements. For TOF and DI-SCLC, there is a minimum requirement for the film thickness due to the penetration depth of incident laser and RC effect of the samples, respectively.⁷ For OTFT, it can be used to determine the carrier mobilities in thin film (∼100 to 200 nm). However, the transport direction of the carriers is parallel to the substrate.⁸ The extracted value may be different from OPV devices, in which the carriers transport in a direction perpendicular to the substrate. CELIV is a popular technique to measure the organic thin film in a perpendicular direction.⁹ For CELIV to be useful, free carriers are required in the organic materials. Otherwise, an additional step is needed to generate the free carriers in samples.

In this contribution, we show that AS is a viable technique to study carrier transport properties of organic thin films. The hole mobilities of poly(3-hexylthiophene) (P3HT) based materials, including pristine P3HT and P3HT:[6,6]-phenyl-C61-butyric acid methyl ester (PCBM), are investigated. In order to obtain ohmic contact, poly(3,4-ethylenedioxythioplene) doped with poly(strenesulfonate) (PEDOT:PSS) is inserted between the anode and the organic layer to reduce the hole injection barrier. In the study of P3HT:PCBM, copper phthalocyanine (CuPc) is inserted between the cathode and the organic layer and acts as an electron blocking layer to reduce the effect of negative capacitance. The chemical structures of different materials are shown in Figs. 1, 1, 1, 1, while the work functions of anodes and the highest occupied molecular orbital (HOMO) of P3HT are shown in Fig. 1.

Fig. 1

Chemical structures of (a) P3HT, (b) PCBM, (c) PEDOT:PSS, and (d) CuPc. (e) Work functions of anodes and HOMO level of P3HT.

2.

Measurement Technique

AS is a very useful technique for determining carrier transport properties in organic semiconducting materials.^{10, 11, 12, 13} Generally, the device structure is anode/organic layer/cathode. To measure a reliable hole mobility, an ohmic contact is required between the anode and organic layer. In addition, there should be a large electron injection barrier by using a large work function cathode. If the sample is subjected to a dc biased voltage V_dc and a small ac signal υ_ac with frequency f, the frequency dependent admittance Y can be measured by an impedance analyzer. The admittance can be described by the following equation:

Fig. 2

(a) Typical frequency dependent capacitance signal. (b) Negative differential susceptance versus frequency plot.

In order to extract the hole mobilities, the negative differential susceptance −ΔB is first computed from the measured C using the following equation:

3.

Experimental Details

In this study, P3HT was purchased from three different corporations, namely, Rieke Metals, Inc. (P200), Lum. Tech. (LT-S909), and Aldrich (445703). PCBM was obtained from Lum. Tech. (LT-S095). The PEDOT:PSS (Clevios P VP Al 4083) solution was obtained from H.C. Starks GmbH. The pre-patterned indium tin oxide (ITO) glass substrates were purchased from Tinwell Technology, Ltd., which were used as the bottom anode in the measurements. Before the deposition of organic layers, the ITO substrates were first cleaned by deconex. Then, deionized water and filtered acetone were used to rinse the substrates in an ultrasonic bath. Finally, the ITO substrates were exposed to UV-ozone for 13 min. in order to remove residual hydrocarbons on the surface.

For AS and J–V measurements, the structure of the device is ITO/PEDOT:PSS/P3HT or P3HT:PCBM/Au. The PEDOT:PSS solution was filtered by a 0.45-μm polyvinylidene fluoride (PVDF) filter and then it was spin-coated on the pretreated ITO for 90 s at 1000 rpm. The PEDOT:PSS layer was annealed at 150ºC for 30 min in ambient in order to remove residual water. The film thickness of PEDOT:PSS was about 90 nm. Prior to the deposition of P3HT-based materials, it was first dissolved in 1,2-dichlorobenzene (DCB) with a concentration of 3 wt. % and stirred for 16 h at 40ºC in dry nitrogen atmosphere inside a glove box. For the P3HT:PCBM blend, the raw materials were mixed in a ratio of 1:1 by weight and dissolved in DCB. The solutions were then spun on the top of the annealed PEDOT:PSS layer. The spinning speed and timewere 600 rpm and 30 s, respectively. Afterward, all the samples were dried naturally inside the glove box for 16 h under dark conditions. The resulting film thicknesses of P3HT-based materials were about 350 to 400 nm. After the spin-coating of organic layers, a gold electrode (the cathode) was coated on the polymer by thermal evaporation under a base pressure of 10^#x2212;6 Torr. The thickness of an Au electrode was 100 nm. The coating rate of Au was kept at 0.5 nm/s. In some P3HT:PCBM devices, 10 nm of CuPc was deposited on top of the P3HT:PCBM layer before cathode fabrication. After all the fabrications, the measurements were performed in vacuum (∼10⁻³ Torr) inside a liquid nitrogen cooled cryostat (Oxford Instrument, Model ITC 601). In AS measurement, the admittances of the devices were measured by an impedance analyzer (Hioki E. E. Corp., Model 3532–50 LCR HiTESTER) as a function of frequency (f = 10³ − 5 × 10⁶ Hz). A fixed ac modulation at 50 mV was superimposed on the dc bias voltage (0 to 10 V) during measurements. For the current-voltage (J–V) measurement, the samples were connected to the source-measure-unit (SMU) (Keithley Instruments Inc., Model 236). The signal (corresponding current) was then captured by driving a voltage to the sample from the SMU.

