Electron Microscopes Stand Corrected
Correction for the spherical aberration of lenses in electron microscopes is allowing scientists to push the resolution into the sub-angstrom range, opening the doors to new frontiers.
Microscopes are indispensable tools to researchers in various scientific fields, allowing them to examine the relationship of a material's properties to its structure. In these fields, the goal of microscopists is to achieve atomic scale resolution to view the arrangement of materials' building blocks.
However, the resolution of electron microscopes (EMs) has always been hampered by the spherical aberration (Cs) of the lenses, which both focus the beam onto the specimen and the images onto the various detectors. This aberration stems from the outer areas of the objective lenses (the outermost, principal imaging lenses), which overfocus, restricting the beam width. To address this problem, a Cs corrector and associated electronics are used to evaluate the overfocus, and then compensate for it by modifying the beam, thus removing the aberrations.
"Ease-of-use designs and full integration with other microscope functions are required to make aberration-corrected microscopes [like the JEM-2200FS above] accessible to all levels of scientific research," says JEOL's Michael Kersker.
Electron microscopy's path
"Scientists have known about aberration for a long time. We knew what to correct, but we couldn't build it. Today, we have a sufficient number of mechanisms for computer augmentation or computer control to allow us to actually make these corrections," says Nestor Zaluzec, a scientist in the Electron Microscopy Center at Argonne National Laboratory (ANL), Ill.
Focus on these corrections has intensified because "we've become more interested in studying not just the morphology of a material, but also its crystallography, its elemental and chemical composition, as well as its electronic structure. And we want to do that at the highest possible spatial resolution," says Zaluzec. The quest for higher resolution is justified "because we have found by image simulations that defects in gallium nitride, for instance, could be characterized more thoroughly if we could reach 0.6-Å resolution, instead of 'only' 0.78 Å," says Michael O'Keefe, a staff scientist III/physicist in the Materials Sciences division at Lawrence Berkeley National Laboratory (LBNL), Calif.
As a result, scientists are now concentrating their efforts on the sub-angstrom frontier. To cross that barrier, microscopists are going back to look at EMs to study their limitations. "Some of the limitations can be traced down to the instrument. We've finally reached the limit of one of the fundamental considerations, namely, the lenses of the microscope," says Zaluzec.
A TEAM project
The Dept. of Energy's recently funded Transmission Electron Aberation-corrected Microscopy (TEAM) project is targeted at this issue. Its first goal is the design, construction, and operation of an aberration-correcting instrument that will hit the 0.5-Å barrier. The project involves five national labs: ANL (Electron Microscopy Center), Lawrence Berkeley, Calif., (National Center for Electron Microscopy), Oak Ridge, Tenn., (Shared Research Equipment Program), Brookhaven, Upton, N.Y., and the Frederick-Seitz Materials Research Laboratory (Center for the Microanalysis of Materials) at the Univ. of Illinois, Urbana. These facilities have also formed alliances with various industrial microscope manufacturers to exploit their respective expertise.
The pursuit of the TEAM's goal will be achieved by looking at the different components of EMs: the illumination source, lenses that focus the illumination source onto the sample, sample area, lenses that take the signal from the sample unto a detector, and the various detector systems used to measure all the signals.
"Most of today's commercial aberration-corrected microscopes focus on enhancing the signal from the sample to the detector," says Zaluzec. The TEAM project's researchers, on the other hand, are examining all five elements, viewing the instrument as a "system" for doing analysis. "We're looking at the post-specimen, pre-specimen, the illumination source, specimen stage, and the detectors" to figure out what is needed to reach the 0.5-Å barrier and determine the limitations.
These limitations involve not only the lenses, but also how to guarantee a sufficient stability for the electron beam and the stage holding the sample at 0.5 Å. At that level, the environmental conditions in the room where the instrument is located become a concern. "A person just walking through the room, 1 m away from the instrument will be enough to affect the temperature of the room, causing air currents and/or vibrations, which will affect the instrument, even though it may weigh several tons," says Zaluzec.
Aiming at a resolution below 0.9 Å, Zeiss's SATEM is equipped with a corrected 90°-Omega Filter, allowing energy-filtered TEM applications.
