Search

Electron-beam-induced deposition of various materials from suitable precursors has represented an established branch of nanotechnology for more than a decade. A specific alternative is carbon deposition on the basis of hydrocarbons as precursors that has been applied to grow various nanostructures including masks for subsequent technological steps. Our area of study was unintentional electron-beam-induced carbon deposition from spontaneously adsorbed hydrocarbon molecules. This process traditionally constitutes a challenge for scanning electron microscopy practice preventing one from performing any true surface studies outside an ultrahigh vacuum and without in-situ cleaning of samples, and also jeopardising other electron-optical devices such as electron beam lithographs. Here we show that when reducing the energy of irradiating electrons sufficiently, the e-beam-induced deposition can be converted to e-beam-induced release causing desorption of hydrocarbons and ultimate cleaning of surfaces in both an ultrahigh and a standard high vacuum. Using series of experiments with graphene samples, we demonstrate fundamental features of e-beam-induced desorption and present results of checks for possible radiation damage using Raman spectroscopy that led to optimisation of the electron energy for damage-free cleaning. The method of preventing carbon contamination described here paves the way for greatly enhanced surface sensitivity of imaging and substantially reduced demands on vacuum systems for nanotechnological applications.

The information depth of imaging with backscattered electrons in the scanning electron microscope was measured directly by comparison of a planar view of a steel sample with buried precipitates with an image of a section cut through precipitates by means of a Focused Ion Beam attachment. The information depth was determined across the electron energy range of units of keV and the results were compared with information depths calculated from two published equations and simulated using widely available Monte Carlo-based software. Recommendations are given for routine estimations of BSE information depths.

Low voltage electron microscopes working in transmission mode, like LVEM5 (Delong Instruments, Czech Republic) working at accelerating voltage 5 kV or scanning electron microscope working in transmission mode with accelerating voltage below 1 kV, require ultrathin sections with the thickness below 20 nm. Decreasing of the primary electron energy leads to enhancement of image contrast, which is especially useful in the case of biological samples composed of elements with low atomic numbers. As a result treatments with heavy metals, like post-fixation with osmium tetroxide or ultrathin section staining, can by omitted. The disadvantage is reduced penetration ability of incident electrons influencing the usable thickness of the specimen resulting in the need of ultrathin sections of under 20 nm thickness. In this study we want to answer basic questions concerning the cutting of extremely ultrathin sections: Is it possible routinely and reproducibly to cut extremely thin sections of biological specimens embedded in commonly used resins with contemporary ultramicrotome techniques and under what conditions? Microsc. Res. Tech. 79:512-517, 2016. © 2016 Wiley Periodicals, Inc.

Progress in materials science is inseparably connected with development of new analytical methods which make possible to observe the materials microstructure with high sensitivity. The aim of the present study is shown that scanning low energy electron microscopy (SLEEM) has a significant impact in advance of a fundamental understanding of the evolution of microstructure upon semi-solid processing. This paper deals with the application of the ultra high vacuum scanning low energy electron microscopy (UHV SLEEM) to the study of microstructure of X210Cr12 steel after the formation in semi-solid state and the study of the annealing of deformed metastable austenite. Examples from these specimens show that the contrast between differently oriented grains in polycrystalline materials is very sensitive to the parameters such as energy of the primary beam, working distance and detection of high angle backscattered electrons.

Summary Commercially available graphene samples are examined by Raman spectroscopy and very low energy scanning transmission electron microscopy. Limited lateral resolution of Raman spectroscopy may produce a Raman spectrum corresponding to a single graphene layer even for flakes that can be identified by very low energy electron microscopy as an aggregate of smaller flakes of various thicknesses. In addition to diagnostics of graphene samples at larger dimensions, their electron transmittance can also be measured at very low energies.

