Elsevier

Ultramicroscopy

Volume 108, Issue 3, February 2008, Pages 179-195
Ultramicroscopy

An electron microscope for the aberration-corrected era

https://doi.org/10.1016/j.ultramic.2007.07.010Get rights and content

Abstract

Improved resolution made possible by aberration correction has greatly increased the demands on the performance of all parts of high-end electron microscopes. In order to meet these demands, we have designed and built an entirely new scanning transmission electron microscope (STEM). The microscope includes a flexible illumination system that allows the properties of its probe to be changed on-the-fly, a third-generation aberration corrector which corrects all geometric aberrations up to fifth order, an ultra-responsive yet stable five-axis sample stage, and a flexible configuration of optimized detectors. The microscope features many innovations, such as a modular column assembled from building blocks that can be stacked in almost any order, in situ storage and cleaning facilities for up to five samples, computer-controlled loading of samples into the column, and self-diagnosing electronics. The microscope construction is described, and examples of its capabilities are shown.

Introduction

Correction of electron-optical aberrations is a field with a 60-year history [1], [2], which has blossomed in the last 10 years. Aberration correctors have significantly improved the performance of scanning electron microscopes (SEMs) [3], transmission electron microscopes (TEMs) [4] and scanning transmission electron microscopes (STEMs) [5], [6], [7]. Aberration-corrected electron microscopes are now finding their way into mainstream usage in large numbers. The attainable STEM resolution has improved by about 2.5×, and the current available in a given-size small probe has increased by about 10×. Further major improvements are expected when a new generation of instruments such as the one described here come fully on-line.

The microscope column itself, on the other hand, has not improved to a similar extent. The electron guns, round lenses, objective lens (OL) polepieces, sample stages, and detectors available in present-day instruments are quite similar to the best available 20–30 years ago. This is exemplified by the Topcon 002B, introduced in 1986, which featured double-winding round lenses able to maintain constant heat output even as their magnetic excitation was changed, a condenser-OL with spherical aberration coefficient Cs=0.4 mm and chromatic aberration coefficient Cc=0.8 mm (at 200 kV), an illumination system consisting of five lenses, toroidal deflectors with minimized hysteresis, and a reasonably stable side-entry sample stage [8], [9]. It was a well-rounded instrument, optimized for 2 Å resolution, which was the norm for the top instruments of its day. Even the best non-aberration corrected dedicated STEM, the 300 kV VG HB603 [10], did not manage to improve on many of these characteristics.

The lack of progress in microscope design is surprising in view of the fact that aberration-corrected instruments are now approaching 0.5 Å resolution, and that atomic-resolution elemental mapping by electron energy-loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy (EDXS), which place serious demands on long-term stability, are about to become established techniques. The continued use of aging design elements has resulted in a situation whereby today's highest-performance microscopes are so sensitive that the designers of the microscopes’ foundations need to be concerned about the pounding of ocean waves on a shore 30 miles distant [11]. The microscopes need to be housed in custom-built labs with special vibration-free foundations, thorough sound shielding, minimized stray magnetic fields, and air temperature that is kept constant to within 0.1 °C [11], [12]. Even with these precautions, they remain sensitive to adventitious disturbances such as pressure changes due to doors opening and closing, and the low frequency magnetic fields due to passing trucks [12].

At Nion, we have decided to adopt a different approach: design an entirely new instrument in a way that takes the increased demands into account right from the start. Other fundamental goals of the design were to give the instrument as much flexibility as possible, and at the same time to automate its basic operation so that even novice microscopists can use the instrument near its full potential. One more very important goal was to make the microscope completely computer-controlled, so that its full functionality, including sample exchange, would be available to remote users over the Internet.

Section snippets

Design overview

The new microscope has been developed on two parallel but converging paths. Its electron-optical column was designed, built, and tested using a 100 kV VG cold field emission gun (CFEG) as the electron source, and is now fully functional. At the same time, a 200 kV CFEG is being developed separately. This gun is now fully built up, but it is still being tested. It will be the subject of a separate publication.

Fig. 1a shows the new column, and Fig. 1b shows its schematic cross-section. The VG 100 kV

Overall performance

There are many operating modes that the new microscope is expected to be able to perform, such as STEM and CTEM imaging, convergent and parallel beam diffraction, and EELS analysis. Most of the modes have been described in the preceding sections that covered the individual components of the column. The descriptions make it clear that the new microscope was designed to be very flexible already in its standard configuration.

The microscope offers even more flexibility if its electron-optical

Optics flexibility

The design aims of flexible condenser and projector optics have been largely met. Fig. 8 shows a pair of diffraction patterns acquired from the same area of [1 1 0]-oriented Si. For the first pattern (Fig. 8a), the probe was set up in a convergent mode, as required for STEM imaging. This gave a Ronchigram showing complicated interference patterns between the unscattered beam and the many diffracted beams. Note that even though the central disk spans a diameter of 50 mrad, the fringes seen in the

Conclusion

A new aberration-corrected STEM column has been designed and is now operational. It incorporates many new features that make it well suited for the high performance made possible by aberration correction, as described above.

The construction of the column gives it probe-forming and data-collecting capabilities that are essentially only limited by the characteristics of its CFEG. The flexibility of the column gives it many new operating modes, previously available only to those using custom

Acknowledgment

We are greatly indebted to our collaborators, especially to Drs. P.E. Batson, A.L. Bleloch, and S.J. Pennycook, and to Profs. L.M. Brown, C. Colliex and J. Silcox, for giving us the opportunity and challenge to build the new microscope, and for their many specific inputs. Partial funding from NSF (grant number DMR-9977547) and in kind-contributions from Gatan Inc. and Sandia National Laboratory are gratefully acknowledged. Further sincere thanks are due to Dr. A.L. Bleloch for his help in

References (24)

  • O.L. Krivanek et al.

    Ultramicroscopy

    (1999)
  • M.A. O’Keefe

    Ultramicroscopy

    (2001)
  • D.A. Muller

    Ultramicroscopy

    (2006)
  • O.L. Krivanek

    Ultramicroscopy

    (2003)
  • R. Vincent et al.

    Ultramicroscopy

    (1994)
  • Z. Yu et al.

    Ultramicroscopy

    (2003)
  • O. Scherzer

    Optik

    (1947)
  • P.W. Hawkes et al.
    (1996)
  • J. Zach et al.

    Optik

    (1995)
  • M. Haider

    J. Electron Microscopy

    (1998)
  • N. Dellby

    J. Electron Microscopy

    (2001)
  • P.E. Batson et al.

    Nature

    (2002)
  • Cited by (283)

    • Principles of Electron Optics, Volume 3: Fundamental Wave Optics

      2022, Principles of Electron Optics, Volume 3: Fundamental Wave Optics
    View all citing articles on Scopus
    View full text