Pushing Perovskites Past Photovoltaics

Hybrid organic-inorganic perovskite materials have taken solar cell research by storm over the past decade, and now researchers are bringing them to other photonic applications. From a material perspective, photovoltaics and LED applications in particular are very similar explains Yizheng Jin, from Zhejiang University in Hangzhou, China. Other than indirect-bandgap silicon, "researchers recognize that a good photovoltaic material, which possesses low defect density, by definition should be a very good luminescent material," he says.

It's relatively easy to produce halide perovskites with the formula ABX3 from solution by mixing low-cost salt solutions together that "work amazingly," says Jin. The A component is typically an organic cation, such as methylammonium or formamidinium, while the B component is a metal cation such as lead or tin. The X is usually a halide ion, like iodide or chloride. It doesn't take strict conditions to grow very good semiconductor crystals, Jin says. Therefore many groups are now involved in studying the material, including outside photovoltaics, some of whom presented at SPIE Photonics West in February.

"The remarkable performance of perovskite solar cells can be largely attributed to long carrier lifetimes and suppressed non-radiative recombination rates," explains Yitong Dong, from the University of Toronto, Canada. Both properties come because perovskites can tolerate defects better than other semiconductor materials, thanks to their electronic band structure. Such defect tolerance also enables higher photoluminescence efficiency in light-emitting applications, Dong explains. The perovskite structure is also immune to forming trap defects that in traditional semiconductors restrict movement of charge carriers, he adds.

Perovskite materials also have narrow photoluminescence (PL) linewidth, which gives "purer" color emission with similar efficiency, useful in both displays and lasers, Dong adds. Researchers have demonstrated green and red LEDs with narrow linewidth, below 20nm, and high efficiency, he notes. That compares with emission linewidths above 40nm for OLEDs and above 25nm for quantum-dot (QD) LEDs. For lasers, scientists can make single crystals and nanocrystals of perovskites with low defect densities. For example, they have made lead halide perovskite nanowire devices that have very low lasing thresholds around 200nJ/cm2 and high Q factors, around 3600, Dong highlights. Researchers have also achieved continuous-wave lasing at around 100K with organic-inorganic perovskite thin films.

Sam Stranks' team at the University of Cambridge has produced green LEDs using CsPbBr3 perovskite. Credit: Miguel Anaya, University of Cambridge

The unique way defects behave in perovskites also produce sharp colors, adds Sam Stranks, from the University of Cambridge, UK. "We don't see any emission from defects, which would lead to a broader emission spectrum," he says. The high-quality, pure colors that perovskites offer can also span the entire wavelength spectrum, Stranks adds. "You can tune the color by changing the composition, in principle continuously, all the way from the UV to near-infrared," he says.

Spots and funnels

Stranks' team intentionally used compositional variation within perovskite materials to create very bright "hot spots, where there's lots of emission from particular small regions in the sample," he says. Their project team presented their work in Session 3 of Physics, Simulation, and Photonic Engineering of Photovoltaic Devices IX at Photonics West. "We can funnel the charges to particular regions in the sample where they're extremely emissive," Stranks says. His team is also studying nanostructured materials for blue LEDs, which are more confined. "That pushes the emission energy further into the blue," Stranks says. "It is a really nice true blue, about 460nm, that we want for the display industry." However with efficiency at just 1%, this is still a very early-stage technology.

Working in Ted Sargent's Toronto group, Dong has helped to develop LEDs that also "funnel excitons to the lowest bandgap emitter embedded in a solid-state mixture perovskite material." The devices are based on "quantum- size-tuned grains," Dong explains. ‘These concentrate charge carriers, ensuring high luminescence quantum yield," he says. "We have made high-efficiency bright red and green LEDs based on this strategy." The Toronto team has also investigated interactions between electrons and vibrations in chemical bonds holding perovskite materials together. They varied the A cations, showing that different chemical structures could reduce the detrimental effect that vibrations have on LED performance. "We demonstrated perovskite crystals with high photoluminescent quantum yields and narrow emission linewidth, promising for LEDs and lasing materials," Dong says.

Other potential applications that Sargent's group has investigated include using single crystals of perovskite for high-gain, high bandwidth photodetectors and electro-optic modulators. They are potentially promising as photodetectors thanks to their high carrier mobilities and large absorption cross sections, Dong says. The heavy elements that perovskites often contain also mean they could be used in gamma-ray and x-ray scintillators, he adds.