Summary The advent of new imaging technologies in Scanning Electron Microscopy (SEM) using low energy (0–2 keV) electrons has brought about new ways to study materials at the nanoscale. It also brings new challenges in terms of understanding electron transport at these energies. In addition, reduction in energy has brought new contrast mechanisms producing images that are sometimes difficult to interpret. This is increasing the push for simulation tools, in particular for low impact energies of electrons. The use of Monte Carlo calculations to simulate the transport of electrons in materials has been undertaken by many authors for several decades. However, inaccuracies associated with the Monte Carlo technique start to grow as the energy is reduced. This is not simply associated with inaccuracies in the knowledge of the scattering cross-sections, but is fundamental to the Monte Carlo technique itself. This is because effects due to the wave nature of the electron and the energy band structure of the target above the vacuum energy level become important and these are properties which are difficult to handle using the Monte Carlo method. In this review we briefly describe the new techniques of scanning low energy electron microscopy and then outline the problems and challenges of trying to understand and quantify the signals that are obtained. The effects of charging and spin polarised measurement are also briefly explored. SCANNING 9999:1–17, 2016. © 2016 Wiley Periodicals, Inc.

An overview of recent developments in very low energy scanning electron microscopy is presented. Electron optical aspects are briefly summarized including the low energy beam formation in a cathode lens equipped column, comparison of the sequential and overlapped electric and magnetic fields in the objective lens, and detection issues including extension to the transmitted electron mode as well as to the multichannel detection of signal sorted according to the polar angle of emission. In addition to the acquisition of contrasts specific for the very low energy range, advantages of detection of electrons backscattered to large angles from the surface normal are demonstrated on selected application examples.

The use of the scanning low energy electron microscopy (SLEEM) has been slowly making its way into the field of materials science, hampered not by limitations in the technique but rather by relative scarcity of these instruments in research institutes and laboratories. This paper reports the results obtained from an investigation of the microstructure of ultra fine-grained (UFG) copper fabricated using equal channel angular pressing (ECAP) method, namely in the as-pressed state and after annealing. SLEEM is very sensitive to the perfection of crystal lattice and using SLEEM, local strain can be effectively imaged.

Study of the grain structure in the equal channel angular pressing processed copper by means of the cathode lens equipped ultrahigh vacuum scanning low energy electron microscope is reported. The grain contrast was found achieving its maximum at electron energies below about 30 eV where it alternated its sign and exhibited dependence on electron energy specific for the grain orientation. The energy dependence of the electron reflectance seemed to be capable of serving as a fingerprint enabling determination of the crystalline orientation. In the cathode lens mode at hundreds of eV fine details of the microstructure are also observable including twins and low angle grain boundaries. This is explained by acquisition of high-angle backscattered slow electrons, normally not acquired in standard scanning electron microscopes. The very low energy electron reflectance is promising as an alternative to the EBSD method owing to its high resolution and fast data acquisition.

By introducing a cathode lens below or inside the objective lens of a scanning electron microscope, many experiments similar to those typical of the LEEM method can be performed. The conditions for the diffraction of slow electrons are modified by the convergence of the primary beam, and challenges include the necessity of managing the signal species propagating along the optical axis in a direction opposite to that of the primary beam. However, even a simple implementation, providing the integral dark-field signal only, has not only delivered plenty of results in the very low energy range below 50 eV, but the performance in the range of hundreds of eV and units of keV has also been substantially improved. The scanning LEEM method is illustrated using experimental results acquired by additionally employing multichannel detection and detection of transmitted electrons.

In the present talk, MgAl2O4 in the Al2O3/Al-1.0mass%Mg2Si alloy composite was also observed by a scanning electron microscope equipped with the low energy electron (SLEEM) adaptation aiming at examination of its morphology and orientation relationships to the Al2O3 particles. Owing to its much smaller interaction volume of signal exciting electrons in the target and hence more localized information, together with a favorable combination of secondary (SE) and backscattered (BSE) electron signals, the SLEEM method provided much better readable and detailed images of all particles, their shapes and mutual orientations, in comparison with conventional SE and BSE images at the electron energies usually used in the SEM. MgAl2O4 (spinels) were formed on facets of Al2O3 as small particles, and their shape well corresponded to an octahedron consisting of 8 equiaxial triangles.

Low voltage electron microscopes working in transmission mode, like LVEM5 (Delong Instruments, Czech Republic) working at accelerating voltage 5 kV or scanning electron microscope working in transmission mode with accelerating voltage below 1 kV, require ultrathin sections with the thickness below 20 nm. Decreasing of the primary electron energy leads to enhancement of image contrast, which is especially useful in the case of biological samples composed of elements with low atomic numbers. As a result treatments with heavy metals, like post-fixation with osmium tetroxide or ultrathin section staining, can by omitted. The disadvantage is reduced penetration ability of incident electrons influencing the usable thickness of the specimen resulting in the need of ultrathin sections of under 20 nm thickness. In this study we want to answer basic questions concerning the cutting of extremely ultrathin sections: Is it possible routinely and reproducibly to cut extremely thin sections of biological specimens embedded in commonly used resins with contemporary ultramicrotome techniques and under what conditions? Microsc. Res. Tech. 79:512-517, 2016. © 2016 Wiley Periodicals, Inc.

