Charge transport in mesoscopic hybrid solar cells

Organic and hybrid electronics is a rapidly growing field. It attracts both academic and industrial interest due to its potential for low-cost, high-throughput processing of optoelectronic devices. For both commercial and environmental reasons, developing inexpensive routes to efficient solar cell technologies is critical. Currently, the two main types of solution-processable solar cells are dye-sensitized solar cells (DSCs)¹ and molecular or polymer blend bulk heterojunction (BHJ) cells.^2,3 The highest reported solar power conversion efficiencies for BHJ cells range between 4 and 5%,^4,5 compared to 10–11% for liquid electrolyte DCSs.⁶

DSCs clearly appear to be winning the efficiency race. But the redox active liquid electrolyte used in these cells is industrially less attractive owing to problems of leakage and corrosiveness. A small number of research groups have recently been working on DSCs that incorporate organic hole transporters instead of liquid electrolyte.⁷ This class of solid-state DSCs has improved considerably over the last few years, and and their conversion efficiencies now exceed 5%.⁸ By the same token, their operational physics remains uncertain, and elucidating it should yield considerable improvements in performance and stability while enabling the fabrication of commercially viable products.

Figure 1. Schematic representation of a dye-sensitized conductivity device.⁹ Au: Gold. Ti: Titanium. e⁻: Electron. h⁺: Electron hole.

Our research is mainly concerned with solid-state DSCs. The active layer consists of light-absorbing molecular sensitizers sandwiched between negative (n-type) mesostructured nanocrystalline titanium dioxide (TiO₂) and a positive (p-type) molecular hole transporter.⁷ The essential processes that govern efficiency are the generation of charge from absorbed photons and the subsequent collection of charge from the device. Unlike inorganic positive-negative (pn) junction solar cells, only majority carriers are involved in the photovoltaic process, namely, electrons in the TiO₂ holes in the hole transporter. This results in significantly longer electron lifetimes. In turn, requirements on the charge carrier mobility of the selected materials are also less stringent, because the longer the charge life, the more time it has to diffuse out of the device. To direct research efforts most effectively, it is advantageous to `dissect’ the device and characterize the optoelectronic properties of each component individually. By fabricating in-plane device structures composed of the same materials as the out-of-plane photovoltaic diode, we can selectively probe the transport in either phase of the composite. This helps to understand the transport function and characteristics in each active material.

We study conductivity devices that consist of mesoporous TiO₂-coated glass slides, dyed and filled with a molecular hole transporter. Capping the composite films with gold electrodes that define a channel allows us to selectively probe hole transport through the organic phase. Alternatively, we use prepatterned titanium bottom electrodes to selectively contact the nanocrystalline TiO₂ and test the transport through this phase. Figure 1 shows a schematic of the device structure.

Figure 2. Estimated electron and hole mobility in the active layer of the solid-state DSC as a function of incident light intensity. Vs: Volts per second.

In this type of configuration, we can vary and investigate many different material and operational parameters. For example, we recently showed that introducing ionic additives in the hole-transporter phase significantly enhances conductivity, while redox-active chemical oxidants produce negligible enhancement. These findings enable subsequent optimization of hole-transporter composition, leading to direct improvements in solar cell efficiency.¹⁰ In another recent study,⁹ we investigated charge transport through a molecular hole transport material (HTM) such as the one shown in Figure 1. We showed that hole mobility increases by many orders of magnitude when the illumination intensity is increased from dark to 100mWcm⁻². HTMs are essentially phototransistors. Besides investigating their physics, we are also optimizing them for light-sensing responsivity and potential exploitation as photodetectors.

Interestingly, and quite unexpectedly, the conductivity and charge carrier mobility through the molecular HTM typically used in a DSC, namely, 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene (spiro-MeOTAD), are both significantly greater than through mesoporous TiO₂ (see Figure 2).¹¹ This implies that new HTMs with enhanced mobility are unlikely to improve cell performance. Increased electron mobility in mesoporous TiO₂, however, may enhance performance if transport is a limiting factor. To understand whether electron diffusion is sufficiently fast to enable the majority of charges to be collected before recombining in the device with opposite charges, we performed transient current collection and voltage decay experiments on complete solar cells. From these measurements, we extracted charge lifetimes, effective diffusion coefficients, and diffusion lengths under working conditions. Both methods yielded qualitative agreement for estimating the charge transport in the composite. Critically, we observed that the charge recombination and collection lifetimes became comparable as the potential in the cell approached that of an open circuit under standard operating conditions. This implies that many charges recombine within the composite when operating at high potential. Hence, significant improvements in fill factor and open-circuit voltage should be possible either by enhancing mobility in mesoporous TiO₂ or by slowing electron-hole recombination at the dye interface.

In subsequent work, we studied charge collection and recombination kinetics under various conditions. We observed that, under short-circuit conditions, the charge recombination lifetime approached 1s with an electron diffusion length as long as 20μm.¹² These very encouraging results suggest that overcoming issues such as infiltrating the mesoporous TiO₂ with hole-transporter and composite formation should yield a hybrid cell with efficiency as high as that of the liquid electrolyte DSCs (∼11%), thus making it a real commercial prospect.

To meet future global energy demands and achieve sustainable growth, alternative energy sources must be exploited. Solar energy is an abundant source, and photovoltaics fabricated by conventional semiconductor industrial techniques are relatively efficient. At present, however, they are too expensive. Novel, chemically processed organic and inorganic semiconductor materials and composites, such as our HTMs, offer a low-cost solution for the design of large-area solar cell arrays. We have also investigated charge transport in hybrid composites of solid-state DSCs and found that mobility in the organic phase is significantly faster than in the inorganic material. This is surprising, since molecular semiconductors usually have poorer electronic properties than the inorganic variety. We note, however, that TiO₂, the inorganic n-type material employed in our devices, is conventionally considered to be an insulator. Directing research efforts to improve the electron transport in TiO₂ should therefore be more useful than synthesizing new hole transporters with enhanced mobility.