CORE CONCEPTS

Core Concept: Capturing atoms in motionDanielle VentonScience Writer

Improvements in electron sources, lenses,and detectors have helped physicists makethe leap from seeing atoms as static toseeing them in motion. The latest advan-ces in the field of atomic microscopy have

culminated in the development of dynamictransmission electron microscopes (DTEMs),allowing researchers to see at a resolutionthat most would not have fathomed even adecade ago.

At Pacific Northwest National Labora-tory (PNNL), in Richland, WA, a team ofmicroscopists is closing in on a prizedtarget: the ability to film chemical reac-tions at the atomic scale. Scientists of allstripes, from disease researchers to bat-tery developers, are hungry for this capa-bility, according to project leader NigelBrowning.“The fundamental step in most chemical

interactions involves moving atoms around,”Browning says. “We generally know the ini-tial and final state, and have to guess at thesteps in between.”However, once you can seeevery step in a reaction, he explains, “youthen can say, if I alter that step in the mid-dle, I can get a different output.”Browning’s work is part of a larger push

within the microscopy field to capture atomsin motion by taking increasingly rapid pic-tures. The US National Institutes of Health(NIH) helped fund the purchase of PNNL’snew DTEM, and once it is operational inearly spring 2015, it will likely be used fora number of NIH projects.“The way people look at biological systems

in microscopes, currently, is that they haveto freeze them,” says Browning, referring tocryo-electron microscopy. “You can’t seeprocesses in action.” The Brownian motionof water molecules inside liquids knockssamples around so much that achievinghigh temporal resolution has been impos-sible before now.Transmission electron microscopes (TEMs)

were first developed in the 1930s, allowingthe first close images of cells and molecules,down to a resolution of about 10 nm,about twice the diameter of a hemoglobinmolecule. TEMs work by firing electrons,emitted by an electrostatic field, througha sample to form an image on film. Re-searchers have boosted spatial and tem-poral resolution over the decades by usinghigher-energy sources of electrons. In the1930s, TEMS could see a few nanometersacross; now they can zoom in on half anangstrom. Temporal resolution has re-mained largely unchanged: pictures everyfew milliseconds.DTEM uses photoemission—a property

that causes metals, such as tantalum, to emitelectrons when light shines on them—togenerate extremely large, quick pulses ofelectrons (more than 109 of them). The

Atomic microscopy, done using dynamic TEMs such as this one, have the potential to helpmaterials scientists to optimize reactions in areas ranging from battery design to medicalresearch. Image courtesy of Pacific Northwest National Laboratory.

www.pnas.org/cgi/doi/10.1073/pnas.1502841112 PNAS | April 21, 2015 | vol. 112 | no. 16 | 4835–4836

CORE

CONCE

PTS

Dow

nloa

ded

by g

uest

on

Mar

ch 1

0, 2

021

rapid bursts of electrons, sourced fromUV lasers, allow the researchers to captureatomic-scale images, down to about 2 Å(0.2 nm) every 100 ns. This speed makesthe DTEM under construction at PNNL10,000 times faster than any comparableelectron microscope in the world, accord-ing to Browning. And that, he says, is fastenough for “chemical reaction movies.”Bioremediation is one arena rife for

DTEM. As a microbe metabolizes a pol-lutant, this microscope offers a dynamicview of how compounds are processed andhow microbial communities develop inthe presence of pollutants. Ultimately, this

understanding could help scientists optimizeenvironmental cleanup projects.And there are other approaches in the

works. Ahmed Zewail, winner of the 1999Nobel Prize in chemistry (1), is working ondeveloping ultrafast electron microscopy, atechnique that uses just one electron perpulse. To form an image, the sample is

assaulted with electrons about one billiontimes. Meanwhile, the sample is pumpedwith a laser to induce a transient effect.Zewail’s method works best to measureeffects that are reversible (i.e., do not includestructural rearrangements), such as phenom-ena related to plasma oscillations (2).Andy Lupini, a materials scientist at Oak

Ridge National Laboratory, is part of a differ-ent team working to take “movies” of atomsin action. However, instead of being inter-ested in reactions on the surface, his teamimages atoms diffusing within a crystal: atechnique better suited to solids. In late2014, Lupini and collaborators reported thefirst observations of diffusion inside analuminum nitride crystal (3). Understandingthis movement will, they hope, lead tomore advanced light-emitting diodes.“Years ago [Richard] Feynman told phys-

icists that the best thing they could do forbiologists and other scientists, would be tomake the electron microscope much better,”says Lupini, referring to a speech at the 1959American Physical Society meeting duringwhich Feynman alluded to closer examina-tion of cellular components such as DNA,RNA, and proteins (4). “We’re part of tryingto make that happen.”Today’s high-powered electron microscopes

are 100 times more powerful than thoseFeynman had access to. And, when com-pleted, the PNNL microscope will be 10,000times faster than any comparable electron mi-croscope in the world. As the age of atomicmovies dawns, researchers are able to see pro-cesses they had only been able to imagine.

1 Nobel Media AB (2014) Ahmed Zewail - Facts. Nobelprize.org. Available at www.nobelprize.org/nobel_prizes/chemistry/laureates/1999/zewail-facts.html. AccessedFebruary 22, 2015.2 Center for Ultrafast Science & Technology, Ultrafast electronmicroscopy. CalTech.edu. Available at www.ust.caltech.edu/press/uem1.html. Accessed February 22, 2015.

3 Ishikawa R, et al. (2014) Direct observation of dopant atomdiffusion in a bulk semiconductor crystal enhanced by a large sizemismatch. Phys Rev Lett 113(15):155501.4 Feynman RP, There’s plenty of room at the bottom - Aninvitation to enter a new field of physics. Nanotechnology.Available at www.zyvex.com/nanotech/feynman.html. AccessedFebruary 15, 2015.

DTEM will capture extremely quick-succession images of atoms as they interact at solid–solid, solid–liquid, and solid–gas interfaces. In this illustration, a conventional microscopeimage shows the latter stages of how nanoparticles grow and agglomerate within a liquid.Once online, the DTEM will provide atomic scale snapshots of this growing process, spaced onlya few nanoseconds apart. Image courtesy of Pacific Northwest National Laboratory.

4836 | www.pnas.org/cgi/doi/10.1073/pnas.1502841112 Venton

Dow

nloa

ded

by g

uest

on

Mar

ch 1

0, 2

021