A method for scanning electron microscopy imaging of nonconductive specimens, based on measurement and utilisation of a critical energy, is described in detail together with examples of its application. The critical energy, at which the total electron yield curve crosses the unit level, is estimated on the basis of measurement of the image signal development from the beginning of irradiation. This approach, concentrated onto the detected signal as the only quantity crucial for the given purpose of acquiring a noncharged micrograph, evades consequences of any changes in an irradiated specimen that influence the total electron yield curve and possibly also the critical energy value. Implementation of the automated method, realised using a cathode lens-equipped scanning electron microsope (SEM), enables one to establish a mean rate of charging over the field of view and its dependence on the electron landing energy. This dependence enables one to determine the energy of a minimum damage of the image of the given field of view. Factors influencing reliability and applicability of the method are discussed and examples of noncharged micrographs of specimens from both life and material science fields are presented.

Studies concerning assessment of the image quality in scanning electron microscopes and studies evaluating the detective efficiency of the secondary electrical (SE) detectors in these microscopes must be based on statistics of SE emission. The vast majority of previous studies have applied Poisson statistics, although their prerequisites have not been satisfied in must cases. This paper is concerned with the limits to the applicability of Poisson statistics to SE emission. Adequate definition of a non-Poisson factor in the variance of the number of SEs emitted is discussed. and a simple formula for this factor is derived for the low yield case in which both the primary and the backscattered electron are assumed not to release more than one SE. These conditions are met with conductive specimens composed of light elements at primary electron (PE) energies of tens of keV. For the lightest specimens, such as carbon, the non-Poisson factor can even be neglected for PEs >10 keV.

Although doped regions in semiconductors have been shown to give a different secondary electron yield in low-voltage scanning electron microscopy, the basic interpretation of this contrast has been difficult. It is accepted that this contrast stem from electronic phenomenon rather than atomic number differences between differently doped regions. However, the question is whether variations in the patch fields above the sample surface, balancing variations in the inner potentials, or surface coatings and/or surface states are the mechanisms responsible for the observed contrast. The present study reports on comparative experiments of these two models and demonstrates that the image contrast can be controlled by the presence of thin-surface metallic coatings. These results are the first evidence of the adlayer contacts, i.e., the subsurface electric fields instead of the patch fields above the surface, being responsible for the secondary electron contrast of doped semiconductors imaged in low voltage scanning electron microscopes under standard vacuum conditions, and they pave the way for the routine use of this method in semiconductor research and industry.

Significant differences in backscattered electron (BSE) yields exist between the surfaces cleaned by methods used in electron microscopy and spectroscopy. These differences have been observed for Au, Cu and Al specimens, and are interpreted on the basis of simulated BSE yields. Composition and thickness of the surface contamination layers, responsible for the differences, are estimated. The results (7 nm of carbon on Au or 3 nm of oxide on Al) remain within expectation and indicate that the BSE yield measurements and BSE images should be interpreted cautiously. Peculiar results are obtained for Cu, perhaps due to a different cleaning procedure. A new concept of an information depth for the BSE signal is introduced as a depth within which the total BSE yield can be modelled as composed of the yields of layers proportional to their thickness weighted by the escape depths. This concept proved satisfactory for thin surface layers and brought the information depth values 2 to 4 times smaller than first estimated, i.e. half the penetration depth.

