Multi-modal characterization of nitrate reduction nano-catalysts with periodic strain distribution

Yiyuan Tao, Xingyu Zheng, Shi Huang, Ershuai Liu, Yangjian Lin, Qike Jiang, Long Yang, Walter S. Drisdell, T. D. Xu, Yao YangNature Communications 2026 · DOI

TL;DR

Researchers have developed an optimized 4D-STEM imaging technique to map lattice strain in industrial-sized (500 nm) nanocatalysts with unprecedented precision. By discovering and tuning "ripple-like" strain patterns through copper doping, the team achieved record-breaking nitrate-to-ammonia conversion efficiency and over 1,000 hours of stable operation in a commercial-style reactor.

What Problem Does This Paper Address?

Electrocatalytic nitrate reduction (NO₃⁻RR) offers a "two birds, one stone" solution: it removes toxic nitrate from wastewater while producing ammonia—a critical fertilizer and carbon-free fuel—without the massive carbon footprint of the traditional Haber-Bosch process.

A key strategy to boost these catalysts is strain engineering—intentionally distorting the crystal lattice to change how molecules stick to the surface. However, there’s a major "scale gap." Most high-resolution strain imaging techniques only work on tiny particles (under 100 nm). Industrial-grade catalysts are often much larger (up to 500 nm), making them "blind spots" for traditional microscopy. Without being able to see the strain in these larger particles, researchers couldn't accurately correlate physical structure with chemical performance.

Methods

The authors bridged this gap using an optimized four-dimensional scanning transmission electron microscopy (4D-STEM) approach.

  • The Innovation: They combined nanobeam electron diffraction (NBED) with an exit-wave power-cepstrum (EWPC) method. This allowed them to map strain across 500 nm nanocubes with 0.6 nm spatial resolution and a precision of 0.55%.
  • Multi-modal Characterization: To get a complete picture, they didn't just look at strain. They used X-ray Absorption Spectroscopy (XAS) for electronic states, Pair Distribution Function (PDF) analysis for bulk atomic arrangements, and high-magnitude EDX for elemental mapping.
  • Theory-to-Experiment: They used Density Functional Theory (DFT) to create a "thermodynamic adsorption-energy landscape," directly linking the physical strain they saw under the microscope to the energy required for the chemical reaction.

Datasets

While this is an experimental physics/chemistry paper rather than a machine learning study, it relies on several massive datasets:

  • 4D-STEM Datasets: Arrays of diffraction patterns recorded at 256 x 256 scanning positions.
  • Synchrotron Data: High-energy X-ray PDF and XANES/EXAFS spectra from the SPring-8 and SSRL facilities.
  • Performance Benchmarks: The catalyst was evaluated against a comprehensive library of previously reported NO₃⁻RR catalysts (compiled in the paper's Supplementary Table 10).

Key Insights and Ideas

  • The "Ripple" Discovery: The most striking finding was that these nanocubes aren't uniform; they exhibit periodic, "ripple-like" fluctuations in both lattice strain and elemental concentration.
  • Cu Doping as a "Strain Smoother": By adding copper to a cobalt-tin hydroxide base (creating CuCoSn(OH)₆), the researchers found they could suppress heterogeneity. The Cu-doped version had longer strain periods and smaller amplitudes, leading to a more "homogeneous" surface.
  • The Strain-Activity Link: The team established a linear correlation between lattice strain and nitrate adsorption energy. Essentially, the "smoother" the strain ripples, the more efficiently the nitrate sticks to the active sites, accelerating the reaction.
  • Economic Viability: The paper calculates an estimated production cost of **$0.831 per kg of NH₃**, which is significantly lower than the current commercial price ($1.0–$1.5) of Haber-Bosch ammonia.

Key Results

  • Record Stability: The catalyst operated for over 1,000 hours at 200 mA cm⁻² in a single-layer membrane electrode assembly (MEA) without significant degradation.
  • High Efficiency: Achieved a 92.6% Faradaic efficiency for ammonia production.
  • Industrial Scaling: In a scaled-up, stacked MEA system, the catalyst produced ammonia at a rate of 2.82 g h⁻¹, demonstrating its readiness for pilot-scale applications.
  • Precision Imaging: Demonstrated the ability to map strain in particles 5x larger than previously possible with similar resolution.

Limitations and Future Work

  • Static vs. Dynamic: The authors acknowledge that their DFT models are "idealized and static." In a real reactor, the catalyst surface might undergo "dynamic reconstruction" (changing shape while working) that isn't fully captured by these models.
  • Material Scope: While highly effective for perovskite-type hydroxides, it remains to be seen if these specific "ripple" patterns are universal across other classes of industrial catalysts.
  • Future Direction: The researchers suggest this "universal framework" of multi-modal characterization should now be applied to other complex, multi-metallic catalysts to unlock "structure-property" laws that have remained hidden due to imaging limitations.

Summary updated

Abstract

Strain engineering serves as a pivotal strategy to optimize catalytic activity in electrocatalysis. However, the catalyst sizes under industrial conditions are usually large and even beyond nanometer regime. The critical methodological limitations on strain imaging of such catalysts with both large field of view and high spatial resolution obscure the mechanistic understanding of strain-performance correlations. Here, we present an optimized four-dimensional scanning transmission electron microscopy (4D-STEM) method to acquire strain mapping of both bulk and surface across particles up to 500 nm with 0.6 nm spatial resolution and 0.55% precision. We observe the ripple-like periodic strain coupled with elemental fluctuations inside a perovskite-type hydroxide CuCoSn(OH)<sub>6</sub> and find it correlated to electrocatalytic nitrate reduction (NO<sub>3</sub><sup>-</sup>RR) absorption energy to achieve the 92.6% Faradaic efficiency and long-term test over 1000 h at membrane electrode assembly (MEA) for ammonia electrosynthesis. This universal framework design offers a practical method that not only develops an advanced measurement combining multi-modal characterization techniques but also reveals the intrinsic structure-property constitutive law of industry-level catalysts.
BibTeX
@inproceedings{tao2026multi,
  author = {Yiyuan Tao and Xingyu Zheng and Shi Huang and Ershuai Liu and Yangjian Lin and Qike Jiang and Long Yang and Walter S. Drisdell and T. D. Xu and Yao Yang},
  title = {{Multi-modal characterization of nitrate reduction nano-catalysts with periodic strain distribution}},
  booktitle = {Nature Communications},
  year = {2026},
  doi = {10.1038/s41467-026-70447-7},
}
Figure