4.

Results and Discussion

4.1.

Evaluation of Hole Transport Properties of Different Brands of P3HT

Figure 3 shows the experimental J–V characteristics of different brands of P3HT films. The thicknesses are in the range of 349 to 408 nm. The solid line is the simulated SCLC (J_SCL) curve, which is computed by:¹⁴

[TeX:] \documentclass[12pt]{minimal}\begin{document}\begin{equation} J_{{\rm SCL}} = \frac{9}{8}\mu _0 \varepsilon _0 \varepsilon _r \exp ( {0.89\beta \sqrt F })\frac{{F^2 }}{d}, \end{equation}\end{document}$$J_{\mathrm{SCL}}=\frac{9}{8}{\mu }_0{\varepsilon }_0{\varepsilon }_r\exp \left(0.89\beta \sqrt{F}\right)\frac{F^2}{d},$$

where μ₀ is the zero-field mobility, β is the Poole–Frankel slope, ε₀ is the permittivity in vacuum, and ε_r ≈ 3 is the dielectric constant. μ₀ = 7.59 × 10⁻⁴ cm²/Vs and β = −2.33 × 10⁻³ (V/cm)^−1/2 were obtained from TOF experiments.¹⁵ It represents the theoretically maximum current that can be present in the film. For the P3HT obtained from Rieke and Lum. Tech., there are excellent agreements between the data and the theoretical J_SCL. The J–V data for P3HT from Aldrich are slightly lower than the ideal J_SCL. However, the departure from the J_SCL is well within half an order, especially at high voltages. It is conceivable that shallow traps are present for this batch of P3HT. Additional evidence from AS will be provided below.

Fig. 3

J–V characteristics of different brands of P3HT films at T = 292 K. The solid line is the computed J_SCL with μ₀ = 7.59 × 10⁻⁴ cm²/Vs and β = −2.33 × 10⁻³ (V/cm)^−1/2 from Ref. 15.

For ease of discussion below, we label the different brands of P3HT by Brands A, B, and C that correspond to P3HT from Ricke, Lum. Tech., and Aldrich, respectively. Figure 4 shows the frequency dependent capacitances at T = 292 K and V_dc = 2 V, normalized to the geometric capacitance C_geo. At high frequency (400 kHz), a clear C_geo can be clearly observed for all samples. As the probing frequency is reduced, the capacitance decreases. For samples A and B, the capacitances remain nearly constant between 10⁴ to 10⁵ Hz. In contrast, for sample C, the capacitance in the same range of frequencies reaches a minimum at about 90 kHz, and increases again. From Sec. 2, we conclude that carrier trapping and dispersion are more severe in sample C, when compared to samples A and B. This conclusion is also supported by the J–V data in Fig. 3. Due to the trapping effect, the capacitance increases again at low frequency.

Fig. 4

(a) Frequency dependent capacitance and (b) negative differential susceptance plots of different brands of P3HT films at T = 292 K and V_dc = 2 V. The capacitance is normalized to the geometric capacitance.