The real point resolution of an SEM image is treated in a two-dimensional model where the decisive quantity is the root-mean-square distance of the emitted electron from the pixel centre. This quantity is computed taking into account the direct illumination of the specimen surface by the primary spot the dimensions of which are given by the electron optical column and the indirect illumination from a virtual source of the backscattered electrons in the specimen depth the properties of which depend on the specimen. Because the second order movement of the full distribution is considered here instead of some measure of the central narrow peak of secondaries, the backscattered electron influence significantly deteriorates the resulting resolution. Reasonable approximations regarding both contributions, particularly the Monte-Carlo modelling of the backscattering process by algorithms providing acceptable results down to 1 keV and the approximate relations describing the secondary and backscattered electron emission again down to 1 keV, enabled us to bring the considerations up to numerical results for some typical instruments, namely a "cheap" SEM, a "top" SEM and a low-energy SEM (LESEM) adapted from the cheap SEM by using the cathode lens. The optimum landing energy providing the best point resolution, computed for the individual chemical elements, falls into the range 1 to 5 keV for the cheap SEM and it remains around 1 keV (with some uncertainty caused by the approximations mentioned) for both the top SEM and LESEM. Similarly, the real resolution for the elements ranges between 16 and 45 nm when the cheap SEM with a 3.4 nm nominal spotsize at 30 keV is used, between 3.5 and 9 nm for the 0.7 nm top SEM and between 5.5 and 7 nm for the cheap SEM adapted to LESEM.

Summary The edge effect in electron microscopical images is proposed to be measured in length units by using a ratio of the edge or near-to-edge signal enhancement or depletion integrated over all this feature along a line-scan across an edge, to the mean signal away from the edge, giving the result in length units. This approach is applied to quantification of the edge effect in Auger electron images, which is examined by using specimens with sharp edge terminated overlayer terraces, both chemically homogeneous and heterogeneous. The following combinations of overlayer/substrate materials were used: Si/Si of three different overlayer thicknesses, W/W and W/Si, and line-scans across the edges were recorded in both the low- and the high-energy Auger electrons mode, i.e. Si LMM and Si KLL or W NOO and W MNN. The experimental results are presented for the 3-, 10- and 20-keV primary electron energies. Owing to a low signal-to-noise ratio in the measured data, basic relations between the effect appearance and the experimental conditions were revealed only: on both Si and W homogeneous specimens with a surface step, the edge enhancement is the dominating subphenomenon while at the W terrace edge on the Si substrate, the ‘penetration’ of the radiation characteristic to both areas separated by the step, to the neighbouring feature, is observed as the most significant effect. The quantification has shown that the effect is, first of all, proportional to the step height, amounting to one-third to up to the full height, while the material dependence was weak, equally to the dependence on the Auger electron energy. The primary electron energy dependence is increasing in accordance with expectation. The results indicate that the effect cannot be modelled simply by the interaction volume cut by the surface step but phenomena such as subsurface electron channelling along the sidewall have to be taken into account.

Owing to diverse possible mechanisms of the e-beam damage, some changes either in the crystalline structure, chemical composition or film thickness occur as the proper “damage” of the virgin state. Changes in the chemical bonds are often connected with some change in thickness or structure so that the other two are the most important items to be detected, preferably in real time of the electron-microscopical or electron-spectroscopical examination. We propose to measure the film thickness on the basis of the most probable energy loss due to the electron pass through the film and back, i.e. on the basis of the position of the broad background maximum in the electron spectra of stratified specimens. A change in crystallinity can be sensed through the elastic peak intensity. Placing a measurement window between these features in the energy scale and measuring the energy filtered background during the irradiation, we get a pronounced dependence of the signal on the dose which unambiguously reveals the damage limit in the form of a curve knee. In some cases, two stages of damage are detectable.

At a landing energy of 10 eV it is possible to achieve spatial resolution of the same order as at the nominal energy, which is usually 15 keV in the classical scanning electron microscope, by taking advantages of the optical properties of the cathode lens. Two different types of the detection system were designed and tested to learn as much about the optical properties of this system as possible and to start to understand the contrast mechanisms at very low energies. Great changes in the contrast take place when the landing energy is changed from 10 eV to an energy of about 2 keV.

Low energy microscopy is treated as the low energy limit of electron microscopy as a whole in all its basic branches, i.e. the emission, transmission and scanning microscopies. The instrumental and methodological aspects are briefly discussed. These include the interaction of electrons with a solid, the contrast formation mechanisms, the instrumentation problems and actual progress achieved in all three microscopies, from the point of view of lowering the energy of electrons, impacting or leaving the specimen, down to the low energy range below 5 keV and the very low energy range below 50 eV.

