Electron microscopy has increasingly found more applications in recent years. Every sample has a mixture of optimized settings that must be utilized to enhance the results of the investigation.
This article will outline all of the key aspects that need to be taken into account when imaging samples and will outline some of the mathematics and physics behind them.
Magnifying glasses were first found in Ancient Greece, where Aristophanes outlined the first attempt to view closer details as a children’s leisure activity. This was when the term ‘magnification’ first entered human language.
The interest in science for the nano and micro world has greatly increased as time has passed, opening up the requirement for a quantification of magnification.
Magnification is defined in modern times as the ratio between two measurements, which suggests that two objects are required for an accurate assessment of the value.
The sample is obviously the first object and the second is an image of it. While the sample will not vary in size, the image can be printed in an infinite number of various sizes.
This means that printing an image of an apple that is the correct size for a regular printer sheet and printing it again to fit on a poster that will be utilized to cover a building, will significantly adjust the magnification value (much bigger in the second example).
An example that is more scientific can be found in the context of microscopy: when storing a digital image of the sample, changing the size of the image results in the magnification number becoming clearly incorrect.
Magnification is a relative value which means it is not helpful in the field of science.
Scientists utilize two parameters that outline the specific imaged area. These are the field of view which is the region that the microscope points at and how sharp the image is, known as the resolution. The formula of magnification also varies accordingly:
The formula still clearly provides a brief description and does not factor in the resolution. This means that the magnification number will change when scaling the same image to a larger screen.
Image Credit: Thermo Fisher Scientific Phenom-World BV
The size of the object to be imaged is defined by the field of view. This value normally varies between several millimeters (a bug) to several microns (the hair of a bug) and a couple of nanometers (the exoskeleton’s molecular macrostructure).
Objects in the range of several hundred picometers can be imaged using modern tools which is the standard size of an atom. How does one define the necessary field of view to image a sample? It depends.
As an example, if the particles have an average size of 1 micron and the application requires them to be counted, it is sufficient to have 20 particles per image, instead of wasting time by imaging one particle at a time.
Even when the empty space between particles is taken into account, a field of view of 25 to 30 microns is adequate for such a sample. If the particle structure is the focus of the investigation, a close up is necessary and the recorded area must be nearer to 2 to 3 microns, if not less.
Images of particles. A close-up of a particle (left) shows the surface topography (FOV=92.7 μm). A larger field of view (right) enables more particles to be imaged (FOV=μm). Image Credit: Thermo Fisher Scientific Phenom-World BV
Resolution is defined as the least possible distance between two objects that still enables the observer to identify them as individual entities in microscopy.
This is where microscopes are involved. Microscopes enable exceptional resolutions to be attained, in some examples helping the user to determine even atoms.
Due to their outstanding resolution, desktop SEMs, especially when contrasted to typical optical microscopes, are incredibly powerful tools to investigate small features.
Desktop SEMs are slowly transforming the industry by changing production standards to a new degree of miniaturization. They have an average resolution that is easily less than 10 nm, and a price range equivalent to that of a high-end optical device.
It is important to consider that a microscope’s resolution is not the size of the smallest attribute that can be imaged. This means that employing a device with a resolution of 10 nm to image and analyze samples with an average size of 50 to 100 nm offers positive results.
More intricate features will appear blurry and will require the use of a much more advanced device to be imaged. In simpler terms, the device’s resolution should be five to 10 times lower than the size of the imaged feature.
Scanning electron microscope images are kept in an image file (such as JPEG or TIFF) with a user-specified number of pixels for the resolution.
An SEM will scan tiny regions with an electron beam, meaning that the segments of the surface will become a pixel of the resulting image. Additional pixels create a more lengthy processing time, and samples can be impacted by a longer process of analysis.
When spots are far enough apart to be distinguished, they are resolved. If they are too close, the edges will seem to overlap, and the objects will be unresolved. Image Credit: Thermo Fisher Scientific Phenom-World BV
1.3 Electron Source
The electron source, also known as a filament, electron gun, or cathode, is one of the most critical components of a desktop SEM.
Its function is to offer a consistent beam of electrons. There are two groups of electron sources utilized in SEM, which vary in the amount of current that they generate into a small beam size (spot), the lifetime of the source, and the stability of the beam.
This article will concentrate on a kind of electron source that is being employed in desktop SEM: the thermionic electron source. More specifically, the differences of two kinds of thermionic electron sources are focused on: tungsten and Cerium Hexaboride (CeB6).
What is a Thermionic Electron Source?
Electrons will be generated through thermionic emission when any solid material is heated. The emission becomes greater when the electrons’ thermal energy is enough to exceed the material’s work function.
The cathode is created from a material with a high melting point and a comparatively low work function as a means to generate a large amount of electrons.
