Super-resolution spectroscopic imaging of plastic solar cells

There has been a need for alternative, renewable, and sustainable energy sources for generations. Notable success in the conversion of solar energy to electricity has been achieved with silicon-wafer based photovoltaic cells. Despite this, researchers continue to search for more flexible, easy to manufacture, easy to mount, light weight, and low-cost photovoltaic alternatives.¹Plastic solar cells based on blends of conjugated polymers and fullerenes are the most promising candidates to meet these requirements. Recently,²an organic bulk-heterojunction (BHJ) photovoltaic material was reported that had almost 100% internal quantum yield, leading to almost 8% power conversion efficiency. In order to enter the market for integrated and roof-top photovoltaic panels, efficiencies above 10% are needed. Thus, progress is still required to reach this benchmark before commercialization can be realized.

Plastic solar cells based on the BHJ configuration, for example poly(3-hexylthiophene) and [6, 6]-phenyl-C61 butyric acid methyl ester (P3HT:PCBM) blend film—see Figure 1 (top)—have demonstrated record power conversion efficiencies. These sandwich-structured cells have a photoactive layer consisting of a polymer blend where electron-hole pairs are created by incident photons at junctions between the donor and acceptor polymer. The donor releases electrons upon illumination, which are collected by acceptor fullerenes and transferred to the cathode. The anode consists of a conducting transparent material—indium tin oxide covered with poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate)—deposited on a glass substrate. The efficiency of exciton creation and dissociation is a major factor affecting electron transfer in BHJ plastic solar cells.³ Whilst exciton creation is determined by the absorption efficiency of the polymer, exciton dissociation at the donor/acceptor interface is directly related to the local morphology and distribution of the photoactive compounds. Due to short exciton-diffusion lengths, typically less than 10nm, the most important photophysical processes take place in a narrow region a few nanometers from the electron donor/acceptor interface.⁴ This narrow band is at the edge of the spatial resolution of even the most advanced optical and electrical measurement techniques.

We developed a novel near-field optical microscopy technique for super-resolution spectroscopic mapping, allowing us to ‘see’ local morphology-related photophysical processes.⁵ We achieved nanometer resolution and an optical signal enhancement of up to one million times. The technique involves local light concentration by the gold tip of the microscope into the nanometer-sized gap between the tip apex and sample surface. This generates an optical near-field that in turn excites the sample: see Figure 1 (bottom). Photons generated by the sample within the near-field are collected by the tip and parabolic mirror, and are then directed onto the detector. High enhancement factors are possible because a perfect optical antenna is formed as the tip is in the focus of the parabolic mirror.

Figure 1. (Top) The sandwich structure of a bulk-heterojunction plastic solar cell. The active composite transfers electrons to fullerenes, which in turn transfers electrons to the cathode. Al: cathode. ITO: indium tin oxide. PEDOT:PSS: poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate). (Bottom) A schematic of the super-resolution microscopy technique. The gold tip concentrates light between its apex and the sample, generating a near-field that excites the sample.

We applied this technique to investigate the affect of thermal annealing on the morphology of P3HT:PCBM blend film.^6,7 Based on simultaneously-recorded morphology and spectroscopic information, we analyzed the interplay between film morphology, local molecular distribution, and P3HT photoluminescence (PL) quenching efficiency. We probed the PL and Raman signals of the electron donor (P3HT) and acceptor (PCBM) at an optical resolution of approximately 10nm, and identified the chemical nature of the different domains. We quantitatively revealed the local PL quenching efficiency, which is related to electron transfer from P3HT to PCBM: see Figure 2 (a-c). Strong PCBM PL emission dominated the optical signals from an anomalous ‘island’ bearing large PCBM aggregates: see Figure 2 (b). Generally, P3HT molecules were distributed homogeneously throughout the film, though P3HT-rich regions were observed, such as in the upper region of Figure 2 (c). The highest P3HT PL quenching efficiency was correlated with PCBM-rich regions: see Figure 2 (d). Notably, lower PL quenching efficiency of P3HT was observed in the lower portion of Figure 2 (d), possibly due to fewer P3HT:PCBM interfaces as a result of enhanced phase segregation. Thus, the observed elevated emission in this region suggests P3HT PL quenching via exciton dissociation and electron transfer to PCBM is hindered.

Figure 2. Photoluminescence (PL) intensity distribution of (a) poly(3-hexylthiophene) (P3HT) and (b) [6,6]-phenyl-C61 butyric acid methyl ester. (c) Molecular distribution of P3HT derived from the intensity variations of its C=C vibrational Raman signal. (d) Normalization of (a) and (c) gives the P3HT PL quenching efficiency. (a-c) Sample area= 1.0×0.75μm.

In summary, with super-resolution spectroscopic mapping we were able to map the local chemical distributions in plastic solar cell blend films. These local morphology-related variations of optical signals (PL and Raman intensities) provide necessary guidance for improving cells by revealing electron transfer efficiency. Systematic super-resolution spectroscopic imaging will hence play a key roll in identifying the parameters for optimizing solar cell blend films. We next aim to implement ultra-short laser pulses to measure PL and quenching dynamics as a function of local chemical composition.