2D perovskites, really?

For a couple of years, more and more papers have been reporting the use of 2D halide perovskites. I’ve found it quite confusing. These paper usually describe slabs of corner-sharing BX₆ octahedra, separated by large organic cations. In most cases, these materials are processed as thin films or large (about 1cm×0.5cm×0.5cm) single crystals. In fact, what is said to be 2D (or quasi-2D) perovskites is always a Ruddlesden-Popper, a Dion-Jacobson or an Aurivillius phase, that are all bulk materials.

In the 2D materials community, and more generally in solid state physics, we describe two-dimensional atomic crystals as sheets thinner than the Fermi wavelength (or in some cases the De Broglie wavelength). In these materials, the electrons are allowed to move in two directions, but are confined in the c direction. Several names describe different structures, all 2D: monolayers, bilayers, van der Waals heterostructure, atomically thin layers. The most representative material of this class is graphene, that consists in a single layer of carbon atoms, that can be seen as a plane pulled out of bulk graphite.

In the perovskite community, the study of these so-called two dimensional (or quasi-two-dimensional) halide perovskites hasn’t lead, to my knowledge, to the report of any 2D electron gas. It would be very surprising to observe some of these, given the high density of defects in these materials that make large mono layers hard to achieve.

I propose to use the terminology layered perovskite, to differentiate this from the 2D materials, and hopefully allow those two different fields to move beyond current limitations and come together. No one would imagine talking about 2D carbon to describe graphite. Nonetheless, this is exactly what the perovskite community has started to do.

As well, it’s been reported “1D perovskites” and “0D” perovskites. I suggest using “perovskite nanopillers” or “perovskite dendritic structures” to describe the 1D perovskite. However, my knowledge of these 1D and 0D fields is too poor to give an accurate proposition.

Addendum (24 January 2019): Aron Walsh tweeted yesterday this commentary from Joachim Breternitz and Susan Schorr, highlighting their view about what could or couldn’t be defined as a perovskite. This reads:

In particular for the Ruddlesden–Popper phases, we would like to draw the attention of the reader to the original publications of Ruddlesden and Popper where the authors point out that these compounds contain perovskite layers. Perovskite layers are, however, not layered perovskites and no matter how small the change in semantics may appear, the latter would not be correct.

Joachim Breternitz and Susan Schorr

Indeed, what has been named layered perovskites in the oxide perovskite community are compounds of the form La2-xBaxCuO₄, Sr2VO₄, LaSrMO₄ where alternating A-site cation are considered as different layers. This is also quite confusing given that layered materials are usually considered as materials forming strong bonds in two directions and a weaker bond in the third direction, as in lead iodide (PbI₂), molybdenum disulphide (MoS₂) or tungsten diselenide (WSe₂). However, the discussion remains open!

Leave a comment

Leave a Reply