The electron beam being projected onto the sample is made by the emitted electrons that are accelerated from the source’s high negative potential to ground potential at the anode within the electron column.
This phenomenon can only occur within the vacuum of an electron column and by utilizing lenses to direct the beam.
Prior to investigating and contrasting the tungsten and CeB6 source, it is beneficial to understand which features of an electron source are critical when evaluating its performance. The most essential features will now be described:
1. Brightness of the Electron Source
The definition of brightness is the beam current per unit area per solid angle. The higher the availability of current or electrons in a small spot area, the more effectively high-resolution (quality) images at high magnification can be achieved.
The brightness enhances linearly with the acceleration voltage. For example, each electron source is ten times as bright at 10 kV as it is at 1 kV.
The electron beam’s spot size can be made smaller to enhance the resolution, but at a certain stage, the weakness is the signal-to-noise-ratio needed to take a high-quality image.
2. Source Size
As previously mentioned, a small spot size provides a better image resolution and high-quality images as a result of this.
The lenses (mostly enabled by the condenser lens) within the electron column are in charge of demagnifying the beam diameter (or spot size). A source with a smaller physical size results in less (complicated) demagnification.
3. Source Temperature
The source temperature is the operational temperature, which combats the work function as a means to generate electrons. For thermionic sources, the operational temperature is between 1800 and 2800 Kelvin.
- Electron Beam Energy Spread
The electron beam energy spread is the spread in electron energies exiting the source.
Chromatic aberration is the leading aberration at low acceleration voltage when the source’s energy spread is significant. Chromatic aberration is a factor that results in a less focused beam because of electrons leaving the source with a slightly different energy.
Lifetime indicates the lifespan of an electron source until it is damaged or requires replacement.
A durable source is preferred and it is beneficial if the moment of replacement can be predicted. The tungsten and CeB6 comparison can now be outlined using the most crucial elements of an electron source as reference.
Cross-section view of an electron column with a schematic view of the source assembly. Image Credit: Thermo Fisher Scientific Phenom-World BV
Tungsten filament. Image Credit: Thermo Fisher Scientific Phenom-World BV
CeB6 filament. Image Credit: Thermo Fisher Scientific Phenom-World BV
Image from TiO2 powder made with CeB6 system. Image Credit: Thermo Fisher Scientific Phenom-World BV
Image from TiO2 powder made with tungsten system. Image Credit: Thermo Fisher Scientific Phenom-World BV
Tungsten filaments are commonly utilized in scanning electron microscopy.
Of every metal in their purest form, tungsten has the lowest vapor pressure, the highest melting point, a very high tensile strength and the least thermal expansion, which are ideal characteristics for creating an electron source.
As can be seen in the comparison, tungsten has some major weaknesses when compared to a Cerium Hexaboride (CeB6) electron source:
1. Brightness of the Electron Source
When brightness is compared, the tungsten source offers 106 A/cm² sr. The smaller work function of a CeB6 filament creates higher beam currents at smaller cathode temperatures than tungsten, which results in stronger brightness at all acceleration voltages.
To make this simpler: a CeB6 cathode offers ten times the brightness than tungsten at 107 A/cm² sr. This gives the CeB6 source two benefits over a tungsten source:
A better signal-to-noise ratio at the same spot size is offered as there is more current available in the same focused spot.
A better resolution can be attained because at the same signal-to-noise ratio, the CeB6 spot can be made smaller.
2. Source Size
The source size of tungsten is elliptically shaped with a dimension varying from 50 μm to 100 μm, reliant on the source operating conditions and configurations.
In comparison to a CeB6 source, which has a dimension of less than 25 μm, extensive electron optic demagnification is needed for a tungsten source to attain a small electron probe required for high resolution in SEM.
3. Electron Source Temperature
The tungsten filament’s operational temperature is around 2800 Kelvin, whereas the CeB6 source has an operational temperature of 1800 Kelvin. The variation in temperature has a direct influence on the source.
- Electron Beam Energy Spread
The tungsten source has a larger energy spread than the CeB6 source due to the higher temperature setting.
The energy spread of a tungsten source is normally around 2.5 eV, where the CeB6 is around 1 eV, which creates a higher image quality, particularly at lower acceleration voltages.
- Electron Source Lifetime
A tungsten filament functions at temperatures that are white-hot, meaning that it eventually evaporates over time. The tungsten wire gradually becomes thinner and damages which constantly occurs during imaging.
The upper section of the electron column can potentially be contaminated when the tungsten wire breaks. When replacing the tungsten filament, this is why it is advisable to clean or replace alternative source-related elements within the column as well.
The benefit of a CeB6 source is that its lifetime ending can be predicted because it gradually degrades over time. It is easy to know when it is time to replace the CeB6 filament and this can be done in between operating times.