Divide and conquer
Once a basic aberration-corrected microscopy platform is developed, the TEAM will then customize it to satisfy the research interest of each of the five national labs. For instance, one prototype, the first to be built, will have an image resolution at the 0.5-Å level and will go to the Berkeley lab. Another prototype will be optimized for in situ application of magnetic and electrostatic fields. This tool will go to ANL, where scientists are interested in analytical spectroscopy and in situ effects. "It will allow us to visualize and measure the effects of internal magnetic fields of materials at the atomic level," says Zaluzec.
The ANL researcher cautions, however, that it will not be possible to produce one machine that will offer all the capabilities found in each of the customized microscopes. "It's a classic error that everybody makes," says Zaluzec. "Some people in the field think they can make one machine that does everything. But to do that, you always make a compromise." The solution being undertaken involves having a different instrument at each of the five labs that is optimized for a specific type of work. As a result, high-performance microscopy user facilities will be scattered across the country. Moreover, these facilities will be available to scientists in the U.S. and remotely accessible to others around the world through Telepresence. LBNL's O'Keefe also warns that with each enhancement in resolution, new challenges will arise. "Resolutions to 0.4 Å will require correctors for additional aberrations, including four-fold-possibly five-fold-astigmatism, fifth-order spherical aberrations, and perhaps others. Fortunately, the fourth and fifth order aberrations are well understood" and can be corrected.
Another reason for this divide-and-conquer approach is the high cost of such aberration-corrected machines. Their price is expected to be two to three times that of high-resolution commercial microscopes. Another factor to consider is that these microscopes will not be accessible to the average person. "You need a lot of expertise not only to run these machines, but also to understand what's going on. Highly-skilled people are needed to optimize the experiment, to make sure the right measurement is done," says Zaluzec.
Microscopists working with scanning electron microscopes (SEMs) have always been trying to overcome the challenge associated with viewing biological samples. The hurdle stems from the vacuum in the microscope's viewing chamber, which leads to the evaporation of biological tissues' liquid water. Typically, researchers bypass this problem by treating the samples with drying solvents, quick-freezing them, or covering them with a very thin gold layer. However, a new option has emerged, which allows the viewing of hydrated biological specimens, thus eliminating the need for tissue-distorting prep.
Developed by scientists at the Weizmann Institute of Science, Rehovot, Israel, the method involves the use of a thin but strong polymer capsule, which envelops the specimen, enabling it to endure the vacuum's force. The capsule works well with SEMs because its material allows the electrons to pass through. The technique promises to be especially beneficial to the study of lipids, which are destroyed by the other prep treatments.
Zaluzec's interest in the aberration-corrected microscopes is not in the 0.5-Å image resolution, however. His interest lies in the spectroscopy and in situ fields, specifically in intensely focused electron probes. "Imagine a 0.5-Å probe walking on your sample. And at each point in your sample, not just getting an image, but also doing analytical work. For example, measuring the composition, the electronic structure, and the crystal structure," says Zaluzec.
The use of aberation correction features allows the size of the probe to be smaller and increases the number of electrons in the probe. The hurdle remains, however, in figuring out how small of a probe can be built, and the number of electrons that can fit in it. Once these two factors are determined, the probe can then be used to hit the sample, and a detector can perform measurements. It is at this stage that the work becomes interesting to Zaluzec.
"The space in the lens where the sample goes in is very tiny. The tinier it is, the smaller the aberration, and the better the image. But the tinier you make it, the harder it is to get detectors in." In today's standard microscopes, that lens space is about 0.5 mm. Zaluzec would like to see the space increased to 10 mm to bring in a variety of detectors. He is willing to forego absolute resolution in exchange for "more space to make the microscope into a pico-scale lab to do analytical work." He envisions such an instrument being available in less than 10 years.
On the industrial front, as microscope manufacturers await the results of the TEAM project, some organizations have already broken the 1-Å image resolution barrier. The first such company is FEI, Hillsboro, Ore., whose nanotechnology center has crossed the barrier using a 200 kV transmission EM (TEM). The Tecnai F20 ST TEM combines a Cs multipole corrector system from Corrected Electron Optical Systems (CEOS) Gmbh, Heidelberg, Germany, with its own proprietary electron beam monochromator technology, integrated in the software. This combination enables in situ observation of sample reaction to different conditions, 3-D reconstruction with tomography, as well as scanning probe applications, with sub-angstrom resolutions.
