Teaching of transmission electron microscopy usually includes both lectures on the contrast theories, electron diffraction, etc., and practical hands-on operation of the microscope. The number of students attending the lectures is normally unlimited, but at the microscope, only a few persons can work at the same time. Since the microscopes are expensive, it would be of a great help if cheaper 'training' microscopes with basic imaging and diffraction capabilities were available. These functions, in fact, can quite easily be realized with fast personal computers and work stations, where the simulation of transmission electron micrographs and related diffraction patterns can help the student better understand the image formation processes. Adding text, audio and video help capabilities to the program, it can be made an efficient supplemental teaching tool.

TemTutor for Windows is based on microScope for Windows, which is a BF/DF TEM micrograph simulation program for dislocations and stacking faults.

The most useful classical definition of resolution is a property of an instrument alone, independent of the sample. Because of this it is not possible to define resolution simply for HREM or for most of the new microscopies. For weak phase objects, the best current HREM machines offer a point resolution very close to 0.lnm . For the first time, recent off-axis electron holography has extended resolution beyond the point resolution limit, in fulfillment of Gabor’s original aim. Other recent approaches to super-resolution have used through-focus series under highly coherent conditions, sums of images obtained under different illumination angles, dark-field STEM imaging with high-angle detectors,thin annular detectors in STEM, or the use of coherent nanodiffraction patterns. Aberration correction by electron optical means has undergone a recent revival, and the effects of third-order astigmatism are under study. If third-order spherical aberration is corrected, the projected charge density approximation may be more useful than the weak-phase object approximation, since the Scherzer focus then no longer applies.

The resolution of the high-resolution transmission electron microscope is limited by the specimen as well as by the HRTEM. For specimens that beam-damage, image resolution depends upon electron energy and electron dose. For small-cell crystalline specimens, Bragg’s law quantizes allowable reso-lutions, preventing image resolution from reaching instrumental resolutiqn. Specimen thickness be-comes increasingly important at higher resolutions. For a resolution of d(Å), we need the specimen to diffract at u=1/d. Consideration of the intersection of the Ewald sphere with the specimen shape trans-form (fig. 1) shows that thickness must be less than For a resolution pr 1.0A at 300keV, thickness must be less than 100A; to achieve 0.7A requires halving this value to 50A (fig.l).

Resolution in an image depends on the spatial frequencies of the information (diffracted waves) trans-ferred from the amplitude spectrum (specimen exit-surface wavefunction) into the image intensity spec-trum (Fourier transform of the image intensity).

As first pointed out by Lord Rayleigh a century ago, incoherent imaging offers a substantial resolution enhancement compared to coherent imaging, together with freedom from phase contrast interference effects and contrast oscillations. In the STEM configuration, with a high angle annular detector to provide the transverse incoherence, the image also shows strong Z-contrast, sufficient in the case of a 300 kV STEM to image single Pt and Rh atoms on a γ-alumina support. The annular detector provides incoherence by virtue of its large central hole, which is equivalent by Babinet's principle of complementarity to a bright field detector of the same size. For weakly scattering specimens, it shows greater contrast than the incoherent bright field image, and also facilitates EELS analysis at atomic resolution, using the Z-contrast image to locate the probe with sub-À precision. The inner radius of the annular detector can be chosen to reduce the transverse coherence length to well below the spacings needed to resolve the object, a significant advantage compared to light microscopy.

Many of the problems of transmission electron microscopy (TEM) are due to the fact that wave optics which governs the interaction of electrons with the specimen and the imaging process definitely is brought to an end with the detection of the final electron image. Unfortunately, resolution is limited by an increasing number of aberrations. Furthermore, wave optical tools in the electron microscope which are needed for example to produce phase contrast better than that given by the phase contrast transfer function, for distinction of amplitude contrast and phase contrast, or to measure phases in Fourier space, are only poorly developed. Since subsequent image processing of electron diffraction patterns or real space images can never compensate for the loss of phase information, the phase has to be recorded also, i.e. one has to work holographically to collect and reconstruct all data of the object structure.