High resolution can be obtained in the scanning electron microscope (SEM) for a very low landing energy of electrons, even for that below 50 eV, when a cathode lens is used with the specimen as a cathode held at a high negative potential but the detection of signal electrons is totally different compared with classical SEM. Primary electrons with an energy of the order of tens of keV are decelerated in the field of the cathode lens to a very low landing energy and signal electrons originating in the specimen are accelerated and collimated by the same field to a narrow beam with an electron energy nearly the same as that of the primary beam. To detect these signal electrons we must deflect them from the optical axis without deteriorating the properties of the primary beam. The design of a novel type of separator of the primary and signal electrons consisting of two stages, each of them formed by the electric and magnetic crossed fields, is presented, together with calculated trajectories for both primary and signal electrons.

The aim of Auger electron imaging is to obtain quantitative surface elemental distribution maps at high spatial resolution. The realization of this goal is complicated by many instrumental effects and by spurious data processing contributions giving rise to an image contrast unrelated to the specimen surface composition. The critical properties of scanning Auger microscopy that may cause such a false information or imaging artefacts are reviewed. Instrumental or data processing related effects appear in the case of the beam current variation, of the background slope effect, and of the use of a combined peak to background ratio. The second set of artefacts are mainly due to the significant differences between the penetration depth of the exciting primary electrons and the escape depth of the Auger electron signal. In this case the net effect is a surface elemental contrast which is dominated by the substrate or by the overlayer rather than by the surface under investigation. In addition, there are also topographical effects of the specimen under test which normally affect the Auger yield and hence the contrast in the image. Methods for the successful suppression of some of these artefacts are outlined. They are based on the creation of reference images from complementary signals acquired by additional detection channels in parallel with the Auger signal of interest.

We demonstrate the possibility to examine the free-standing foils of thicknesses in units of nm in the scanning low energy electron microscope, using both reflected and transmitted electrons. Very high contrast has been obtained in dependence on the thickness and structure of the foil. A contribution of secondary electrons to the forward scattered electron signal is discussed and a way of suppressing it is presented. Examples of reflected, total transmitted and dark field transmitted electron signal for two graphene-like samples are shown. Dependence of the transmitted signal on the electron energy is observed.

The topographic contrast in a scanning electron microscope with very slow electrons is examined in case the cathode lens is used for adjusting the landing energy. Measurements are presented regarding the behaviour of planar contrasts at low energies and a significant improvement of the surface detail visualization is demonstrated, together with the possibility of controlling the electron impact angle.

Significant contrast differences between linescans in an energy-filtered total electron emission taken at several tens of electron-volts and at several hundreds of electron-volts show up remarkably small differences in surface chemical composition, i.e. local differences in the work function on a partially cleaned surface, so that comparison of such scans can be used as a test for the cleanliness of a rough surface. The comparison can be quantified by a single numerical quantity which allows one to monitor the cleaning process.

Minimum necessary adaptations are described that are sufficient for obtaining very low energy electron micrographs (VLEEMs) from commercially available routine scanning electron micrographs (SEMs) with the electrons accelerated to an energy of the order of tens of keV. A cathode lens inserted into the specimen chamber enables one to decelerate electrons in front of the specimen surface to a desired low landing energy, which can be freely varied even down to zero. When a potential slightly more negative than the accelerating voltage is applied, a scanning mirror electron microscopy mode can be effected. The achievable point resolution at very low energies proves to be not too dependent on the objective lens parameters, so that the physical limit of aberrations of the homogeneous field of the cathode lens is nearly attainable. The detection efficiency for the standard Everhart-Thornley secondary electron detector is discussed, and results for the routine Tesla BS 340 SEM are presented.

We have developed a Scanning Low Energy Electron Microscope (SLEEM) based on the Cathode Lens (CL) principle [1]. A resolution of 4.5 nm at 20 eV, 0.8 nm at 200 eV and 0.5 nm at 15 keV primary beam energy can nowadays be obtained in a commercially available instrument [2]. One of the main advantages of operation at low energies is the decrease in the interaction volume from approximately 1 μιη at 10 keV to 10 nm at 100 eV. The material contrast can be optimised and the charging effect suppressed at a tailored electron energy. Wave-optical contrasts are also available beneath 50 eV. The specimen may be immersed in a strong magnetic field in addition to an electrostatic field in order to obtain a small spot size across the whole energy range. The same fields influence the signal trajectories, so we can choose which part of the angular and energy distributions of emitted electrons are to be collected. Certain arrangements provide strong crystallographic contrast. Imaging conditions have been tailored to various material types. Experiments have been performed in an experimental ultrahigh vacuum (UHV) SLEEM and in an XHR SEM Magellan 400L.

