Electron microscopy (EM) is a fantastic tool that enables biologists to capture images of their samples at a greater resolution than with a light microscope. There are several types of EM and each of these can provide different information about your sample. The large field of EM is expanding all the time and there are many advanced instruments that I do not describe below. In this article I will discuss the basics, the main types of EM, what information you can obtain and the challenges faced by biologists.
Magnification and resolution
Often the term magnification is used when discussing the power of a microscope. However, magnification is not the main issue affecting microscopes. It is the resolution. Resolution is the ability to distinguish two objects as separate. Imagine a car coming towards you at night. Initially you would see a single headlight and at some point you would be able to separate the light into two distinct headlights. This is the minimum resolvable distance or resolution. In microscopy we refer to the Airy disc, a diffraction pattern created by imaging a point object. The resolution of two objects in a microscope depends on them being sufficiently separate so that the diffraction patterns do not merge, known as the Rayleigh criterion.
Ernst Abbe (1873) was able to determine the optimum resolution that a microscope can achieve, and the equation for this is the Abbe diffraction limit. Microscopes are designed to minimize variables so that the main limiting factor is the wavelength used to image the sample. All electromagnetic radiation (e.g., light, x-rays, radio waves, etc.) has a set wavelength. Using a form of electromagnetic ration with a smaller wavelength increases the resolution that can be achieved. Light microscopes can achieve a resolution of 200 nm. Super-resolution light microscopy allows some biologists to go beyond this limit and a few X-ray microscopes are now also available. However, EM remains the main technique used by biologists for high resolution imaging of molecules, viruses and cells.
Electrons are negatively charged sub-atomic particles that have a wavelength. The wavelength of an electron is determined by the de Broglie wavelength and is linked to the accelerating voltage (AV) used to form the electron beam.
For example, an AV of 100,000 volts results in an electron wavelength of 0.0037 nm and a resolution that is up to 100,000 times smaller than can be achieved with light.
While it is possible to achieve sub-atomic imaging with an electron microscope, it is not possible to image biological samples at this resolution. As we will see, the biological tissue is always the limiting factor!
There are several different types of EM. These can be split into two main categories, transmission electron microscopes (TEM) and scanning electron microscopes (SEM). The main differences between these are in the optics (Fig. 1), how the signal is detected and the type of information you can obtain.
Both types of EM have an electron gun, which contains an electron source (a filament that produces a cloud of electrons), a Wehnelt cylinder (to form the beam) and an anode (to accelerate the beam). There are three main types of electron source; a tungsten filament, a lanthanum hexaboride (LaB6) crystal and a field emission filament. Differences among the filaments are shown in table 1.
|Energy range (eV)|
Both TEM and SEM use electromagnetic lenses to focus the beam of electrons. Electrons travel along the magnetic field and can be focused in the same way that light is focused using glass lenses. Apertures are associated with the lenses and are thin plates of molybdenum with several small bores (usually a range of 10-300 µm in diameter). Apertures are used in an EM to control the coherence of the beam, which affects resolution, and the amount of contrast in the signal.
Transmission electron microscopy
A TEM transmits the beam of electrons through a thin sample onto a screen or a camera/detector (Fig. 1). It has a large number of lenses. The condenser lenses (2-4 depending on the microscope) are responsible for the amount of illumination that reaches the sample and control beam intensity or brightness. The objective lens focuses the beam of electrons onto the sample and applies a small amount of magnification. The intermediate and projector lenses magnify the beam and project it onto the camera (CCD or film) or screen to form an image.
It takes only a few seconds to obtain a micrograph (microscope image). The image is a result of the projected beam intensity: Transmitted electrons are detected as light areas in the micrograph; darker areas occur where electrons have been scattered or absorbed by the sample, thus reducing the number of electrons reaching the camera or screen. This is known as bright field imaging and is the most common type of imaging for biological samples.
TEMs are often classified based on the accelerating voltage (AV) they are capable of. A routine TEM for biological imaging should be capable of an AV of up to 120 kV. Most thin-section TEM will be conducted using 80-100 kV. Advanced TEM techniques may require instruments capable of an AV between 200 kV and 3 MV, which represent a resolution 100,000 to 3 million times smaller than light microscope resolution.
Scanning electron microscopy
A SEM focuses the beam of electrons into a small spot that scans across the surface of a sample (Fig. 1). The condenser lens assembles the electrons into a fine beam. The objective lens focuses the beam onto the sample. Deflection coils cause the beam to move in a rectangular X and Y direction, producing a raster scan across the surface of the sample. The signal is transmitted to a computer screen. Reducing the area being scanned results in an increase in magnification).