Aberration correction in electron microscopy is a subject with a 60 year history dating back to the fundamental work of Scherzer. There have been several partial successes, such as Deltrap's spherical aberration (Cs) corrector which nulled Cs over 30 years ago. However, the practical goal of attaining better resolution than the best uncorrected microscope operating at the same voltage remains to be fulfilled. Combining well-known electron-optical principles with stable electronics, versatile computer control, and software able to diagnose and correct aberrations on-line is at last bringing this goal within reach.

We are building a quadrupole-octupole Cs corrector with automated aberration diagnosis for a VG HB5 dedicated scanning transmission electron microscope (STEM). A STEM with no spherical aberration will produce a smaller probe size with a given beam current than an uncorrected STEM, and a larger beam current in a given size probe.

The resolution of any microscope is usually taken as the benchmark of its quality. When thinking about the scanning electron microscope(SEM) a comparison of its resolution to the resolutions quot-ed for other comparable instruments such as the transmission EM or the scanning tunneling micro-scope has unfortunately led some users to the view that the SEM is only a second best choice. In fact the spatial resolution that can be anticipated for an optimized SEM is highly competitive with its peers making it clearly the microscopy of choice.

The resolution of the SEM depends on two classes of factors; those which are inherent to the use of electrons as an imaging tool, and those which relate to the performance of the instrumentation (and possibly of the operator) used to perform the microscopy. The use of electrons as the scanning probe, and the choice of secondary electrons as the usual imaging mode, potentially affects the achievable resolution in three ways.

The resolution in conventional light microscopy was defined by Abbe as: Wavelength (λ) and Numerical Aperture (NA) have a direct influence while the factor (c) is dependent upon the geometry of the illumination and observation beam paths. Some typical examples are:

High illumination apertures in darkfield or oblique illumination or specific ratios between illumination and observation aperture can substantially improve resolution. Various shapes of the illumination aperture, annular, point, centered or off axis at different azimuths directly influence numerical aperture and coherence and with this resolution, contrast and depth.

In mirror objectives low apertures become vignetted, the resolution is enhanced at the expense of contrast. In studies of Hopkins and Barham the factor “c” can range from 0.5 to 0.8. For the practically attainable condition, where illumination and observation aperture are identical, “c” becomes 0.61 and we have partial coherence.

Exceeding the observation aperture on the illumination side reduces “c” further but diffraction significantly reduces contrast.

The National Center for Electron Microscopy has recently acquired a field-emission TEM to form thebasis of a project to achieve a resolution of one Ångstrom. To reach this resolution, both instrumental and environmental factors need to be considered. We have designed and constructed a new building to provide a suitable environment for this instrument, with emphasis on providing isolation from external influences detrimental to the achievement of ultra-high resolution. Such influences include mechanical vibration, temperature fluctuations, acoustic noise, and stray electromagnetic fields.

The microscope chosen for the one-Ångstrom project is a Philips CM300 Ultra-Twin equipped with a field-emission gun. Pre-installation specifications provided by Philips for this 1.7Å-resolution TEM specify maximum-allowable values for vibration levels in three mutually-perpendicular directions. In the most critical direction (console left to right), vibration is required to remain below 0.8)μm/sec in the frequency range from 1Hz to 5Hz, although allowed to rise to 6μm/sec above 10Hz (Region I in fig. 1). Even when resolution is not a critical requirement, vibration must be minimized at 2.5Hz (Region II in fig.1).

The most interesting structures in materials science are non-periodic areas where the crystalline structure is disturbed such as interfaces, defects or dislocations. These non-periodic structures can be hidden by artefacts, caused by aberrations, and therefore they can be easier analysed if an aberration free imaging system can be used. Therefore, in order to improve the point resolution and to obtain easier access to the hidden information, a spherical aberration corrected 200 kV TEM, following a proposal by Rose, was set up.

Phase contrast in a transmission electron microscope (TEM) is obtained, as it was shown by Scherzer, due to the phase shifting power of the wave aberrations as there are: the defocus and the spherical aberration. The defocus can be optimised in terms of the well transferred bandwidth of spatial frequencies (Scherzer defocus) Δfsch = (Cs*λ.)1/2. The upper limit of the spatial frequency without a contrast reversal when choosing a Scherzer defocus is called the point resolution d ≈ 0.71 (Cs λ3)1/4.