The local density of electron states is important characteristics of solids, which can be utilized in materials science when distinguishing between different crystal orientations in a polycrystalline matter, mapping the local density of dopants in semiconductors, etc. Non-imaging application of the reflection of slow electrons (VLEED — very low energy electron diffraction) knows so called energy band structure region on the intensity vs. energy curve for the specularly reflected (00) spot, which appears below the threshold where the first nonspecular diffracted beam appears [1]. The incoming electron wave impinging on the sample surface has to “convert” into electron waves of the crystal periodicity, i.e. into Bloch states. This condition makes penetration of incident electrons proportional to the local density of states coupled to the incident wave. Electron reflection becomes then inversely proportional to the local density of states and the reflected flux can measure the density. Crucial condition is sufficiently low absorption of injected hot electrons, which is available below about 20 to 30 eV above the Fermi level.

Some problems of scanning Auger electron microscopy (SAEM) concerning instrumentation and experimental techniques are described and briefly discussed in the light of the fundamental limitations of the method set by the nature of the phenomena themselves.

Differently doped areas in silicon can show strong electron-optical contrast in dependence on the dopant concentration and surface conditions. Photoemission electron microscopy is a powerful surface-sensitive technique suitable for fast imaging of doping-induced contrast in semiconductors. We report on the observation of Si (100) samples with n- and p-type doped patterns (with the dopant concentration varied from 10(16) to 10(19) cm(-3)) on a p- and n-type substrate (doped to 10(15) cm(-3)), respectively. A high-pass energy filter of the entire image enabled us to obtain spectroscopic information, i.e. quantified photo threshold and related photoyield differences depending on the doping level. Measurements have confirmed the possibility of resolving areas at a high contrast even with the lowest dopant concentration when employing the energy filter. The influence of electron absorption phenomena on contrast formation is discussed.

One of the well-proven and efficient methods of obtaining a very low-energy impact of primary electrons in the scanning electron microscope is to introduce a retarding field element below the pole piece of the objective lens (OL). It is advantageous to use the specimen alone as the negatively biased electrode (i.e., cathode of the cathode lens). The optical power of the cathode lens modifies some of the standard parameters of the image formation such as relation of working distance to OL excitation or magnification to the scanning coils current, the impact angle of primary electrons, and so forth. In computer-controlled electron microscopes these parameters, particularly with regard to focusing and magnification, can be corrected automatically. Derivation of algorithms for such corrections and their experimental verifications are presented in this paper. Furthermore, a more accurate analytical expression for the focal length of an aperture lens is derived.

Extended abstract of a paper presented at Microscopy and Microanalysis 2013 in Indianapolis, Indiana, USA, August 4 – August 8, 2013.

Extended abstract of a paper presented at Microscopy and Microanalysis 2013 in Indianapolis, Indiana, USA, August 4 – August 8, 2013.

Extended abstract of a paper presented at Microscopy and Microanalysis 2012 in Phoenix, Arizona, USA, July 29 – August 2, 2012.

Scanning electron imaging of plan views of boron-doped patterns in silicon is examined, together with the mechanism of formation of the electronic contrast in this kind of structures. Main to-date published results are critically reviewed and new data are presented concerning the secondary, backscattered and total-emission electron contrasts, including their qualitative and quantitative behaviour, particularly in the low energy range achieved with the help of the cathode lens (the scanning low energy electron microscopy mode, SLEEM). Surface analysis of the structure by means of Auger electron spectrometer has been performed, too, both before and after ion beam bombardment. The scanning electron microscope micrographs, acquired after the oxide mask removal in HF, are examined in a variety of detection modes, aiming at identification of the signal component primarily bearing the contrast. The energy dependence of the contrasts is presented as well as its change owing to alteration in the vacuum conditions. The most important findings include an extremely high contrast obtained in the SLEEM mode and even more enhanced under medium vacuum conditions at which the carbonaceous layer of surface contamination plays its role. The observed phenomena are partly explained in the frame of the “flat band” model of a passivated surface. The increased contrast in the SLEEM mode is understood as connected with the above-surface electric field of the cathode lens, generating space charge layers inside the semiconductor. In addition, charge carriers, injected via the primary electron beam, are considered as influencing the contrast vs. energy dependence.

