figurePerovskite solar cell fabricated in Yang Yang’s lab at UCLA. [Jingjing Xue]

In 1839, German mineralogist Gustav Rose received a curious delivery from the chief mines inspector of the Russian Empire, August Alexander Kämmerer. It contained a sample of metallic cube-like crystals embedded in a chunk of skarn—a type of rock altered by hot, chemically active fluids—from the Southern Ural Mountains.

A decade earlier, Rose had traveled to the Urals on a rockhounding expedition that resulted in the discovery of several new minerals. Given Rose’s experience with the region’s geology, Kämmerer requested that he determine the physical properties and chemical composition of these strange-looking crystals.

Upon investigation, Rose found that they consisted of a mineral with cubic symmetry composed of calcium and titanium oxides—a brand new discovery. At Kämmerer’s suggestion, Rose named the mineral perovskite, after the Russian politician and mineralogist Count Lev Alekseyevich Perovski.

For the next hundred years, perovskite appeared to have little use except as a pigment. But gradually, the material—and others that share its crystal structure, also known as perovskites—revealed intriguing physical properties that opened the door for a host of applications.

Most notably, perovskite-based solar cells have drummed up excitement as the fastest-improving solar technology to date in terms of efficiency. Since their invention in 2009, the power-conversion efficiency (PCE) of perovskite photovoltaics has skyrocketed from 3.8% to 25.2%. With this latest record-setting value, achieved in late 2019, perovskite cells are approaching the top PCEs achieved by today’s single-crystalline silicon cells.

Commercialization still faces a number of challenges, including stability issues, cost-efficient scalability and the toxicity of lead-containing perovskites.

While other types of solar cells have even higher efficiencies, perovskite photovoltaics possess the practical advantages of a cheap starting material and ease of fabrication. For instance, the world record PCE for any solar cell is currently 47.1%, set by a six-junction III-V semiconductor device. But the cost of III-V materials means that the top echelon of solar cells are only practical for niche applications like aerospace. Meanwhile, the manufacturing costs of perovskite cells are estimated to be about half the amount of crystalline silicon cells.

“Perovskite solar cells are well known for their solution processability, which can potentially allow for simple and low-cost manufacturing,” says Yang Yang, a professor of engineering at the University of California, Los Angeles, USA. “In contrast, conventional photovoltaics generally require high temperatures, high vacuum, or ultra-clean environments to manufacture, which adds costs and complexity.”

Research in the field continues to progress at a furious pace, with large-area technology and perovskite–silicon tandem solar cells both active areas of interest. Commercialization still faces a number of challenges, including stability issues, cost-efficient scalability and the toxicity of lead-containing perovskites. But some researchers believe that, if these hurdles can be overcome, perovskites could be leading candidates for the next generation of commercial solar cells.

figureCrystals of perovskite. Inset: An ABX3 perovskite structure, where A is an organic cation, B is metal cation and X is halogen anion. The ionic nature leads to ion migration during operation of perovskite solar cells and contributes to stability issues. [Rob Lavinsky, iRocks.com, CC-BY-SA-3.0 / Y. Chen et al., RSC Advances, 8, 10489 (2018) - Published by The Royal Society of Chemistry, CC BY 3.]

Coming around to perovskites

While Gustav Rose is credited with the discovery of perovskite, his brother was the first to determine its chemical composition. In 1844, mineralogist and analytical chemist Heinrich Rose deduced that the chemical formula for the new material was CaTiO3. In the century that followed, scientists payed perovskite little attention, and less than 100 studies were published in the first 90 years of its known existence.

The advent of X-ray crystallography elucidated the crystal structure of hundreds of raw materials in the 1920s, including several metal oxides with the formula ABO3, in which A and B are cations. Other studies focused on compounds with the formula ABX3, with X as a halide anion instead of an oxide anion. These were found to follow the same pattern as CaTiO3, and the term “perovskite” from then on referred to both the original mineral and its structure type.

The first perovskite boom came in the 1940s, when the excellent dielectric properties of metal-oxide perovskites were uncovered. In particular, barium titanate had a hundredfold-higher dielectric constant compared to any other material at the time, making it ideal for capacitors in both military and commercial electronics. They continue to be used in many ferroelectric, piezoelectric and dielectric applications.

Generally, metal-oxide perovskites lack the semiconducting properties needed for a photovoltaic material. Organometal–halide perovskites, on the other hand, do fulfill this requirement. In the mid-2000s, Japanese engineer Tsutomu Miyasaka tinkered with the idea of using an organolead–halide perovskite to replace the sunlight-absorbing dye in dye-sensitized solar cells (DSSCs). Such cells, invented in 1988 as a low-cost alternative to conventional silicon solar cells, are made of a porous layer of titanium dioxide covered with a molecular dye.

