The same reaction produced either gold-coated copper oxide cubes or hollow gold boxes depending on whether it was observed with an electron beam or soft X-rays. The findings show that the beam used for real-time imaging can alter the chemistry around a sample and influence the final product.

Observing a reaction under a microscope is usually associated with passive recording. But in electron and X-ray microscopy, the image is generated by a beam that delivers energy to the sample. When the reaction takes place in a liquid, that energy can split water molecules and change the local chemical environment.

The microscope therefore ceases to be merely an observer and can influence the outcome of the experiment.

A team from the Institute of Nuclear Physics of the Polish Academy of Sciences, the University of Warsaw and Jagiellonian University investigated this phenomenon in a study published in the scientific journal Small (doi: 10.1002/smll.75209).

The scientists examined a reaction between copper(I) oxide nanocubes and a solution containing gold ions. During the reaction, the gold ions gain electrons and are deposited as metallic gold, while copper in the oxide loses electrons and can dissolve into the solution.

If the copper oxide cube dissolves as the gold is deposited, a hollow gold shell is left behind.

The researchers used two imaging techniques to observe the process. In a transmission electron microscope, the sample was enclosed in a tiny liquid cell separated from the microscope's vacuum by thin membranes, with an electron beam passing through the cell.

In the second experiment, the sample was scanned with soft X-rays at the SOLARIS synchrotron in Kraków. This technique also allowed the researchers to determine the composition of the material and the oxidation state of the copper.

The starting point was the same, but the results were different.

Under the electron beam, gold accumulated on the surface of the cubes, forming a coating, while the copper oxide interiors remained intact. The result was a core-shell structure with a gold casing.

During X-ray imaging, the copper template dissolved as the gold was deposited, leaving behind a hollow gold box.

The difference was not a quantum “observer effect”, but radiation-induced chemistry.

Both beams caused radiolysis of the water, breaking it down into highly reactive substances. Under the experimental conditions, the electron beam produced a more acidic environment, favouring the acceptance of electrons by gold ions. Gold therefore deposited on the surface while the dissolution of copper oxide was inhibited.

The X-rays, by contrast, created a more alkaline environment that promoted copper oxidation and allowed the interior of the cube to dissolve.

The researchers tested this explanation by changing the composition of the solution inside the electron microscope. After sodium hydroxide was added, the copper oxide began to dissolve and hollow structures formed even under the electron beam.

This confirmed that the outcome was determined by the chemical environment created around the nanocube. Although the detailed course of the reaction differed from that observed under X-ray irradiation, models of ion transport, surface reactions and water radiolysis were consistent with the explanation.

The findings raise questions about the reliability of real-time measurements made with liquid-cell microscopes. Such techniques allow researchers to observe processes including nanoparticle formation, corrosion, catalyst performance and transformations in materials used in the energy sector.

But if the imaging beam changes the acidity or redox potential of the solution, part of the process being recorded may be an artefact of the measurement rather than an intrinsic property of the material under study.

The type and dose of the beam, as well as the composition of the solution, must therefore be treated as experimental variables, with their effects checked through control experiments and modelling.

At the same time, the researchers point out that the same effect could be used deliberately to control chemical reactions and fabricate nanostructures.

The thin walls, large surface area and internal cavities of gold nanoboxes can facilitate chemical reactions, making similar structures potential candidates for use as catalysts. However, their practical application was not the focus of this study. Instead, the gold structures served as a way to determine whether the observation method itself was changing the reaction.

The project was carried out by scientists from the Institute of Nuclear Physics of the Polish Academy of Sciences, where the nanocubes were prepared and modelling was performed. The Biological and Chemical Research Centre of the University of Warsaw participated in the electron imaging, while X-ray measurements were conducted at the DEMETER beamline of the SOLARIS synchrotron at Jagiellonian University.

Computational work was carried out at Cyfronet AGH, and the project was funded by the Polish National Science Centre.

The study focused on a single model system, so the findings do not mean that electrons and photons will affect every liquid-phase reaction in the same way. (PAP)

kmp/ zan/

tr. RL