Publisher Summary This chapter discusses the advances in scanning electron microscopy (SEM). Alternatives to the previously characterized main body, physical part, or column of the SEM have progressively been introduced, including (1) scanning columns for electron spectrometers, such as those suitable to be inserted inside a cylindrical mirror energy analyzer, (2) testers and lithographs for semiconductor technology, and (3) miniaturized versions. The issues regarding further development of the classical general-purpose SEM are discussed in the chapter. Computer-aided design methods for calculation of electrostatic and magnetic elements and simulation of electron trajectories have enabled significant progress in tailoring the column design to prescribed electron optical parameters. Complete computer control of the device opens approaches to full utilization and easy adjustment of all possible operation modes. These two aspects are most important with regard to the recent development outlined in the chapter, and some new ideas have proved viable, particularly that of variable beam energy along the column. In SEM, new technologies—such as rare-earth based permanent magnets and particularly the family of various micro- and nanotechnologies that have projected themselves into the SEM instrumentation—are used.

Extended abstract of a paper presented at Microscopy and Microanalysis 2011 in Nashville, Tennessee, USA, August 7–August 11, 2011.

A novel principle of an in-lens detector of very slow electrons is described and the detector efficiency discussed. The detector was built into a coaxial column for a Cylindrical Mirror Analyser for Auger electron microanalysis. In order to obtain a very low energy scanned imaging, a cathode lens was formed between the final electrode of the column and a negatively biased specimen. The signal electrons accelerated within the cathode lens field enter the column and after being mirrored back impact a micro-channel-plate based detector fitted around the optical axis. The acceptance of the detector, expressed as a ratio of the number of electrons impacting the detector to the full emission of a cosine source, was calculated to be 0.86 for 1 eV and 0.985 for 10 eV electrons. Then, the efficiency of conversion into output pulses is 0.35 and 0.31, respectively; these parameters are superior to those of conventional SEM detectors for secondary electrons. Micrographs taken at low energies ranging down to units of eV are presented.

The edge effect, which manifests itself especially in an edge over-brightening in images formed by shallow escape depth signals, is addressed. Attempts had been made to utilize the ratio technique which was developed for suppression of other spurious contrasts in Auger micrographs; it consists in relating the Auger signal to a suitable energy filtered background signal. A systematic comparison of a quantified edge effect in both the Auger images and the energy selected background images is reported here. Surface steps on Si and W surfaces and a W terrace edge on a Si substrate were used at 3 to 20keV primary electron energy, both low and high energy Auger electrons of Si and W were registered together with a background of up to 3 keV. Results have shown that the edge effect is less pronounced with the background so that any ratioing leads to an undercorrection. The best suppression (nearly acceptable for low energies, both primary and Auger ones) is obtained when using the background between 50 and 200 eV as the reference signal.

If any not absolutely smooth and homogeneous nonconductive surface is observed in SEM, a locally variable charging appears which produces local electrical fields that destroy the image. The surface charge depends on the total electron yield σ, i.e. the number of outgoing electrons divided by the number of the incoming ones, on the primary current, resistance between the field of view and the earth, surrounding gas pressure and time. A charge balance is established between primary and emitted electron currents, the specimen current taking charges off the surface, and the charged particle bombardment from the atmosphere, which can be neglected in high vacuum, σ changes with the landing energy of primary electrons: near to zero energy, σ < 1, and the surfaces charge negatively. For nearly all materials, between two characteristic energy values E I and E II , σ exceeds unity and the charge is positive, while above E II , σ < 1 again and a negative charge appears. E I can be found usually between 80 and 300 eV and E II amounts to one or a few keV and depends on the electron impact angle. A novel method is proposed consisting in approaching one of the critical landing energies E I or E II where σ → 1 and the surface charge formation is suppressed to a level corresponding to the intensity of spontaneous discharging. The retarding field principle of low energy SEM realization, which minimizes the point resolution changes with the landing energy, makes both E I and E II equally well accessible. In order to find the critical energy without getting the specimen intensively charged, we recommend to measure, quickly enough from the point of view of the charging process duration, the time dependence of the signal in a single pixel by stepwise approaching the landing energy of the slowest change, i.e. the acceptably approximated critical energy. A model of the image signal vs. time development is presented, together with a description of algorithms and software for the critical energy measurement. The preliminary results presented confirm our basic ideas regarding the method.