The imaging of biological macromolecules at high resolution is predicated on the use of imaging modalities that combine a sufficiently high signal-to-noise ratio (S/N) to observe the desired detail with a sufficiently low dose not to have perturbed the structure at the resolution of that detail. For 3D structure determination the S/N has to be sufficiently high to permit the accurate position and orientation determination for individual molecules or assemblies.

For 2D crystal specimens the position is determined a priori by the lattice, orientation by the externally chosen tilt of the specimen, and signal-to-noise in a single unit cell enhanced by the translocational redundancy of the many unit cells of the crystal. This is close to an ideal specimen, which has permitted imaging with resolutions of 3.4 À at doses of 20-35 e/Â2[l]. Specimens with high internal symmetry, icosahedral or helical structures, are examples which confer intermediate structural redundancies to assist in reducing the requires electron dose.

Six years ago there was a symposium held at the 1991 EMS A meeting to discuss the issue of “Resolution in the Microscope”.1 In this paper, we will look at resolution in near-field imaging, a blossoming field, and see whether any of our concepts have changed.

It has been only within the last decade that the concept of super-resolution microscopy in the near field has been vigorously pursued and experimentally demonstrated. (For reviews on the subject, the reader is referred to the proceedings of the second and third international conferences on near field optics.) However, as in most areas of microscopy, the idea is not new, but rather rediscovered after decades of dormancy.

The idea of optical resolution unhindered by far-field diffraction limitations was conceived more than a half-century ago by E.H. Synge4 in a paper entitled “A Suggested Method for Extending Microscopy Resolution into the Ultra-Microscope Regime”.

In far-field microscopes, the spatial resolution is ultimately limited by the wavelength of the radiation used. While near-field and related microscopes can improve upon this, they can only do so with thin specimen regions. Thin specimens can also be studied at atomic resolution using electron microscopes. To achieve improved resolution on micrometer-thick specimens, another alternative is to use significantly shorter photon wavelengths. We discuss here the use of soft x-rays for microscopy and their resolution limits.

Image formation requires resolution and contrast. by using soft x-rays with a photon energy between the K absorption edges of carbon and oxygen, one is able to image hydrated biological specimens with high contrast. The contrast is such that no addi-tional staining is required, while efforts are also underway to utilize gold and luminescent probes for selective labeling. In addition, x-ray sources have high spectral resolution and good signal-to-background relative to electron microscopes which allows for elemental and chemical state mapping of major constituents.

Scanned Probe Microscopy first received widespread recognition in the form of scanning tunneling microscopy (STM) images clearly showing atomic resolution of the Si 111 surface in the characteristic 7×7 surface reconstruction. For this sample, STM imaging under carefully controlled ultrahigh vacuum conditions reveals the clear image of each atom position within the surface unit cell with excellent contrast and clearly atomic resolution. Over the past few years, versions of scanning force microscopy (commonly referred to as atomic force microscopy or AFM) have become much more widespread than STM. A very common, and very difficult question, is: What is the resolution of AFM? The simple answer is that SPM in general, and STM and AFM in particular, routinely obtain sub-angstrom resolution—in the z axis, or the sample height direction. This high resolution capability is easily demonstrated by scanning a cleaved crystal of known lattice spacing and observing single and multiple atomic steps.

The atom probe field ion microscope can resolve and identify individual atoms. This ability is demonstrated in a pair of field ion micrographs of an Ni3Al specimen, Fig. 1, in which the individual atoms on the close packed (111) plane are clearly resolved. Comparison of these two micrographs reveals that an individual atom was field evaporated between the micrographs. Due to the hemispherical nature of the specimen, the ability to resolve this two dimensional atomic arrangement is only possible on low index plane facets. The spatial resolution in field ion images is determined by a number of factors including specimen temperature, material, microstructural features, specimen geometry, and crystallographic location.