The Tecnai "allows direct interpretation of the images without artifacts at resolutions below 1 Å," says Dominique Hubert, TEM marketing manager at FEI's Netherlands site. As a result, boundaries and interfaces can be viewed without delocalization effects. With this ability, "the doors have been opened for researchers working in nanotechnology development to explore materials down to the next level of resolution, which had, until now, been impossible," says Hubert. And since in nanotechnology macroscopic physical characteristics closely depend on the microscopic local structure, significant improvements in this area are expected.
Another company focusing on nanoengineering structures through the use of sub-angstrom microscopes is Carl Zeiss, Oberkochen, Germany. At their Nano Technology Systems division, the Sub-Angstrom-TEM (SATEM) and Sub-Electronvolt-Sub-Angstrom Microscope (SESAM) projects were designed to attain the highest resolution for energy filtering TEMs, thus enhancing imaging resolution and analysis. Partly government funded, these instruments resulted from a partnership between Zeiss and the German Society for Physics, Bad Honnef. Not commercially available, the SATEM and SESAM have, however, been delivered to three German universities.
"In microscopy, it's a constant drive for better resolution," says Jan Pieter Vermeulen, marketing director at Zeiss. "If you can image atoms, then you can image the structure. If you know how it's built, either in nature or by man, you can reproduce it, nanoengineer it." This ability to image, build, and manipulate oat the nanoscale can produce more effective products in different fields. For instance, in the sector of drug discovery, if the structure of a specific drug is identified, it can then be manipulated to penetrate the membrane more effectively, creating better treatments. Another example involves semiconductors, which today are composed of 20 to 30 layers on top of each other. If the structure at the interface between the different layers is better discerned, a semiconductor with more homogeneous layers can be fabricated. Besides possessing fewer defects, "such a semiconductor makes better use of resources, leading to faster processes," says Vermeulen.
Looking ahead, Zeiss will try to improve the resolution of EMs by getting better field emission sources, monochromators, in-column filtering, energy filtering, and Cs correctors.
"The Cs corrector represents a major step in a new direction in electron optics and will be important in all future microscope designs," says Michael Kersker, VP/product manager of TEM and SP Product Group at JEOL-USA, Inc., Peabody, Mass. "When one uses a Cs corrector, the magnetic constraints that were imposed on non-corrected instruments are relaxed. This means the magnetic field can be weaker, the physical consequence of which is more room in the microscope for experiments that can be done while the samples are being observed at high resolution," says Kersker.
At JEOL, all of the correctors that were incorporated into their 200 kV field emission TEMs and scanning TEMs (STEMs) were designed and built by CEOS. The integration yielded TEMs with a resolution of 1 Å, and STEMs, whose resolution is 0.8 Å. "We expect the STEM resolution to ultimately be on the order of 0.5 Å and the TEM to be better than 1.0 Å," says Kersker.
Such STEM Cs-corrected microscopes have already been delivered to different labs, among them the microscope facilities at Lehigh Univ., Bethlehem, Pa. At these facilities, the instruments are used by researchers for chemical analysis to discover a single impurity atom in a sample.
Argonne National Laboratory, 630-252-2000, www.msd.anl.gov/groups/emc
Brookhaven National Laboratory, 631-344-8000, www.bnl.gov
Carl Zeiss SMT AG, 49-7364-20-44-88,www.smt.zeiss.com
CEOS Gmbh, 49-6221-894670,www.ceos-gmbh.de
FEI Co., 503-726-7500, www.feicompany.com
Frederick-Seitz Materials Research Laboratory, 217-333-1370, www.mrl.uiuc.edu
German Society for Physics, 22-24-92-32-0, www.dpg-physik.de
JEOL-USA, Inc., 978-535-5900, www.jeol.com
Lawrence Berkeley National Laboratory, 510-486-4000, www.lbl.gov
Lehigh Univ., 610-758-3000, www.lehigh.edu
Oak Ridge National Laboratory, 865-574-6106, www.ornl.gov
Weizmann Institute of Science, 972-8-934-2111, www.weizmann.ac.il