Miyasaka and his colleagues finally succeeded in building the first perovskite solar cell in 2009, consisting of a thin film of nanocrystalline perovskite deposited by spin coating onto a thick layer of titanium dioxide. The PCE barely reached 3.8%—nevertheless, a milestone in photovoltaics history had been reached.

“The tremendous success of perovskite [photovoltaics] was quite unexpected by our group,” Miyasaka and his team write in a 2019 review. “With a long silence (almost no citations) for about 3 years after our first publication and a series of presentations at international conferences, we had not anticipated [perovskite solar cells] becoming the subject of such a huge amount of research activity.”

Aside from having low efficiency, the first perovskite solar cells failed to inspire excitement from the field because of their poor stability. Traditional DSSCs immersed the cell’s contents in a liquid electrolyte to act as a hole transport layer and close the circuit. However, being an ionic crystal, organolead-halide perovskite dissolves in a polar solution.

figurePerovskite solar cells quickly climbed in efficiency after 2012, with the current record standing at 25.2% power conversion efficiency. [Adapted from NREL Best Research-Cell Efficiencies, 2020 / Courtesy of the National Renewable Energy Laboratory, Golden, CO, USA] [Enlarge graphic]

A major breakthrough occurred in 2012, when Nam-Gyu Park and his colleagues—including Michael Grätzel, a professor at the Swiss Federal Institute of Technology Lausanne (EPFL) and the co-inventor of DSSCs—successfully replaced the liquid electrolyte with a solid hole conductor.

“The stable and high-efficiency solid-state perovskite solar cell was first developed by our team,” says Park, a professor of chemical engineering at South Korea’s Sungkyunkwan University. “We reported for the first time a 9.7%-efficiency perovskite solar cell showing 500 hours of stability without encapsulation, after which [the field of] perovskite photovoltaics surged swiftly.”

In the seven years that followed, the PCE of perovskite solar cells continued to climb. Fabrication advances of a high-quality perovskite film, interfacial engineering of perovskite with both electron and hole transport layers, and other factors contributed to this rapid ascent. The highest efficiency recorded so far for a single-junction perovskite solar cell, 25.2%, was achieved by two research groups—Korea Research Institute of Chemical Technology and Massachusetts Institute of Technology, and Korea University—and independently certified by the U.S. National Renewable Energy Laboratory (NREL).

“Efficiency is expected to be further increased according to the theoretical studies, probably over 30% from a single-junction perovskite solar cell,” Park says. “[With such high efficiencies,] perovskite solar cells are very promising for commercialization.”

figureTandem solar cell made by spraying a thin layer of perovskite onto a commercially available CIGS (copper indium gallium selenide) solar cell, developed at UCLA by Yang Yang’s group. [UCLA Samueli Engineering]

An Achilles’ heel: Instability

A high conversion efficiency is only one piece of the ideal solar cell, however, and perovskite photovoltaics still suffer from a problem that has plagued the technology since its inception.

“The efficiency of perovskite solar cells is now competitive with commercialized photovoltaic technologies based on conventional silicon solar cells, but the major obstacle remaining before commercialization is possible is their well-known instability issues,” says Yang. “The ionic nature of perovskites results in detrimental ion migration during operation and contributes to many of the observed issues.”

The organic–inorganic hybrid perovskites employed in solar cells possess ions with a relatively low activation energy, which leads to severe ion migration in the presence of an electric field. The exact mechanism behind this phenomenon isn’t completely understood, which makes it difficult to fix this type of intrinsic instability. The leading hypotheses relate to ion diffusion through Schottky defects or grain boundaries, but more work is needed to find evidence of these migration channels.

figureA sealed 6.5×6-cm, lightweight, flexible perovskite solar mini-module. The device employs a hydrophobic, cross-linked polymer to protect it from water. [M.K. Nazeeruddin, EPFL]

External stability—or how photovoltaic devices handle being exposed to the elements—is another can of worms. To match the standards set by silicon devices, commercial outdoor solar cells must produce stable power for roughly 25 years under environmental stresses like heat, light, humidity and oxygen. The longest lifetime reported for perovskite solar cells is less than two years, with many cells degrading significantly even after a few hundred hours of usage, which hurts their performance and efficiency.

One of the technology’s biggest environmental foes is moisture. The high polarity of water molecules wreaks havoc on the hydrogen bonds between the organic and inorganic units in hybrid perovskites, causing irreversible degradation of the compound into its smaller components. For example, methylammonium lead halide (MAPbI3) will hydrolyze into PbI2 and CH3NH3I (MAI) in the presence of water, which further breaks down into CH3NH2 and HI.

“Hybrid perovskites quickly decompose in air and humid environments, increasing their sensitivity during both processing and the operational lifetime,” says Mohammad K. Nazeeruddin, professor of chemistry at EPFL. “This greatly limits prospects for commercialization.”