The spatial resolution of the data obtained in atom probe and 3 dimensional atom probe compositional analyses can be evaluated with the use of field evaporation or field desorption images. The field evaporation images are formed from the surface atoms with the use of a single atom sensitive detector whereas the field ion image is formed from the projection of a continuous supply of ionized image gas atoms.

It has long been realized that imaging with hollow cone illumination (HCI) should, in theory, improve the directly interpretable resolution of TEM by as much as 100% (albeit at the expense of contrast). The principle of HCI was first proposed by Scherzer in 1949 and then reinvented by Hanssen and Trepte in 1971. As opposed to axial illumination, HCI effectively eliminates zeroes and reversals of the transfer function providing direct interpretability of the resulting images. In addition to the substantial resolution enhancement, HCI should reduce significantly the phase-contrast noise inherent in axial HRTEM images. However, there are experimental obstacles for high resolution HCI which make its practical application very difficult to implement. To our knowledge, all observations using HCI so far have not shown all of the expected improvement predicted theoretically. This is believed to be due to the fact that accurate coma-free alignment is required to substantially improve the resolution.

Most scanning electron microscopy is performed at low magnification; applications utilising the large depth of field nature of the SEM image rather than the high resolution aspect. Some environmental SEMs have a particular limitation in that the field of view is restricted by a pressure limiting aperture (PLA) at the beam entry point of the specimen chamber. With the original ElectroScan design, the E-3 model ESEM utilised a 500 urn aperture which gave a very limited field of view (∼550um diameter at a 10mm working distance [WD]). An increase of aperture size to ∼lmm provided an improved but still unsatisfactory field of view. The simplest option to increase the field of view in an ESEM was noted to be a movement of the pressure and field, limiting aperture back towards the scan coils1. This approach increased the field of view to ∼2mm, at a 10mm WD. A commercial low magnification device extended this concept and indicated the attainment of conventional fields of view.

We have investigated the amplification properties of several imaging gases in the Environmental SEM (ESEM) with the intent of forming a set of general guidelines for the selection of gases. In the ElectroScan ESEM, a gas ionisation cascade is used to amplify the secondary electron signals emanating from the specimen surface. The presence of gas in the chamber also gives rise to a pressure dependent background signal derived from ionisation events between gas molecules and high energy primary beam and backscattered electrons. The fraction of secondary electron signal decreases as the pressure is raised. This point is illustrated in figures la and lb which show the calculated fraction of signal contributed by secondary, backscattered, and primary electrons as a function of pressure in helium and water vapour. Helium yields a very pure secondary electron signal over the entire range of pressures shown. Unfortunately, helium does not provide a great deal of signal amplification compared to water vapour (figure 2).

The mechanism of the contrast in ‘environmental’ or ‘gaseous’ secondary electron images in the environmental scanning electron microscope is at best poorly understood. The original theory suggested a simple gas amplification model in which emitted secondary electrons ionise the chamber gas, leading to signal amplification and finally measurement at a biased detector. This theory is being advanced but little attention has as yet been paid to the factors which influence the actual secondary emission, although unusual contrast effects have been noted in one case. The conven-tional view is that the positive ion product of the gas-electron interaction results in charge neu-tralisation at the sample surface.

The implantation and trapping of charge in non-conductive materials was recently described, in reference to electron range measurements. This work demonstrated that trapped charge influ-enced the secondary electron yield, with enhanced secondary electron emission above the region of trapped charge. The consequence is that the distribution of the trapped charge is seen as a bright circle on the surface of the specimen, centred on the point of beam exposure (Fig.l).

We have performed a theoretical and experimental study of the the effect that a surface layer of condensed water has on the emission of secondary electrons from the surface. This is an issue of considerable interest to users of the Environmental SEM (ESEM) when imaging wet samples. Previous work has been performed to investigate the effect of a layer of water on back scattered electrons (BSE), but secondary electron (SE) imaging is more commonly used in ESEM, so an understanding of the interactions of SE with water is important. The aim of this work is to quantify the thickness of water through which imaging is possible, by considering both the interactions of secondary electrons with the water, and the interactions of the water layer with the sample, which may affect the secondary electron emission coefficient, δ.

The effects that a surface layer of water may have on electron emission from a sample surface can be split into three regimes.