Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Surface science

Catalytic hotspots get noisy

Identifying and imaging catalytically active sites on solid surfaces is a grand challenge for science. A microscopy technique has been developed that images 'noise' to detect active sites with nanometre-scale resolution. See Letter p.74

Electrocatalysts speed up reactions involving electrons and chemical species, and such reactions include those key to clean-energy technologies1. Discovering and engineering electrocatalysts is therefore a crucial step in the development of a carbon-neutral economy based on renewable fuels. The problem is that we often don't know how electrocatalysts work at the molecular level, or even where on the catalysts' surface reactions take place. On page 74, Pfisterer et al.2 report that the structure and location of the active sites on electrocatalyst surfaces — the positions at which the catalytic rate is the fastest — can be directly imaged with nanometre-scale resolution.

Heterogeneous (solid) electrocatalysts have many applications, notably in fuel cells, where they oxidize hydrogen gas to produce protons (H+ ions) and reduce oxygen gas to water, generating electricity as the overall output. Electrocatalysts are also used to make hydrogen gas from water, liquid hydrocarbon fuels (such as methanol) from carbon dioxide, and fertilizer (ammonium nitrate) from nitrogen gas. But many electrocatalytic processes have efficiencies much lower than their thermodynamic limits, because of the slow kinetics of the catalytic reactions. Learning how to design faster electrocatalysts is therefore an important challenge in science and engineering.

Yet understanding electrochemical reactions on solid surfaces has been difficult. First, there is a lack of analytical tools that can follow molecular reactions on surfaces while the catalysts are working. The development of new analytical techniques is thus imperative3. Second, small changes in an electrocatalyst's surface structure can have drastic effects on the reaction rate — and even apparently perfect surfaces on single crystals of an electrocatalyst contain a variety of defects. Consequently, a few active sites, often located at these defects, catalyse essentially all of the chemical reaction. The analytical tools used to understand heterogeneous electrocatalysis must therefore not only provide atomic-level information about the structure at the surface of a catalyst under electrochemical conditions, but also distinguish between the majority of inactive sites and the minority of active ones.

The invention of scanning tunnelling microscopy4 (STM) in 1981 enabled the direct observation of surface structures at the molecular scale. STM operates by scanning a sharp metal tip over a surface while recording a current that results from the 'tunnelling' of electrons between the tip and the sample under an applied voltage. Variations in the measured current form the basis of the image. But although STM is a powerful method for imaging surface structures, and in some cases reactivity5, it has not been used to directly map catalyst activity in situ or under electrochemical conditions.

A complementary technique known as scanning electrochemical microscopy6 (SECM) was reported in 1989, and does allow surface catalytic reactivity to be mapped. In SECM, electrocatalytic products are sensed by reacting them electrochemically on a small metal electrode that is scanned over the electrocatalyst surface. Optimization of the experimental conditions, especially the development of nanoscale electrode tips, has enabled researchers to map the catalytic activity of single gold nanoparticles 10 nanometres in diameter7. However, the resolution of SECM is fundamentally limited: as the tip is moved closer to the surface to increase resolution, direct electron transfer to the substrate dominates the current signal, drowning out the electrochemical current that is used to map catalytic activity8.

Pfisterer and colleagues' technique combines STM's ability to achieve molecular resolution with SECM's ability to monitor where products evolve from a catalyst's surface in situ. In principle, the method is simple. A scanning tunnelling microscope in an electrochemically active environment is scanned across a catalytic surface and the tunnelling current is measured, as in conventional STM experiments. But when a voltage is applied to the catalyst such that it starts generating a reaction product, the microscope measures increased 'noise' in the tunnelling current if the tip is over an active site (Fig. 1).