To match the standards set by silicon devices, commercial outdoor solar cells must produce stable power for roughly 25 years under environmental stresses.

Tackling these external stability issues is an active area of research, with scientists testing strategies like incorporating protective interlayers, device encapsulation to shield it from the environment and mixing hydrophobic additives within the active material. Nazeeruddin and his colleagues recently fabricated a perovskite solar cell in humid ambient conditions by adding a polymerizable ionic liquid during processing of the perovskite film. A highly hydrophobic, cross-linked polymer protected the perovskite from interactions with water, and the resulting device achieved a PCE of 19.9%.

“The degradation processes occurring within the cells are still not fully understood, although important steps have been achieved by improved encapsulation techniques and most importantly, by including engineered interfaces with additives,” says Antonio Urbina, professor of electronics at the Technical University of Cartagena, Spain. “These steps contribute to passivation—and therefore reduction—of unwanted surface recombination and, on the other hand, prevent degradation by moisture or oxygen penetration into the cell.”

The trouble with lead

Another commercialization roadblock is the toxicity of lead-based halide perovskite solar cells, which have demonstrated superior efficiency and stability compared with lead-free or tin-based approaches.

“Most of the technological proposals for perovskite active layers include lead halides, and therefore prevention for lead contamination during processing, operation or decommissioning at end of life must be taken into consideration,” says Urbina.

Lead is a cumulative toxicant that affects almost every organ and system in the body. Young children are the most susceptible to the effects of lead, with even low levels resulting in behavior and learning problems, slowed growth, and anemia. Adults exposed to lead can suffer from hypertension, decreased kidney function and reproductive issues.

Unsurprisingly, a solar-cell technology based on soluble lead compounds raises alarms as both an occupational and environmental hazard. But how much lead is really contained in the average perovskite solar cell? How does it compare with other lead-containing technologies in everyday use, such as batteries? And will that amount of lead have a big impact on the environment?

Before dismissing lead-based materials altogether, Urbina and his colleagues decided to find the answers to these lingering questions by performing a life cycle assessment (LCA) of perovskite photovoltaics.

“This methodology aims to quantify all environmental impacts of perovskite solar cell manufacturing, use and recycling,” he says. “We analyze in detail the impact in several categories—such as cumulative energy demand, human toxicity, climate change, etc.—and recommend routes to minimize the global impacts of the technology by moving toward greener manufacturing processes.“

In a 2015 study, the researchers found that scaling up current manufacturing routes for perovskite solar cells would harm both freshwater ecosystems and human health. The majority of these environmental impacts come from the perovskite layer for both vapor-deposition and spin-coating methods. According to Urbina, when methylammonium iodide (CH3NH3I) combines with lead chloride (PbCl₂) during this process, the compound dominates both human toxicity cancer effects and freshwater ecotoxicity, mainly due to the electrical energy input required for synthesis. However, lead itself would contribute surprisingly little to these impact categories—less than 5% for human toxicity cancer effects and less than 0.01% for freshwater ecotoxicity.

Despite these encouraging results, the field is actively exploring lead-free perovskite materials, such as tin-based and germanium-based perovskites, as an eco-friendly alternative. PCEs of these devices have been limited to single-digit percentages, which has led others to look into the partial substitution of lead with tin to create lead–tin alloy perovskites. This low-lead (rather than lead-free) approach has resulted in devices with much better efficiency, with the current record PCE at 21.08%.

Another avenue of research aims to keep any lead safely contained inside the solar cell with different encapsulation methods. For instance, researchers at Northern Illinois University, USA, and the U.S. Department of Energy recently developed transparent lead-absorbing films that did not affect cell performance or long-term stability. These films are made with off-the-shelf materials and could be readily applied to current devices.

figureA team at Helmholtz-Zentrum Berlin, Germany, holds the tandem perovskite–silicon solar cell PCE record, currently at 29.15%. The cell, pictured here, was realized on a typical laboratory scale of one square centimeter. [Eike Köhnen/HZB]

Evolving in tandem

Another emerging perovskite photovoltaic technology with the potential to lower environmental impact is the tandem solar cell. These multi-junction cells allow a top perovskite cell to harvest the blue portions of sunlight while letting red and near-infrared light pass through to be absorbed by the silicon cell below. A 2017 LCA found that manufacturing tandem cells can reduce environmental impact by up to 30% compared with two single-junction devices, mostly due to the exclusion of extra glass.

Much to the excitement of the field, researchers in Germany presented a perovskite–silicon tandem solar cell in January 2020 with a record-breaking 29.15% conversion efficiency.

Much to the excitement of the field, researchers at Helmholtz-Zentrum Berlin, Germany, presented a perovskite–silicon tandem solar cell in January 2020 with a record-breaking 29.15% conversion efficiency. This PCE beats out the top values for individual single-junction crystalline silicon (27.6%, according to the NREL’s Best Research-Cell Efficiency Chart) and perovskite solar cells operating on their own.

