EUVIMEDEUVIMED entdecken
Crossref

So Many Atoms and So Little Time

Thoams F Kelly

Microscopy and Microanalysis · 27.07.2026 · 32 (Supplement_1)

DOI: 10.1093/mam/ozag053.002 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Microscopy and Microanalysis, online erschienen am 27.07.2026. Verfasst von Thoams F Kelly. Die Arbeit nennt 11 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Humans are explorers. We rely on vision as our primary means of learning about and understanding our environment; without it, our knowledge of the world would be far more limited. The human eye can see details over about five orders of magnitude in length from the super-meter scale down to the sub-millimeter scale. But human knowledge today spans 60 orders of magnitude in length. How do we know about these other length scales? Simply put, telescopes on the large scale (25 orders) and microscopes on the small scale (12 orders). Knowledge of subatomic phenomena comes from other sources and accounts for another 23 orders of magnitude. Let’s focus on the small scale (pun intended). What is the role of microscopy in the world? If we cannot see something, then our understanding of the entity may be incomplete and could be wrong. Conversely, the ability to acquire deep knowledge about an object can lead to opportunities to improve, optimize, and discover. For example, before the invention of the light microscope (a mere 350 years ago), early theories of some diseases like the bubonic plague (∼1350 CE) were far from correct and now seem almost comical. Leading scientists of the time pointed to astronomical events as the cause of the plague. The culprit, bacteria, were not known at the time. Knowledge of microbes such as bacteria was missing because they are smaller than the human eye can resolve. These early theories may seem silly now, but they were the best achieved with the information available. Without knowledge, we are left with only our imagination. Microscopy can be thought of as a feedback mechanism for gaining knowledge about a material. With it, we gain essential insight. Without it, we are flying blind. Nature and humans today both engineer materials and devices at the atomic scale. If we are making things at the atomic scale, then we require atomic-scale microscopy. So too, if we are to fully understand natural structures, we must understand them at the atomic scale. This begs the question: Where should atomic-scale microscopy be headed? To answer this, an actual example of how atomic-scale microscopy may be invaluable can be illustrative. In the 1980s, rapid solidification processing of materials was producing many new effects on materials’ microstructures. The resulting properties were outstanding in many cases. It was essential then that the materials’ microstructure be fully understood if we were to control and reproduce these advancements. Fig. 1 is an image of one such microstructure that contained a high number density of very fine (∼10 nm) and novel precipitates [1]. We could quantify their morphology with the electron microscope, but we needed to know their composition. Using one of the finest electron microscopes of the time, a VG HB501 STEM, we tried to determine the composition of the small precipitates with energy-dispersive X-ray analysis. The results were ambiguous. Afterall, we were analyzing 10 nm diameter precipitates in a 100 nm thick foil. Today’s microscopes could do much better but still would not approach the ideal of quantifying a large number of precipitates for detailed composition and internal structure. What we wanted to know ideally was the identity and location of every atom in a large number of these precipitates. In the 1990s, the three-dimensional atom probe became available, and this was a major step in the right direction [2]. With this tool, a three-dimensional image of the identity and location of the atoms in the structure is obtained. There remain, however, three limitations that are noteworthy. 1) The fraction of the atoms that were detected was about 60%. It is now about 80%. This is good but the ideal microscope would detect 100% of the atoms. 2) The spatial positioning precision for the atoms had local and global errors. The ideal microscope would have high precision in atom positioning. 3) The size of a typical image contains between 106 and 108 atoms. Such a microscope would ideally produce much larger images with perhaps 1010 to 1011 atoms. If the analyzed volume is large enough, then it can become representative of the bulk properties and efforts to link atomic-scale microscopy to properties could ensue [3,4]. This would become a structure-properties microscopy. There are efforts to address each of these limitations which, taken together, could lead to the ideal microscope. 100% detection efficiency has been pursued with new detector technology such as superconducting detectors [5]. There are also room temperature approaches that may bear fruit. One possibility is to push tapered MCPs to the limit. Regardless, the heart of an atom probe is the detector and new detector technology is needed to address this and the image size limitation. A key reason that image size is typically 108 atoms or less in atom probes is that the data collection rate is about 104 atoms/s. A thousand-fold increase in the maximum data rate is needed to realistically get to the 1011 atom images. One major limitation of the current detector technology (delay lines) is the poor multihit performance which makes it necessary to limit the probability of atom evaporation during pulsing to small numbers like 10-2 atoms per pulse. Detectors that can correctly encode 10 atoms per pulse would enable this factor of 103 increase in evaporation probability and hence, data collection rate. It is noteworthy that early experimenters in atom probe during the 1960s and 1970s routinely evaporated thousands of atoms per (voltage or laser) pulse but it remains to be seen if this can be done on a wide spectrum of materials and at high repetition rates (like 105 pulses/second) for billions of atoms [6-8]. Independently, efforts have been pursued by several groups to combine the atom probe with an electron microscope to improve the atomic positioning precision and accuracy [4,9,10]. There is no known physics that prevents us from achieving the above improvements in atomic-scale microscopy. Think back to what Erwin Müller must have pondered about someday recording thousands or even millions of atoms from an atom probe. It may take a few decades, but the advances described above are well within the career expectancy of today’s young engineers. The vision expressed here is to open atomic-scale microscopy to much larger dimensions and greatly expand its reach and value. Consider that the atom can be considered the building block of nature. If we can locate each building block of a structure, we can in principle fully understand and potentially reproduce and improve that structure. In Fig. 1, the yellow rectangle depicts the approximate volume of 1011 atoms in this steel. By using these data, we might be able to determine the macroscopic mechanical properties of the steel and apply machine learning to predict new compositions and processes that could result in even better properties. We might call this Structure-Properties Microscopy [3,4]. In this way we can push microscopy to new limits and, indeed, a new paradigm. Wheel, Fire, Microscope [11]. These three inventions have been touted as humanity’s greatest. For the physical and potentially the life sciences, access to atomic-scale information at much larger scales has real potential to be transformative. In atomic-scale microscopy, there is still plenty of room at the top. Yes, humans are explorers and both inner space and outer space await. STEM bright-field image of a rapidly solidified 303 stainless steel powder from the author’s early career. Determination of the atomic-scale makeup of the small precipitates in the structure (region highlighted by yellow arrow) was essential but was very challenging and not sufficiently informative with a field-emission STEM at the time [1]. The yellow rectangle illustrates what a 100 billion atom volume would look like if the specimen were 1 micron thick.

Zitieren

APA

Kelly, T. F. (2026). So Many Atoms and So Little Time. Microscopy and Microanalysis, 32(Supplement_1). https://doi.org/10.1093/mam/ozag053.002

Vancouver

Kelly TF. So Many Atoms and So Little Time. Microscopy and Microanalysis. 2026;32(Supplement_1). doi:10.1093/mam/ozag053.002

BibTeX
@article{Kelly2026,
  author = {Kelly, Thoams F},
  title = {So Many Atoms and So Little Time},
  journal = {Microscopy and Microanalysis},
  year = {2026},
  volume = {32},
  number = {Supplement_1},
  doi = {10.1093/mam/ozag053.002}
}
Kostenlos starten →