Methylammonium lead iodide perovskite can make high-efficiency solar cells, which also show an unexplained photocurrent hysteresis dependent on the device-poling history. Here we report quasielastic neutron scattering measurements showing that dipolar CH3NH3+ ions reorientate between the faces, corners or edges of the pseudo-cubic lattice cages in CH3NH3PbI3 crystals with a room temperature residence time of ∼14 ps. Free rotation, π-flips and ionic diffusion are ruled out within a 1–200-ps time window. Monte Carlo simulations of interacting CH3NH3+ dipoles realigning within a 3D lattice suggest that the scattering measurements may be explained by the stabilization of CH3NH3+ in either antiferroelectric or ferroelectric domains. Collective realignment of CH3NH3+ to screen a device's built-in potential could reduce photovoltaic performance. However, we estimate the timescale for a domain wall to traverse a typical device to be ∼0.1–1 ms, faster than most observed hysteresis.

We have examined methylammonium lead iodide (MAPI) cells of the design FTO/sTiO 2 / mpTiO 2 /MAPI/Spiro-OMeTAD/Au using fast opto-electronic techniques including transient photovoltage, differential capacitance, charge extraction, current interrupt, and chronophotoamperometry. The data allow several important conclusions regarding the physics and proper characterization of these cells. 1) In mpTiO 2 /MAPI cells there are two kinds of extractable charge stored under operation: a capacitive electronic charge (~0.2 µC/cm 2 ), and another, much larger, charge (40 µC/cm 2 ) which could be the result of dipole realignment, or the effect of mobile ions. The capacitive charge is ~10 times smaller than in mpTiO 2 /dye/SpiroOMeTAD cells with similar mpTiO 2 thickness. 2) Transient photovoltage decays are strongly double exponential with two time constants that differ by a factor of ~5, independent of bias light intensity. We show that the fast decay (~1 µs at one sun) can be assigned to the predominant charge recombination pathway in the cell. We examine and reject the possibilities that the fast decay is due to ferro-electric relaxation, or to the bulk photovoltaic effect. We provide two possible schematic electrical models for the cells that reproduce the V oc and TPV data.3) It has been previously observed that MAPI cells frequently show current/voltage hysteresis. For example, an increase in J sc and V oc is observed on the return sweep of a cyclic JV. Our capacitance vs V oc data indicate that the hysteresis involves a change in internal potential gradients, most likely a shift in band offsets at the TiO 2 /MAPI interface. The TPV results show that the V oc hysteresis is not due to a change in recombination rate constant. Calculation of recombination flux at V oc shows that the hysteresis is also not due to an increase in charge separation efficiency, and that charge generation is not a function of applied bias. We also show that the JV hysteresis is not a light driven effect, but is caused by exposure to forward bias, light or dark.

No abstract

Stable and efficient dye‐sensitized solar cells based on water‐containing electrolytes are shown. For water contents up to 40%, no decrease in efficiency is seen. The cells are demonstrated to be stable for long periods of continuous illumination.This work lays a foundation for the further development of water‐based cells for commercial production.

Three organic or hybrid photovoltaic technologies are compared with respect to performance and stability under the harsh regime of concentrated light. Although all three technologies show surprisingly high (and linear) photocurrents, and better than expected stability, no golden apples are awarded.

Electrical and luminescence characterization was performed on 16 dye sensitized solar cells with different formulations, from different industrial and academic sources. Most of the cells were fabricated in pre-industrial pilot lines. The cells were put through a light soaking period up to 150 hours and then re-characterized. The results show the commonly observed increase in J sc with light soaking is due to a decrease in the conduction band energy (with respect to the electrolyte) and an increase in the injection rate and efficiency. The strong correlation between the luminescence decay lifetime (<200 ps to 5 ns) and the photocurrent (7 to 13 mA cm À2 ) shows that the luminescence decay is a useful monitor of injection rates in these cells. The very slow injection shown by some cells implies substantial losses at the injection step. The data point to a need to understand and improve the TiO 2 processing and dyeing conditions in the industrial setting as well as the need to focus injection studies on the full range of dynamics present in the cells.

We have fabricated Dye Sensitised Solar Cells (DSSCs) using only water as the solvent and guanidinium iodide-iodine as the redox couple that operate at 4% energy efficiency under 1-sun illumination. This result is $5 times higher than the best previously reported values. We show that it is critical to facilitate the wetting of water electrolytes into the mesoporous TiO 2 dye films, especially when using hydrophobic dyes. We show chenodeoxycholic acid to be a good surfactant for this purpose. By separate variation of iodide and iodine concentrations in series of cells we show that the optimum concentrations for water based DSSCs are quite different from those used in organic electrolytes. We argue that this is due to the much lower stability constant of tri-iodide in water, relative to organic solvent. Finally we also vary the TiO 2 thickness and pore structure to achieve the above stated efficiency.

Electron transfer from TiO to iodine/iodide electrolytes proceeds via reduction of either I or uncomplexed I (free iodine), but which route predominates has not previously been determined. By measurement of the electron lifetime while independently varying free iodine or I concentrations, we find the lifetime is correlated with free-iodine concentration and independent of I concentration. This trend supports the hypothesis that electron recombination to the electrolyte occurs predominantly by iodine reduction rather than reduction of triiodide.Peer reviewe