A situation where the investigation must be cancelled due to a broken filament will not occur and, more importantly, contamination of the column as a result of debris will not be an issue.
Utilizing a CeB6 source also reduces the requirement to replace different parts related to the source along with the source itself.
When the lifetime for tungsten and CeB6 are compared, the typical lifetime of a tungsten source is around 100 hours, based on the vacuum. A CeB6 source normally offers more than fifteen times the service life, at more than 1,500 hours.
1.4 Acceleration Voltage
The voltage is a signifier of the energy content of the electrons: this will establish the type of interaction that the beam will have with the sample.
As a general rule of thumb, a high voltage is related to a greater penetration beneath the sample’s surface, also called a bigger interaction volume.
The electrons will have a bigger and deeper propagation inside of the sample as a result, and will produce signals in various parts of the affected volume.
The sample’s chemical composition also influences the size of the interaction volume: light elements have less shells, and the energy content of the electron is lower.
This restricts the interactions from the electron beam with the electrons, which can penetrate deeper into the sample as a result, in comparison to an element that is heavier.
When investigating the outcoming signals, various results can be gathered. In desktop instruments, three types of signal are usually identified: secondary electrons (SE), X-rays, and backscattered electrons (BSE).
The Effect of Voltage in SEM Imaging
The influence of voltage within the SE and BSE imaging is comparable: low voltages allow the sample’s surface to be imaged; whereas higher voltages offer further information on the layer underneath the surface.
This is presented in the images below, where low voltages make contamination of the surface of the sample clearly apparent, while higher tensions show the structure of the surface below the layer of contamination.
BSE images of tin balls at 5 kV (left) and at 15 kV (right). With the lower voltage, the carbon contamination on top of the sample becomes visible. When the voltage is increased, the deeper penetration enables the imaging of the tin ball surface underneath the carbon spots. Image Credit: Thermo Fisher Scientific Phenom-World BV
The sample’s nature is also greatly important in the selection of the correct voltage. Many polymers, biological samples, and a host of other (mainly organic) samples are very sensitive to the electron’s high energy content. This sensitivity is further increased by the fact that the SEM functions in a vacuum.
This is the main reason why SEM developers are focused on maximizing the resolution value at lower voltages, offering critical results even with the most intricate of samples.
The key challenge that occurs in this system is the physics principle underlying the imaging method. Similar to photography, there are actually multiple types of aberration and distortion that can influence the quality of the final output.
The chromatic aberrations become less important with higher voltages, which is the key factor that led to the previous trend in SEM that promoted the greatest possible voltage as a means to enhance imaging resolution.
The Generation of X-Rays
The process is entirely different regarding X-ray generation: a higher production of X-rays is the result of a higher voltage.
The X-rays can be acquired and analyzed utilizing an EDS (energy dispersive spectroscopy) detector to carry out compositional analysis on the sample.
The method entails forcing the ejection of an electron in the target sample using the interaction from the electron beam with the electrons (primary electrons).
A charge vacancy (hole) can be produced in an atom’s inner shells. An electron fills this with a greater energy content from an outer shell in the same atom. This system needs the electron to release some of its energy in the form of an X-ray.
The X-ray’s energy can lastly be correlated to the atom’s atomic weight using Moseley’s law which then determines the sample’s composition.
Key Factors in X-Ray Production
- Overvoltage: The ratio between the energy required to ionize the targeted atom and the energy of the incoming beam.
- Interaction volume: How the spatial resolution of the investigation is defined.
The perfect analysis needs a minimum overvoltage value of 1.5, which means that the maximum amount of detectable elements increases as the voltage increases.
A high voltage also means there is a higher chance of sample damage and the interaction volume is larger. This means that the reliability of the sample could be at risk, and the production of X-rays exhibits a much greater volume.
In the example of particles, multilayers, and non-isotropic materials in general, a bigger interaction volume will produce signals coming from parts of the sample with an alternate composition, harming the quality of the results.
The standard tension values recommended for the analysis vary between 10 and 20 kV to counter both effects. Selecting the perfect value relies on a further aspect of EDS analysis that is referred to as ‘peak overlap’.
X-rays produced by electrons traveling from various shells of different elements can have comparable contents of energy.
This needs more refined integration methods to normalize the results and deconvolute the peaks, or the higher energy content lines should be used (coming from one of the two elements with overlapping peaks).
While the majority of EDS software already uses the former method, the latter may not always be possible because the higher energy level line for a popular element like lead would need a voltage greater than 100 kV.