Figure 1: An electrochemical scanning tunnelling microscope that resolves catalytic active sites.

a, Pfisterer et al.2 studied a process in which protons (H+ ions, which come from an electrolyte solution) react with electrons (e) from a platinum surface that is under an applied negative electrochemical potential (not shown) to produce hydrogen gas. Such reactions occur at catalytically active sites, depicted here as steps between terraces of atoms on the metal surface. The authors applied a voltage to the platinum surface to begin the reaction, and simultaneously measured the current generated by the tunnelling of electrons between the tip of a scanning tunnelling microscope and the sample as the tip was moved across the surface. b, The continual adsorption and desorption of reactants and products at the step edge under electrochemical conditions alters the tunnelling parameters, increasing the noise level in the current signal. This identifies the step edges as active sites.

The increased noise originates from changes in the STM current due to the local evolution of products — hydrogen gas, in the case of Pfisterer and co-workers' study. The products randomly modify the tunnelling barrier through which electrons must pass to be measured as STM current. Hence, the tunnelling-current profile differs on catalytic active sites compared with inactive ones, allowing a surface's activity to be mapped directly. Furthermore, the authors show that the magnitude of the noise correlates with the relative activity of different types of active site.

The authors resolve electrocatalytic active-site locations in situ with an unprecedented resolution of about 1–2 nm, but they do not yet reach the ultimate goal of catalysis science: the detection of active sites on solid catalyst surfaces with atomic resolution, in a way that also allows reaction intermediates to be identified and their sequence to be followed. One issue limiting the spatial resolution of the technique is that the products diffuse from the active site, making the area in which noise is observed larger than the true active site (probably by a factor of roughly twice the tip–substrate distance, approximately 1 nm).

The technique might also be limited by the materials to which it can be applied. Active sites on materials that have very low activity, or those where the active sites are distributed across the surface and closely spaced (near to the resolution of 1–2 nm) could be challenging to observe, because changes in noise will be small. Studying samples that are not close to being atomically flat may also be difficult.

Nonetheless, the new work is a breakthrough that opens up many opportunities for research. One system of interest for future study is nickel–iron oxyhydroxide9, which is among the fastest catalysts for water oxidation in alkaline conditions — a key process in the electrochemical splitting of water into hydrogen and oxygen. Does water oxidation occur at the edge or surface of these two-dimensional materials?

In the meantime, Pfisterer et al. have demonstrated that in catalysis, as in life, high levels of activity go hand in hand with high noise levels.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Seh, Z. W. et al. Science 355, eaad4998 (2017).

    Article  Google Scholar 

  2. 2

    Pfisterer, J. H. K., Liang, Y., Schneider, O. & Bandarenka, A. S. Nature 549, 74–77 (2017).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Sambur, J. B. & Chen, P . Annu. Rev. Phys. Chem. 65, 395–422 (2014).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Binnig, G., Rohrer, H., Gerber, Ch. & Weibel, E. Phys. Rev. Lett. 49, 57 (1982).

    ADS  Article  Google Scholar 

  5. 5

    Zambelli, T., Wintterlin, J., Trost, J. & Ertl, G. Science 273, 1688–1690 (1996).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Bard, A. J., Fan, F.-R. F., Kwak, J. & Lev, O. Anal. Chem. 61, 132–138 (1989).

    CAS  Article  Google Scholar 

  7. 7

    Sun, T., Yu, Y., Zacher, B. J. & Mirkin, M. V. Angew. Chem. Int. Edn 53, 14120–14123 (2014).

    CAS  Article  Google Scholar 

  8. 8

    Bard, A. J. et al. Science 254, 68–74 (1991).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Burke, M. S., Enman, L. J., Batchellor, A. S., Zou, S. & Boettcher, S. W. Chem. Mater. 27, 7549–7558 (2015).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Shannon W. Boettcher.

Related links

Related links

Related links in Nature Research

Imaging techniques: Nanoparticle atoms pinpointed

Chemistry: The long and winding road to catalysis

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Dette, C., Boettcher, S. Catalytic hotspots get noisy. Nature 549, 34–35 (2017).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing