Electron Spin Echo of Photoinduced Spin-Correlated Polaron Pairs in P3HT: PCBM Composite

Electron Spin Echo of Photoinduced Spin-Correlated Polaron Pairs in P3HT:PCBM Composite

Abstract A variation of the electron spin echo signal caused by laser pulse in a blend of [6,6]-phenyl C₆₁ butyric acid methyl ester and poly(3-hexylthiophene) was detected. This variation was attributed to light-generated paramagnetic species in P3HT^PCBM blend, with non-equilibrium spin polarization.  The echo-detected electron paramagnetic resonance spectrum of these species closely resembles the time-resolved EPR spectrum of spin-correlated polaron pair PCBM^-/P3HT⁺ and was assigned to this pair. The characteristic times for polarization and coherence decay were measured for the PCBM^-/P3HT⁺ pair at 77 K. These times are long enough, which shows the possibility of the application of the ESE technique for studying spin evolution of light-generated charge transfer intermediates in composites of fullerenes and conductive polymers.

1 Introduction

Composites of fullerenes and conductive polymers are promising materials for plastic solar cells. Since bulk heterojunction is formed in these composites, they are widely used as an active layer in photovoltaic devices. While the light to electricity power conversion efficiency of such devices continues to increase, its presently achieved maximum value is about 7%, which is below the commercially interesting level.

Understanding the nature of the intermediates of light-driven charge separation in polymer/fullerene composites is crucial for optimization of their composition and morphology, which is the main way of increasing power conversion efficiency of plastic solar cells. In polymer/fullerene blends, the light-generated excitons dissociate within a few picoseconds, resulting in the formation of polaron pairs, with each of them consisting of a delocalized cation radical of the polymer and an anion radical of the fullerene. Since these polarons are paramagnetic, electron paramagnetic resonance is a method of choice for the study of post-nanosecond intermediates of light-driven processes in polymer/fullerene composites. The most widely used technique for such studies is continues-wave EPR under stationary light illumination of the sample. However, it lacks temporal resolution and thus does not allow one investigation of polaron pair evolution.

Recently, time-resolved EPR signal was detected for the polaron pair in the blend of [6,6]-phenyl C₆₁ butyric acid methyl ester and poly(3-hexylthiophene), the most popular polymer/fullerene composition at present. The TR ERR had emissive and the spectral shape was characteristic for geminate spin-correlated radical pair for the case of polymer/fullerene blends it is usually called “polaron pair”. It was concluded that for small values of delay after flash, the majority of the polarons are coupled in these pairs charge transfer complexes and experience magnetic interaction exchange and/or dipolar within the pair. For more detailed determination of the type and magnitude of magnetic interactions between the components of polaron pair the electron spin echo experiments are needed. Especially desired is the observation of the out-of-phase ESE, which provides direct information about the magnitude of magnetic interactions and the interspin distance in the spin-correlated radical pair.

However, up to now no ESE signals were reported for the polaron pairs of polymer/fullerene composites generated by laser flash in their non-equilibrium spin state. This caused mainly by small volume of the blend effectively illuminated in the ESE experiment, and by a relatively low sensitivity of the ESE technique, as compared to CW EPR. In the present paper, the polarized ESE signal of spin-correlated polaron pair PCBM^-/P3HT⁺, appearing upon laser excitation of the P3HT:PCBM blend is observed for the first time. The spin relaxation times for this pair are determined.

2 Experimental

2.1 Sample Preparation

The solution of 400 µl toluene and 0.5 mg PCBM and 0.5 mg P3HT was prepared using ultrasonic bath. The solution was put in the quartz EPR tube of 4.5-mm outer diameter, and three freeze-pump-thaw cycles were performed. Toluene was evaporated in vacuum of about 0.1 torr, which resulted in the formation of the P3HT:PCBM blend on the inner wall of the EPR sample tube. The sample was annealed at 450 K during 20 min and the tube was sealed. The thickness of P3HT:PCBM blend was slightly inhomogeneous over the sample, with estimated thickness of 2µm. Since the typical absorption coefficient of the P3HT:PCBM blend at 532 nm is about 10⁵ cm^-1, the thickness of the blend ensures complete absorption of the laser light in our setup.

2.2 Experiments with Stationary Light Illumination

CW EPR and ESE measurements were carried out on an X-band ELEXSYS ESP-580E EPR spectrometer equipped with an ER 4118 X-MD-5 dielectric cavity inside an Oxford instruments CF 935 cryostat.

The following parameters of CW EPR experiments were set: microwave power, 63 µW; magnetic field modulation frequency, 100 kHz; magnetic field modulation amplitude, 1 G.

ESE signal was obtained using a two-pulse mw pulse sequence π/2-t-π-echo, where the π/2 and π pulses were of 24 and 48 ns duration, respectively, and the τ delay was 120 ns. The whole echo signal in the time domain was integrated within an integration time window 80 ns centered at the echo maximum.

Samples were illuminated by continuous light irradiation of an Xe lamp, equipped with a filter transmitting light in the range between 350 and 700 nm. The estimated light power reaching the sample was about 10 mW. Temperature was kept at 80 K by cold nitrogen gas flow.

2.3 Experiments with Laser Pulse Illumination

 ESE experiments were carried out on an X-band Bruker ESP-380E FT EPR spectrometer equipped by a homebuilt rectangular resonator with a circular hole of 4-mm diameter in the center of the front wall. A refractive lens was mounted on this wall, with its optical axis passing through the center of the hole. The sample was placed in the quartz Dewar vessel filled with liquid nitrogen, so the sample temperature was 77 K. The vessel was fixed in the resonator, with the sample positioned at the center of the resonator. Illumination from Nd-YAG laser Surelite I-10 was used, with wavelength 532 nm, pulse duration 10 ns, and pulse repetition rate 10 Hz. The laser light was directed to the refractive lens along its optical axis. After passing through the lens and the hole in the resonator wall, the laser pulse illuminated 0.4 x 1 cm² of the sample area with about 5 mJ/cm² density of energy per flash. Such geometry of the experiment allows nearly complete illumination of polymer/fullerene blend in the sample with easily controlled light intensity. The contact of the sample with liquid nitrogen is also important for successful observation of ESE signal of laser-generated polaron pair. Such contact ensures better heat sink from the sample than nitrogen gas flow. This allows the use of a higher laser flash intensity and to produce higher polaron concentration without sample degradation caused by overheating.

For the ESE detection, the sequence flash DAF-π/2-τ-π-echo was used, where the π/2 and π pulses were of 24 and 48 ns length, respectively, and the τ delay was 320 ns. Such a long τ delay was used because of the resonator ringing, which made the detection of a relatively weak ESE signal at shorter τ delays problematic. The whole echo signal in the time domain was integrated within an integration time window 160 ns centered at the echo maximum.

Quadrature detection was used. The phase for the echo was adjusted by the dark ESE signal of accumulated polarons in the P3HT:PCBM blend see below. All the ESE signals presented in this paper refer to the in-phase component of ESE.

3 Results and Discussions

CW EPR spectra of the P3HT:PCBM blend before and during the stationary light illumination are shown as solid and dashed lines, respectively. The spectrum obtained under illumination is similar to that obtained previously for the P3HT:PCBM blend and is composed of the signals of PCBM^- and P3HT⁺ polarons at stationary concentration. A small dark signal was obtained before the illumination. It probably originates from residual P3HT⁺ polarons.

Figure 2b shows the echo-detected EPR spectrum of the P3HT:PCBM blend under the stationary light illumination. As expected, its shape is close to the integrated CW EPR signal, although some broadening can be noticed for ED EPR spectrum. This broadening is probably caused by a relatively large excitation bandwidth of the mw pulses compared to that of the continuous mw irradiation of CW EPR. Integration over the echo signal is known to reduce this effect. However, at our experimental conditions, the excitation bandwidth broadening of ED EPR spectra is not canceled completely.

It is well known that some fraction of photogenerated polarons in the P3HT:PCBM blend has a relatively long lifetime at cryogenic temperatures in the order of min. In our experiments, they are accumulated under the action of the laser flashes and their electron spins relax to Boltzmann equilibrium with the longitudinal electron spin relaxation time T₁ in the order of tens to hundreds microseconds. The ED EPR spectrum of the accumulated polarons in the P3HT:PCBM blend is shown as dotted line. This spectrum was recorded with synchronization between the laser pulses and mw pulses switched off. Since the repetition time of the laser flashes in our experiment is much longer than T₁ of the polarons in the P3HT:PCBM blend, one can assume that they are at thermal equilibrium at the conditions described. As expected, this spectrum is similar to the ED EPR spectrum of the P3HT:PCBM under stationary illumination. Some dark background contributes to the wings of the spectrum of the accumulated polarons. It is probably caused by paramagnetic centers in the quartz Dewar and quartz sample tube.