Deep neural network analysis of nanoparticle ordering to identify defects in layered carbon materials

Boiko, Daniil A.; Pentsak, Evgeniy O.; Cherepanova, Vera A.; Gordeev, Evgeniy G.; Ananikov, Valentine P.

doi:10.1039/d0sc05696k

Cited by 17 publications

(12 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the 1 wt % Pd on GR, the nanoparticles were larger and had a broader size distribution with a mean of 5.2 ± 0.5 nm; their scattering was even, although it clearly followed the domain-like structure described previously. 45,50 By contrast, 10 wt % Pd on GR represented an assortment of sizes and shapes with significantly larger aggregates and highly developed surfaces. Surprisingly, the large metal agglomerates in 10 wt % Pd on GR consisted of relatively uniform nanoparticles with a narrower distribution than 1 wt % Pd on GR.…”

Section: Resultsmentioning

confidence: 98%

“…Presently, the processing of large amounts of microscopy data has become possible due to the development of machine learning techniques. Neural networks are already used to characterize nanoscale objects in both static images , and very information-rich video recordings − in electron microscopy and microscopy, in general. Thus, these algorithms are an excellent choice for the complete characterization of the nanocatalyst surface.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Toward Totally Defined Nanocatalysis: Deep Learning Reveals the Extraordinary Activity of Single Pd/C Particles

et al. 2022

Self Cite

View full text Add to dashboard Cite

Homogeneous catalysis is typically considered "welldefined" from the standpoint of catalyst structure unambiguity. In contrast, heterogeneous nanocatalysis often falls into the realm of "poorly defined" systems. Supported catalysts are difficult to characterize due to their heterogeneity, variety of morphologies, and large size at the nanoscale. Furthermore, an assortment of active metal nanoparticles examined on the support are negligible compared to those in the bulk catalyst used. To solve these challenges, we studied individual particles of the supported catalyst. We made a significant step forward to fully characterize individual catalyst particles. Combining a nanomanipulation technique inside a field-emission scanning electron microscope with neural network analysis of selected individual particles unexpectedly revealed important aspects of activity for widespread and commercially important Pd/C catalysts. The proposed approach unleashed an unprecedented turnover number of 10 9 attributed to individual palladium on a nanoglobular carbon particle. Offered in the present study is the Totally Defined Catalysis concept that has tremendous potential for the mechanistic research and development of high-performance catalysts.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

Toward Totally Defined Nanocatalysis: Deep Learning Reveals the Extraordinary Activity of Single Pd/C Particles

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…The high specific surface area of the substrate helps to increase the dispersion of metal particles. The defects and doping components are responsible for the strength of the bond between the metal and the substrate, and can also provide specific electronic effects [69]. It is often noted that the presence of a large volume of micropores in the substrate can reduce the activity of the supported catalyst due to mass transfer limitations [70,71].…”

Section: Which Properties Of the Support Determine The Activity Of The Catalyst?mentioning

confidence: 99%

How to Make a Cocktail of Palladium Catalysts with Cola and Alcohol: Heteroatom Doping vs. Nanoscale Morphology of Carbon Supports

et al. 2021

Self Cite

View full text Add to dashboard Cite

Sparkling drinks such as cola can be considered an affordable and inexpensive starting material consisting of carbohydrates and sulfur- and nitrogen-containing organic substances in phosphoric acid, which makes them an excellent precursor for the production of heteroatom-doped carbon materials. In this study, heteroatom-doped carbon materials were successfully prepared in a quick and simple manner using direct carbonization of regular cola and diet cola. The low content of carbon in diet cola allowed reaching a higher level of phosphorus in the prepared carbon material, as well as obtaining additional doping with nitrogen and sulfur due to the presence of sweeteners and caffeine. Effects of carbon support doping with phosphorus, nitrogen and sulfur, as well as of changes in textural properties by ball milling, on the catalytic activity of palladium catalysts were investigated in the Suzuki–Miyaura and Mizoroki–Heck reactions. Contributions of the heteroatom doping and specific surface area of the carbon supports to the increased activity of supported catalysts were discussed. Additionally, the possibility of these reactions to proceed in 40% potable ethanol was studied. Moreover, transformation of various palladium particles (complexes and nanoparticles) in the reaction medium was detected by mass spectrometry and transmission electron microscopy, which evidenced the formation of a cocktail of catalysts in a commercial 40% ethanol/water solution.

show abstract

“…Recently, a solution to this problem—the application of deep learning algorithms—has become increasingly widely adopted. Examples of usages include biomedical applications, 56 analysis of pharmaceutical powders, 57 protein nanowires, 58 catalysts, 59 and analysis in a liquid phase. 60 Besides segmentation and detection tasks, deep learning-based image inpainting has also performed.…”

Section: Introductionmentioning

confidence: 99%

Digital biology approach for macroscale studies of biofilm growth and biocide effects with electron microscopy

Kozlov,

Boiko,

Detusheva

et al. 2023

Digital Discovery

Self Cite

View full text Add to dashboard Cite

show abstract

Deep neural network analysis of nanoparticle ordering to identify defects in layered carbon materials

Abstract: Defectiveness of carbon material surface is a key issue for many applications. Pd-nanoparticle SEM imaging was used to highlight “hidden” defects and analyzed by neural networks to solve order/disorder classification and defect segmentation tasks.

Cited by 17 publications

References 75 publications

Toward Totally Defined Nanocatalysis: Deep Learning Reveals the Extraordinary Activity of Single Pd/C Particles

Toward Totally Defined Nanocatalysis: Deep Learning Reveals the Extraordinary Activity of Single Pd/C Particles

How to Make a Cocktail of Palladium Catalysts with Cola and Alcohol: Heteroatom Doping vs. Nanoscale Morphology of Carbon Supports

Digital biology approach for macroscale studies of biofilm growth and biocide effects with electron microscopy

Contact Info

Product

Resources

About