Figure 4 shows the negative differential susceptance −ΔB versus frequency plot. The average transit time τ_dc can be obtained from the peak of the plot. As a result, the average hole mobilities can be calculated by Eq. 3. Figure 5 shows the extracted hole mobilities at different applied voltages at T = 292 K and the trends of the mobilities are in agreement with the Poole–Frenkel relation [TeX:] $\mu \propto \exp ({\beta F^{1/2}})$$\mu \propto \exp \left(\beta F^{1/2}\right)$. The extracted mobilities are in good agreement with the data obtained independently from TOF and CELIV techniques (as shown in Fig. 5), in which much thicker films are used for measurements.^{16, 17} The excellent agreement among the data sets establishes the reliability of the AS technique for the transport characterization of P3HT films. However, the average hole mobilities of P3HT are found to exhibit a negative field dependence. This phenomenon is widely observed in the P3HT transport measurements.^{16, 17} The origin(s) of this phenomenon is still under debate. One proposed explanation is that the positional disorder of materials is larger than its energetic disorder.¹⁸ Recently, it was observed that P3HT can form nanorods.¹⁹ It is conceivable that these nanorods may influence the low field transport by opening up alternative pathways. These pathways are not necessarily along the direction of the applied field and therefore, carrier mobility may decrease as the applied field increases.

Fig. 5

Field dependent hole mobilities of different brands of P3HT films at T = 292 K. The star and cross symbols are the data obtained from TOF and CELIV from Refs. 16 and 17, respectively.

4.2.

Evaluation of Hole Transport Properties of P3HT:PCBM

Apart from the pristine P3HT, the charge carrier transport properties of P3HT:PCBM are also investigated by AS. The structure of the sample is ITO/PEDOT:PSS/P3HT:PCBM/Au. In this experiment, we chose P3HT from Rieke Metals, Inc. as it provides a better agreement between the experimental data and theoretical simulation for both J–V and AS measurements. Furthermore, P3HT from Rieke appears to be the most popular material provider for OPV fabrication from the literature.²⁰ The frequency dependent capacitances of P3HT:PCBM film at T = 246 K and T = 292 K are shown in Fig. 6. The data appear to be markedly different from the case of pure P3HT films. Between 10⁵ to 10⁶ Hz, the data are similar to the case of P3HT. However, there is a rapid reduction of capacitance in lower frequencies. One proposed explanation is that there is minority carrier injection during the measurement.^{21, 22, 23} In this case, electrons are injected from the cathode and transport via the PCBM domains within the device. Thus, the active layer contains both holes and electrons. At the P3HT/PCBM interface, recombination occurs, resulting in the recombination current. This current generally lags behind the stimulus. As a result, there is a negative contribution to the frequency dependent capacitance. With the presence of the negative capacitance, it is difficult to determine the reliable value of the hole mobility.

Fig. 6

Frequency dependent capacitance of P3HT:PCBM films at different temperatures and V_dc = 6 V (a) without. CuPc layer and (b) with a CuPc layer. In (a), the dotted circle indicates the low frequency regime in which negative contribution to the capacitance occurs. In (b), the inset shows the negative differential susceptance plots of P3HT:PCBM films with a CuPc layer.

In order to suppress this effect, a material with a low electron mobility is inserted between the cathode and P3HT:PCBM layer to impede the injected electrons. In this study, CuPc is chosen since it exhibits electron trapping behavior after air exposure. Figure 6 shows the frequency dependent capacitances at T = 246 K with the presence of a CuPc layer. It can be clearly observed that the effect of negative capacitance is effectively contained. The extracted hole mobilities of P3HT:PCBM are shown in Fig. 7. The mobility of the P3HT:PCBM blend is similar to the case of pristine P3HT. This result indicates that the percolation pathways of P3HT are unaffected by the presence of PCBM domains. Similar results were obtained previously from time-of-flight techniques in which much thicker films were used.²⁴

Fig. 7

Field dependent hole mobilities of P3HT:PCBM films at T = 246 K with the CuPc electron blocking layer. For comparison, the hole mobilities of P3HT (Rieke) is also shown.

In addition to T = 246 K, we also show in Fig. 6 data at T = 292 K. In contrast to the former case (T = 246 K), we still observe a decrease (less severe) of the capacitance at low frequencies. In this case, with a higher temperature, it is likely that there is still non-negligible injection of electrons, contributing to the addition reduction of capacitance. The CuPc layer probably is not the most effective electron blocking material. Materials with either a high-lying LUMO or electron blocking capabilities will be tested to improve the AS signal.

5.

Conclusion

In conclusion, we have demonstrated that AS is a reliable technique to investigate the carrier transport properties of P3HT-based materials in a thin film (350 to 400 nm). For the pristine P3HT, the extracted hole mobilities are in good agreement with the data determined independently by TOF and CELIV techniques. For P3HT:PCBM, negative capacitance occurs at a low frequency due to the injection of electrons from the cathode. This phenomenon can be circumvented by inserting an electron blocking layer (CuPc). Thus, hole mobility can be determined reliably in a blend of P3HT:PCBM.