“Such tandem architectures have probably the greatest potential to disrupt the commercial photovoltaics market, which is now dominated by conventional inorganic semiconductors,” says Yang.

The previous record of 28% was held by a University of Oxford, U.K., spinoff company, well known as one of the major players pushing for commercialization of perovskite photovoltaics. Established in 2010, Oxford PV Ltd. began with a focus on producing single-junction perovskite solar cells based on the research of Henry Snaith, who collaborated with Miyasaka on the transition to solid-state devices. The company pivoted to tandem perovskite–silicon solar cells in 2014.

figureOxford PV, Germany, is working toward the first volume manufacturing line for perovskite-on-silicon tandem solar cells. [Oxford PV]

“Many companies like Oxford PV have been at the forefront to make commercial perovskite solar cells, since preparation of perovskite layers are rather easy compared to other thin-film technologies,” said Pabitra Nayak, a former member of Snaith’s research group and now based at India’s Tata Institute of Fundamental Research. “The expected date [for solar-cell production by Oxford PV] is by the end of this year,” according to Nayak, although he expects that the pandemic may delay things.

The downside of tandem cells is the extra cost of materials and processes needed to produce layers made of two different materials. In addition, the stark difference in lifetimes of silicon and perovskite—at least for the moment—will prove a challenge in terms of cell stability and long-term viability for widespread commercialization.

Toward commercialization

Despite lingering issues with stability and toxicity, the excitement around perovskite photovoltaics remains high enough that commercialization is well underway. Aside from Oxford PV, other startups in the space include Saule Technologies in Poland and WonderSolar and Microquanta Semiconductor in China. Large corporations like Toshiba, Panasonic and Toyota are also investing in the technology.

To become commercially viable, every solar-cell technology must experience the growing pains associated with going from small-area lab cells (typically less than 0.1 cm2) to standard-size modules. For perovskites, the biggest challenge comes from precisely controlling aspects of manufacturing at large scales to obtain uniform thin films and maximum efficiency. In February 2020, Panasonic reached a milestone by achieving 16.09% PCE for a 802-cm2-sized perovskite module. The company employed a large-area coating method based on inkjet printing to improve the uniformity for thickness and crystal layer.

“The industrial manufacturing techniques are mostly based in roll-to-roll technology in which the different layers of the perovskite module can be printed on rigid or flexible substrates by using functional inks,“ says Urbina. “Although it may seem a simple technique with a huge potential for cheap manufacture, it is a high-tech process, in which hundreds of parameters have to be optimized for a reproducible, ISO-certified manufacturing process.”

Commercial photovoltaic technologies are also required to conform to a set of strict standards developed by the International Electrotechnical Commission (IEC). Devices must pass endurance tests to determine the electrical and thermal characteristics of the modules in various environmental conditions. As a step forward, a team of Australian researchers in June 2020 found that encapsulating perovskite solar cells in pressure-tight polymer–glass stacks allowed them to survive the IEC’s heat and humidity requirements for the first time.

Silicon and inorganic thin-film photovoltaic modules, with their impressive warranty of 25 years, set a high bar for perovskite solar modules. Progress is being made in this area to the point where perovskite cells with more than 10,000 hours of laboratory-stable operation have been reported, which is roughly equivalent to more than one year of outdoor operation.

With over 4,000 studies on perovskite photovoltaics released in 2019 alone, according to Web of Science, the remarkable advancements made in the field over a mere decade are expected to continue. Theoretical calculations predict the efficiency of single-junction perovskite cells in the lab to surpass 30%, but at this point, much of the research is focused on extending stability and improving the performance of large-scale modules.

A long road still lies ahead before widespread commercialization, but perovskite solar cells have built-in advantages like abundant, cheap raw materials and simple processing that could offset their drawbacks. The next 10 years will undoubtedly be a crucial make-or-break period for perovskite photovoltaics and determine whether they become the low-cost, high-efficiency solar technology the world has been waiting for.


Meeri Kim is a freelance science journalist based in Los Angeles, CA, USA.

References and Resources

  • H-S. Kim et al. Sci. Rep. 2, 591 (2012).

  • Y. Rong. Science, 361, eaat8235 (2018).

  • A.K. Jena et al. Chem. Rev. 119, 3036 (2019).

  • R. Wang et al. Adv. Funct. Mater. 29, 1808843 (2019).

  • R. Xia et al. Adv. Mater. 2003801 (2020).

  • L. Shi. Science 368, eaba2412 (2020).

  • N.S. Arul and V.D. Nithya (Eds). Revolution of Perovskite: Synthesis, Properties and Applications, Springer (2020).

  • A. Urbina. J. Phys. Energy 2, 022001 (2020).