This example shows an EDS spectrum collected at 15 kV. The peaks highlight the presence of an element, and a complex algorithm is applied to convert the signal coming from the detector into chemical composition. Image Credit: Thermo Fisher Scientific Phenom-World BV
1.5 Current Intensity
The user can manage the size of the electron probe in any contemporary scanning electron microscope. This is mostly obtained by changing the objective lenses of the system and the condenser, and by choosing alternative apertures.
Electrons are moving through electromagnetic lenses (which essentially comprise wire coils within metal pole pieces) and the user can direct the path of the electron by tuning the current that is applied to the lenses.
The spot size relies on the acceleration voltage (as high accelerating voltages reduce the spot size), the working distance (the greater it is, the bigger the spot size becomes), and the objective lens aperture (smaller apertures produce spots of smaller diameter).
The size of the resulting electron probe is a parameter that is much more complicated to manage and predict because it relies on multiple factors that are interconnected.
The relation that outlines the spot size has terms that depend on the gun’s Gaussian diameter, the final aperture’s diffraction effect, chromatic aberration, and spherical aberrations of the beam-forming lens.
It may seem too basic that to have sufficient current on the sample and a small probe, the user just needs to make the convergence angle of the probe larger.
This will maximize the aberrations of the optical elements in the microscope and will widen the beam.
It is apparent that to carry out an analysis with precision, it is essential to know how various parameters determine the features of the electron beam and distinguish the trade-offs between them.
High-Resolution Imaging Versus High-Beam Current
The key factor that determines resolution is the spot size. To take an image with a high-resolution, the spot size must be kept as small as possible to be able to describe and resolve the smallest characteristics of the sample adequately.
It is also critical that the beam holds enough beam current for adequate contrast resolution and signal-to-noise (S/N) ratio.
Since decreasing the spot size also reduces the beam current, users must determine and choose the configurations that will best meet their requirements. If high-magnification images are required, the spot size should be kept as small as possible.
If the user only needs imaging of low magnification, then maximizing the spot size is suggested, so that the images contain more ‘electron juice’ and appear sharper.
The next image shows that images were taken at low magnification but with a larger spot size appear to be smoother and brighter.
The user should change to the smaller spot size as the magnification increases, which offers optimal results when high-resolution imaging is needed.
Wider spot sizes, and higher beam currents as a result increase harm to the sample. This is an element that should be considered, particularly when samples that are beam-sensitive are being imaged.
The four major parameters of the electron beam in a SEM: Accelerating voltage, convergence angle, beam current, and spot size. Image Credit: Thermo Fisher Scientific Phenom-World BV
These SEM images are of tin. At low magnification, a high beam current (a) is preferred. At high magnification, using a smaller spot size (b) allows the user to achieve better spatial resolution. Image Credit: Thermo Fisher Scientific Phenom-World BV
Scanning electron microscopes can be supplied with several different accessories or detectors to carry out various types of analysis or to image non-perfect samples. For example:
- Freezing the sample allows the user to analyze samples that have significant moisture content.
- An EDS detector can offer the chemical composition of a sample.
- Tensile testing gives information on the sample’s behavior when stressed with a large load.
- Motorized tilting systems offer the ability to move the sample while it is in the vacuum.
Countless examples can be cited and various technologies prove to be more beneficial with particular types of samples.
Image Credit: Thermo Fisher Scientific Phenom-World BV
1.7 User Experience and Time to Image
Electron microscopy is now more economically accessible, and the user experience with these devices has been refined so that anyone can operate them.
Scanning electron microscopes are made from intricate electronic components, which means they are very prone to contamination from either exposure or polluted environments.
Decreasing the time that the internal system is exposed to external environments is beneficial. For example, an electron source that lasts longer will help to keep the device in optimal conditions as it does not constantly need the system to be opened to replace components.
Specifying the height of the sample is also a crucially important action when performing SEM analysis: if the sample is too small, the signal will not be sufficiently strong, meaning the image quality and resolution will be reduced.
If the sample is loaded too high in contrast, there is a chance of impacting the detectors. Smart loading systems have been produced to protect the devices from damage and makes it simple to position the sample at the best possible working distance.
The positioning of the columns used to be an essential condition for accurate imaging. Electron columns can now be aligned automatically to save user’s time.
The ease of use will define how much time is required to gather the intended results. A system which is simple to use and has an efficient loading time can deliver results between 1 to 2 minutes. This saves time so that the user can concentrate on more critical tasks.
A scanning electron microscope is a promising tool with endless applications.
It is highly important that the user has a specific idea of what kind of analysis is necessary. It is also important to know how the various beam currents, accelerating voltages, and spot sizes will impact the imaging quality of the SEM. It is fundamental to choose the optimal parameters for any experiment at hand.
This information has been sourced, reviewed and adapted from materials provided by Thermo Fisher Scientific Phenom-World BV.
For more information on this source, please visit Thermo Fisher Scientific Phenom-World BV.