An SEM image is formed from signals that are emitted from the sample as a result of the specimen-beam interaction (Fig. 2). Most biological SEM will generate images using two types of electrons. Secondary electrons (SE) are low energy electrons produced by small energy transfers between electrons from the beam and electrons orbiting atoms in the sample. The energy transfer causes the orbiting electron to leave the atom and become a secondary electron. An outer orbiting electron will then release some energy in order to jump into the gap left by the secondary electron. The second type, backscattered electrons (BSE), are high energy electrons that have passed close to an atomic nucleus and been reflected or “back-scattered” out of the specimen. In addition, there are a few applications that require the detection of characteristic X-rays (energy dispersive X-ray spectroscopy) or photons (cathodeluminescence). There are different types of detectors to collect these signals. Secondary electrons are low energy electrons and only those produced near the surface can be emitted (Fig. 3).
The signal detectors are not cameras and the resolution of an SEM image depends on the spot size of the beam as it hits the sample and the interaction volume between the beam and specimen. The interaction volume directly relates to the AV of the beam. Biological SEM typically uses an AV of 1-5 kV for the best resolution.
The SEM image is inverted compared to the TEM. Bright areas of the image are the result of more electrons being scattered (from topography or heavy element staining). Relatively large biological samples can be imaged using an SEM as we no longer have to transmit the signal through the specimen.
Challenges with biological EM
There are several challenges for researchers working with biological samples in EM:
The electron beam operates within a fairly high vacuum. This causes problems for biological tissue, as evaporating water destroys structures being imaged. This also means that it is not possible to image living tissue with EM. Several protocols stabilise the sample during preparation. Samples can be frozen rapidly and imaged in their hydrated state in a technique called cryo-TEM, or they can be fixed with chemicals, dehydrated in solvents, stained, embedded in a resin and sectioned (Hayat, 2000).
Lack of contrast
Biological tissue does not diffract many electrons. Electron-dense stains or advanced EM techniques are often necessary to visualise biological ultrastructure.
The electron beam must be able to get through the sample in a TEM. Some samples are small enough that they can be imaged whole, such as viruses, but most cells and tissues need to be sectioned into 50-200 nm-thick slices.
Biological samples are non-conductive, which can create issues when bombarded by a negatively charged electron beam. The samples can be unstable and drift, blurring the image, or create distortions in the signal, all of which are features of an accumulation of negative charge or “charging”. Heavy metal staining will help prevent this and in a SEM a conductive coating is applied to the sample to dissipate the charge.
What can you see with an electron microscope?
TEM is appropriate for imaging either very small samples, such as molecules or viruses, or the inside of cells and tissues (Fig. 4). The image produced is analogous to an x-ray and is a projection through the biological sample. Unlike a light microscope and SEM, everything in the image will be in focus due to the relatively large depth of field. TEM images are 2D. It is possible to obtain 3D information of viruses or molecules using a computerised image processing technique called single particle analysis. Larger samples have to be sectioned to between 50 and 200 nm in thickness in order for the beam to penetrate the sample. A series of sections can be imaged and the data assembled into a 3D volume, which is called serial section TEM. 3D data of 200-500 nm-thick sections of cells and tissues can be obtained using a technique called electron tomography.
SEM is used to image the surface of bulk samples, such as bacteria, cells, tissues and organisms (Fig 5). Atomic contrast can be detected using backscattered electrons (BSE) as heavier atoms produce a brighter signal. SEM micrographs have an optical illusion that creates the impression of a 3D image, however, no z dimension data is available thus making the images 2D only. 3D information can be obtained by tilting the sample and creating a stereoscopic image. SEM can be used to image sections and resin embedded samples, creating TEM-like images and 3D data (Fig. 6). This is used in advanced SEM techniques such as array tomography (Micheva and Smith, 2007), serial block-face scanning electron microscopy (SBFSEM) and focused ion beam scanning electron microscopy (FIBSEM).
Abbe, E., (1873) Beiträge zur Theorie des Mikroskops und der mikroskopischen Wahrnehmung. Archiv für mikroskopische Anatomie, 9(1), pp.413-418.
Hayat M., (2000) Principles and techniques of electron microscopy, biological applications. 4th Edition. Cambridge: Cambridge University Press.
Micheva, K.D., and Smith, S.J., (2007) Array tomography: A new tool for imaging the molecular architecture and ultrastructure of neural circuits. Neuron 55:25-36