Multimodal Imaging of Perovskites

In a thin film of a solar-energy material, molecules in twin domains align in opposing orientations within grain boundaries. Strain can change chemical segregation and may be engineered to tune photovoltaic efficiency. (Source: S. Jesse, ORNL / DOE)

A unique combi­nation of imaging tools and atomic-level simu­lations has allowed a team led by the Department of Energy’s Oak Ridge National Labora­tory to solve a long­standing debate about the properties of a promising material that can harvest energy from light. The researchers used multi­modal imaging to see nanoscale inter­actions within a thin film of hybrid organic–inorganic perovskite, a material useful for solar cells. They determined that the material is ferro­elastic, meaning it can form domains of polarized strain to minimize elastic energy. This finding was contrary to previous assump­tions that the material is ferro­electric, meaning it can form domains of polarized electric charge to minimize electric energy.

“We found that people were misguided by the mechanical signal in standard electro­mechanical measure­ments, resulting in the misinter­pretation of ferro­electricity,” said Yongtao Liu of ORNL. Olga Ovchin­nikova, who directed the experiments at ORNL’s Center for Nano­phase Materials Sciences (CNMS), added, “We used multimodal chemical imaging – scanning probe micro­scopy combined with mass spectro­metry and optical spectro­scopy – to show that this material is ferro­elastic and how the ferro­elasticity drives chemical segre­gation.”

The findings revealed that differen­tial strains cause ionized molecules to migrate and segre­gate within regions of the film, resulting in local chemistry that may affect the transport of electric charge. The under­standing that this unique suite of imaging tools enables allows researchers to better correlate structure and function and fine-tune energy-harvesting films for improved perfor­mance. “We want to predic­tively make grains of particular sizes and geome­tries,” Liu said. “The geometry is going to control the strain, and the strain is going to control the local chemistry.”

For their experiment, the researchers made a thin film by spin-casting a perov­skite on an indium tin oxide-coated glass substrate. This process created the conductive, trans­parent surface a photo­voltaic device would need, but also generated strain. To relieve the strain, tiny ferro­elastic domains formed. One type of domain was grains, which look like what you might see flying over farmland with patches of different crops skewed in relation to one another. Within grains, sub-domains formed, similar to rows of two plant types alter­nating in a patch of farmland. These adjacent but opposing rows are twin domains of segre­gated chemicals.

The technique that scientists previously used to claim the material was ferro­electric was piezo­response force micro­scopy, in which the tip of an atomic force micro­scope (AFM) measures a mechanical displace­ment due to its coupling with electric polarization – electro­mechanical displace­ment. “But you’re not actually measuring the true displace­ment of the material,” Ovchin­nikova warned. “You’re measuring the deflection of this whole ‘diving board’ of the canti­lever.” Therefore, the researchers used a new measure­ment technique to separate cantilever dynamics from displacement of the material due to piezo­response– the Inter­ferometric Displace­ment Sensor (IDS) option for the Cypher AFM, developed by Roger Proksch, CEO of Oxford Instru­ments Asylum Research. They found the response in this material is from cantilever dynamics alone and is not a true piezo­response, proving the material is not ferro­electric.

“Our work shows the effect believed due to ferro­electric polari­zation can be explained by chemical segre­gation,” Liu said. The study’s diverse micro­scopy and spectro­scopy measurements provided experimental data to validate atomic-level simu­lations. The simu­lations bring pre­dictive insights that could be used to design future materials.

“We’re able to do this because of the unique environ­ment at CNMS where we have characteri­zation, theory and synthesis all under one roof,” Ovchin­nikova said. “We didn’t just utilize mass spectro­metry because [it] gives you information about local chemistry. We also used optical spectro­scopy and simu­lations to look at the orien­tation of the molecules, which is important for under­standing these materials. Such a cohesive chemical imaging capa­bility at ORNL leverages our func­tional imaging.” (Source: ORNL)

Reference: Y. Liu et al.: Chemical nature of ferroelastic twin domains in CH3NH3PbI3 perovskite, Nat. Mat., online 27 August 2018; DOI: 10.1038/s41563-018-0152-z

Link: Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, USA

Speak Your Mind