But possibly most impressively, Sargent's team has shown that perovskite quantum dots (QDs) are promising materials for lasing. "They have shown a low gain threshold as well as high Q-factors," Dong explains. "One benefit is low-cost synthesis, as many groups have demonstrated scalable, low-cost, room temperature, perovskite QD synthesis." However perovskite QD stability poses a challenge for applications. Halide perovskites are known for degrading rapidly on exposure to moisture, for example. The Toronto researchers are therefore exploring coatings and other methods to protect them.

More generally, researchers have explored integrating perovskite into silica- alumina materials to boost stability, Dong explains, but the insulating shell harms optoelectronic device performance. Chemical modifications known as passivation likewise seek to protect perovskites. However, they fail to prevent a unique problem originating from their ionic semiconductor structure, namely that the halide X ions migrate through the material. Chloride and iodide-based perovskites have bandgaps in the UV and near-infrared regions, Dong says. Mixing them can therefore provide red and blue LEDs and lasers. "Under heat and electric field, phase segregation happens, and this results in emission shifts," Dong observes. "Compared with perovskites in PV devices, size-confined perovskites are required in light-emitting devices for higher exciton binding energy. This brings more interfaces and potentially more defects, facilitating ion migration."

Using polarizing filters, 3D-printed perovskite nanowire LEDs enable adjustable multicolor displays. Image: N. Zhou et al., Sci. Adv. 5, eaav8141 (2019) Credit: Sameer Khan/Fotobuddy

Seeking protection

Dong also warns that fabricating perovskite LEDs with both high carrier mobility and high photoluminescent quantum yields remains hard, especially blue devices. "The labile surfaces of halide perovskites make inorganic passivation such as core-shell structures a true chemical challenge," he says. Perovskites' electronic structure also makes finding suitable transport layers with good hole injection efficiency difficult, Dong adds.

Nevertheless, the Toronto group worked with Zhanhua Wei's group at Huaqiao University in Quandong, China, on fabricating and passivating perovskite thin films. Together they produced perovskite LEDs with external quantum efficiency exceeding 20 per cent.

Jin notes that encapsulation techniques developed for OLED technology are potentially suitable for perovskites. OLEDs must withstand "strict and harsh conditions," with packages that ensure low oxygen and water concentration, Jin says. As such he feels extrinsic instability is "solvable," but intrinsic instability arising from ion migration is a "more challenging problem" in LEDs and lasers compared to photovoltaics, because the electric field strength is higher. LEDs "put about 2-4V into a 100nm layer," Jin says. "In photovoltaics there is 1V over about 500nm."

Nevertheless Jin believes "that we can conquer this problem in the near future. As long as we can adjust the intrinsic ion migration problem under working conditions, I am very optimistic on this material for light-emitting applications," he says. That's partly because the Zhejiang University team recently developed an approach to make efficient blue perovskite LEDs without needing problematic mixed halides. "Our approach is to use the quantum confinement effects to enlarge the bandgap of the bromide perovskites," Jin says. To achieve that confinement, the researchers used bromide perovskite nanocrystal QDs. Working together with Richard Friend's University of Cambridge team, and other co-workers from China, the approach attained 9.5 per cent external quantum efficiency.

Another key problem often brought up for halide perovskites in photovoltaics is the fact that they almost all use lead as the B atom, with other options like tin performing less well. Yet as lead brings toxicity concerns, many think perovskite devices should avoid lead. Jin thinks that this should be easier in light-emitting devices. Photovoltaic cells have more stringent restrictions on charge transport, to enable generated current to flow and leave the device, and bandgap, to absorb the right color light.

"For LED material, the restriction on the material choice is a little bit looser," Jin says. "We can find some other leadfree materials." Jin goes further still, suggesting perovskite semiconductor devices may succeed in niche light-emission and light-detection applications before they do so in photovoltaics. "I think the challenge of photovoltaic applications is still huge, because crystalline silicon is very good, very stable," Jin says. "There are so many things that semiconductors can do, and perovskite is a very good solution process semiconductor."

