Posted on 25 Jan 13:50
Over the last few months, we've been improving our purification process for MAI and have come across some interesting results that may be connected to several of the urban legends around perovskite fabrication.
As we've been improving our purification and drying process for MAI, we reached a point where when we mixed it with lead chloride, a cloudy solution was formed. Our first thought as experimentalists was to look for a source of contamination - possibly of the solvents or a stable, solvated MAI crystal with sufficient thermodynamic stability to prevent complete drying in the vacuum oven. After repeated attempts, we still had the same result, despite our analysis indicating that we had a highly pure material. This was somewhat of a mystery and also delayed the shipment of some perovskite inks (I101 and I201), for which we would like to apologise to any affected customers.
To resolve this, we started a series of control experiments and tested the unpurified MAI feedstock. It became apparent that with the highest purity MAI, the lead salts (PbCl2 and PbI2) had issues dissolving. Meanwhile, it has been found that a small amount of hydroiodic acid significantly increases the solubility of the commonly used lead salts [1,2]. This may well be related to observations made by a number of our academic collaborators that not all batches of lead iodide would dissolve successfully, and indeed that 99% pure powder often proves more soluble than 99.999% pure powder. Whilst the exact reason behind this is unknown, it seems sensible to suggest that an impurity in the PbI2 fabrication process is important for its solubility in DMF and DMSO. This issue may well be similar to that which we are currently noting with our most pure MAI batches as it appears that lead PbCl2 is also easier to dissolve in the presence of slightly impure MAI. In addition, it may explain why different groups using different MAI fabrication procedures can experience different solubility levels for their perovskite precursor solutions, an effect which may impact upon the characteristics of thin-film formation.
We are currently working on testing this theory and are looking at introducing two grades of MAI purity to account for this; one allowing for easy ink preparation, and another higher purity MAI that may require the use of additives such as HI for high concentration ink preparation. The latter is likely to be important for optimising lifetime and moving towards the commercialisation of perovskite solar cells.
 Planar CH3NH3PbI3 Perovskite Solar Cells with Constant 17.2% Average Power Conversion Efficiency Irrespective of the Scan Rate, J. Heo et al. Advanced Materials, 27, 3424-3430 (2015)
 Hysteresis-less inverted CH3NH3PbI3 planar perovskite hybrid solar cells with 18.1% power conversion efficiency, J. Heo et al. Energy & Environmental Science, 8, 1602-1608 (2015)