Why stable perovskites are a bigger deal than you think

This post is a continuation of my earlier post on overcoming the SC limit using perovskite-silicon tandem solar cells. To further explore this near-future technology for large-scale solar energy, I will briefly explain what perovskites are, how their discovery shook up the photovoltaic community, and where their development currently stands.

Perovskite

To begin with, perovskite is not a single material, but a whole class of materials that have a specific crystal structure, namely ABX3, where "A" and "B" stand for positively charged ions (cations) and "X" for anions. The name comes from the Russian mineralogist Lev Perovski, once calcium titanium oxide with the chemical formula CaTiO3 was discovered in the Ural Mountains in 1839. In optoelectronics, the organometallic halide perovskite composition is the most promising, where “A” is an organic cation such as methylammonium or formamidinium, “B” is typically lead and X is a halide anion such as iodide or bromide.

Figure 1. Perovskite crystal structure.

Interestingly, synthetic semiconducting perovskites first entered the field of photovoltaics to increase the absorption of dye-sensitized solar cells (DSCs, an old-fashioned, highly inefficient organic solar cell technology). In 2009, a group from Japan reported organic–inorganic lead halide perovskite compounds that efficiently sensitize TiO2 for visible-light conversion and achieve a “remarkable” power conversion efficiency (PCE) of 3.8%. [1] (Today, this work is cited over 21,000 times)

A short time later, it was shown that it is in fact not the perovskites that improve DSC efficiency, but that DSC severely limits the efficiency of perovskites. In 2012, H. Snaith demonstrated that they are not only able to sensitize a semiconductor, but - even more interestingly - can also transport electronic charges to the electrodes of the solar cells themselves, enabling much higher device efficiencies. This was published in Science [2], and let the genie out of the bottle, sparking unprecedented research interest in perovskite solar cells (PSCs). In a single decade (2009-2019), 16103 research papers were written. [3]

While it took 15–42 years for traditional PV technologies to mature, PSC technology has done the same within 10 years. Today, PSCs boasts a record PCE of 26.1%, outperforming other thin-film technologies such as CIGS, CdTe or amorphous Si that have developed since the 1970s. [4]

So, what makes perovskites so special?

Perovskites possess multiple favorable optoelectronic properties, such as a long carrier lifetime, a long carrier diffusion length, a low exciton binding energy, out of which I would like to highlight most:

High absorption coefficient. Typically 10^4 to 10^5 cm^-1, whereby the layer thickness of a few hundred nm is sufficient for photovoltaic applications. Notably, this is a thousand times thinner than the typical crystalline Si in conventional solar cells.
Defect tolerance. In typical semiconductors, the valence band maximum (VBM) is formed from bonding orbitals and the conduction band minimum (CBM) from antibonding orbitals. When these bonds are broken, dangling bonds form near the original bonds, creating defects deep in the semiconductor bandgap. Perovskites instead form their bandgap from antibonding orbitals at both the VBM and CBM, so breaking these bonds creates states outside the bandgap, either shallow defects or states within the valence band. The result is that polycrystalline perovskite films with a defect density 10^6 times greater than that of single crystal silicon can produce similar solar cell performance.

Figure 2. Taken from [5].

Bandgap tunability. Compared to a fixed bandgap of Si, basically, any desired bandgap of perovskite can be achieved with a simple composition technique, as shown in the graph below. This is mainly done by using different ratios of halides (Br, I) as anions in the perovskite nanostructure. This is particularly useful in tandem solar cells, where the perovskite bandgap can be fine-tuned to perfectly complement other absorbing layers.

Figure 3. Taken from [6].

Low manufacturing cost. Above all, perovskites are easy and extremely cheap to produce. They are literally produced by mixing different salts in solutions, which are then simply dried. This is an immense advantage over classical semiconductors such as Si, which require temperatures of ~1500C to turn sand into crystalline material, so to speak.

So, why we still can’t buy PSCs?

Yet everything has its price. Hybrid halide perovskites, which are best suited for photovoltaic applications, are thermodynamically unstable structures [7]. As such, perovskites tend to decompose quickly under the influence of temperature, moisture, oxygen or light. Moreover, they are ionic compounds and ions can migrate within the crystal lattice, further promoting device degradation. The science behind ionic migration and the degradation of PSCs is extremely complex and has yet to be fully understood, but we will not go into detail here. For now, it is important to note that the initial PSCs would degrade after just a few hours of operation.

Many ways have been developed to increase their stability. These include encapsulation, composition optimization, interfacial engineering or the introduction of specific additives or ligands during perovskite synthesis.

For instance, in one of our work [8], we showed that specific ionic liquid additive inhibits methylammonium (MA) permeation in methylammonium lead triiodide (MAPbI3) occurring due to intrinsic thermal degradation, by inhibiting out‐diffusion of the MA+ cation, preserving the pristine material and preventing decomposition.

Other researchers went further and tried literally everything they had on the shelf, including caffeine [9] or sugar [10], as additives - with rather moderate success, steadily increasing their operational stability up to hundreds of hours.

However, in February this year, an article [11] was published in Nature Energy (a journal with a 5-year impact factor of 64, whereas Nature only has 61) and describes a major breakthrough. PSCs treated with a sulfonium molecule showed a performance loss of less than 1% after more than 4,500 hours. This results in a theoretical T80 (time required for a device to degrade to 80% of its initial performance under certain conditions) of more than nine years in continuous sunlight, which is comparable to the T80 of 25 years for Si solar cells. This indicates that efficient and stable PSCs are indeed within reach.

So, when?

Notably, Oxford PV [12], founded in 2010 as a spin-off from the University of Oxford, focuses exclusively on the development and commercialization of perovskite-based solar technology and is by far the leader in this field. The combination of its research and development site in the UK with a pilot and production line in Germany enables the accelerated transfer of its technology into the industrial-scale production of perovskite-on-silicon tandem solar cells. The company has achieved a world record efficiency of 28.6% with its commercial size perovskite-on-silicon tandem solar cell and has a clear roadmap to take this technology to over 30% efficiency. They also plan to deliver market-ready modules as soon as..wait for it... this year!

References:
[1] https://doi.org/10.1021/ja809598r
[2] DOI: 10.1126/science.1228604
[3] https://doi.org/10.1002/pssa.202000441
[4] https://www.nrel.gov/pv/cell-efficiency.html
[5] https://doi.org/10.1038/s41578-023-00582-w
[6] https://doi.org/10.1016/j.nanoms.2019.10.004
[7] https://doi.org/10.1038/srep37654
[8] https://www.researchhub.com/paper/6034452/retarding-thermal-degradation-in-hybrid-perovskites-by-ionic-liquid-additives
[9] https://doi.org/10.1016/j.joule.2019.04.005
[10] https://doi.org/10.1039/D0TC00365D
[11] https://www.researchhub.com/paper/6171613/breakthrough-perovskite-solar-cells-with-less-than-1-efficiency-loss-over-4500-h-operational-stability-tests
[12] https://www.oxfordpv.com/

Preprint

Preprint

Why stable perovskites are a bigger deal than you think