There's "quite a bit of activity" looking at lead-free perovskites for consumer electronics, Stranks notes. "For example, double perovskite structures and other nanocrystal, nanostructured versions of perovskites that are lead-free are starting to come through that are quite interesting," he says. "In terms of performance they are still far behind the lead-based systems. Generally, there is this family of lead-free materials that seem to be very promising for emission, it's just whether we can actually control the emission recombination and process them into devices. There are lots of examples out there where seemingly toxic materials are used but in such low quantities and in well-packaged and well-protected forms that it's not an issue. I wouldn't say it's a showstopper, but it's something that of course the field will keep innovating on."

Researchers at Princeton University have refined the manufacturing of perovskite LEDs. Credit: Sameer Khan/Fotobuddy

Exploration and progress

Stranks believes that the charge densities found in LEDs and lasers will make stability a much harder problem to resolve. "This is the real challenge, to move from something that in a lab we can show as a reasonable efficiency, to show that efficiency can last for a sufficiently long time," he says. "It seems encouraging that from the PV side we have made a lot of progress compared to where we were even three years ago. They're now extremely stable." Design is understandably crucial. "You can take a solar cell as is and run it in reverse and you get light out," Stranks says. "But to make it an efficient light emitter, we do have to design it in a different way. We use tailored charge injection layers, rather than the charge extraction layers used in solar cells. The other factor is, of course, the light out-coupling. We need to ideally design it so that you can maximize the light coming out. That really hasn't been explored that much yet for perovskites."

Mansoor Sheik-Bahae and his group from the University of New Mexico (UNM) in Albuquerque, NM, US, is now exploring perovskites for thermal imaging and non-contact temperature measurements. Existing materials have limitations, he stresses. "You cool to just 20˚C below room temperature and thermal cameras typically are not sensitive anymore." In seeking better techniques, his team looks for materials to detect temperature whose photoluminescence spectrum shifts, broadens, or narrows significantly depending on whether they're heated or cooled. Such materials should have good quantum efficiency, so that they don't generate any heat when their atoms are excited. They should also be resilient to thermal cycling.

Recently, Sheik-Bahae's team has been studying QDs made of conventional semiconductors for this application. QDs can easily be mixed with polymers, coated onto arbitrary objects, and detected with inexpensive commercial CCDs or spectrometers, explains UNM team member Alexander Albrecht. These detectors measure when the QDs are excited by UV, and track changes in their emission. "Because you are detecting visible wavelengths, rather than infrared, you can actually get higher spatial resolution," Albrecht explains.

Semiconductor QDs worked well, but degraded quickly, so the team has now moved to perovskite materials. "They're known for having high quantum efficiency," Sheik-Bahae notes. This application should have fewer problems with ion migration, as the QDs have fewer grain boundaries compared to the thin-film counterpart, and are not exposed to long-term electrical current, or even high-intensity light, according to UNM postdoc Davide Priante.

"That helps us to avoid photodegradation," says Albrecht. "If they are enclosed in a polymer, which we like to do anyway so we can apply QDs to different materials, they are also protected from the atmosphere. So we actually think that the degradation is not a big problem." The UNM team  presented preliminary results from this project in the poster session at Photonic Heat Engines: Science and Applications II at Photonics Wesr.

Though halide perovskite research outside photovoltaics is still in its early stages, companies are showing tentative interest. German industrial giant Siemens was recently involved in an x-ray detector review paper, Stranks notes. In the UK, Richard Friend and University of Oxford perovskite pioneer Henry Snaith have founded a perovskite LED startup called Helio Display Materials. "What's interesting is that there aren't yet lots of startup companies like there have been in PV," Stranks says. "We'll see if that changes though."

Stranks notes that technologies usually take at least ten years from first invention in the lab to commercial products. That would mean that the earliest perovskite LEDs might emerge would be 2022. "But I'm excited," he adds. "It's an area that, if we can stabilize them, if we can get the high performance and the long lifetimes, they could quickly become a mainstream technology."

Andy Extance is a freelance science journalist based in the UK. A version of this article originally appeared in the 2020 Photonics West Show Daily.

Related SPIE content:

5 Things to Know about Perovskite Solar Cells

Nam-Gyu Park: The History and Progress of Halide Perovskite Photovoltaics

Multication perovskites for highly stable and efficient solar cells

Perovskite Solar Cell Fever

Photovoltaics and concentrating solar power: why hybridization makes sense