Transmission Electron Microscopy is an electron microscopy technique in which a beam of electrons passes through an electron transparent specimen and the transmitted and scattered electrons are used to generate images, diffraction patterns or spectroscopic information. Because electrons can interact with matter strongly, specimens generally need to be sufficiently thin to allow an appreciable fraction of the beam to pass through them.
TEM can provide exceptionally high spatial resolution and can reveal internal microstructure, crystal defects, interfaces, nanoparticles, thin films, lattice fringes and, with appropriate instrumentation and conditions, atomic scale arrangements. The technique can operate in conventional imaging modes, diffraction modes, scanning transmission modes and analytical modes. TEM is particularly powerful because image formation and diffraction can be performed on the same nanoscale region, allowing morphology and crystallographic structure to be correlated. Unlike SEM, which primarily examines surfaces and near surface regions, TEM can directly investigate internal structures within thin specimens.
In conventional TEM, electrons are generated by an electron source, accelerated to high energy and focused into a beam using electromagnetic lenses. The beam passes through a thin specimen, where electrons undergo elastic and inelastic scattering. The objective lens forms an intermediate image or diffraction pattern, and additional lenses magnify the information for detection. Regions of the specimen that strongly scatter electrons can appear darker under appropriate bright field conditions because fewer electrons contribute to the transmitted beam.
Dark field imaging instead uses selected diffracted or scattered electrons to emphasise particular structural regions. Image contrast can arise from mass thickness, diffraction, strain, defects, composition and phase. At sufficiently high resolution, phase relationships between transmitted and diffracted electron waves become important, allowing lattice information to be encoded into the image. Interpretation therefore requires understanding electron diffraction, wave interference, lens behaviour, specimen thickness and imaging conditions rather than treating TEM images as conventional photographs.
Electron diffraction is one of TEM's most important capabilities. Crystalline materials diffract electrons according to their atomic arrangement and lattice geometry, producing patterns that can be used to determine crystal structure, orientation, lattice spacing and phase identity. Selected area electron diffraction examines diffraction from a defined region, while convergent beam electron diffraction can provide more local structural information and symmetry data.
Diffraction contrast can also reveal dislocations, stacking faults, strain fields and other defects because local changes in crystal orientation or lattice spacing alter the diffraction condition. High resolution TEM can display periodic lattice fringes whose spacing and orientation correspond to crystallographic planes under appropriate imaging conditions. However, lattice fringes should not automatically be interpreted as direct atom by atom images because image contrast depends on specimen thickness, defocus, aberrations and crystallographic orientation. Computer simulation and comparison with calculated images may be necessary for rigorous structural interpretation, particularly when atomic scale conclusions are being drawn.
TEM specimen preparation is often the most technically demanding part of the technique. Conventional bulk materials must frequently be thinned to electron transparency using mechanical polishing, dimpling, ion milling or combinations of methods. Focused ion beam instruments can extract site specific thin sections from selected regions of bulk specimens, making it possible to investigate interfaces, defects or device structures identified during other measurements. Powders and nanoparticles can sometimes be dispersed onto suitable electron transparent support films with comparatively simple preparation.
Biological specimens may be prepared by fixation, dehydration, embedding and ultrathin sectioning, or frozen approaches may be used for cryogenic studies. Preparation can alter oxidation states, induce amorphisation, redistribute mobile species, introduce contamination or mechanically damage delicate structures. Ion milling can produce preferential sputtering, implantation or surface damage. Consequently, TEM results should be interpreted with knowledge of how the specimen was produced and whether the observed region remains representative of the original material.
Scanning Transmission Electron Microscopy is an important extension of TEM in which a focused electron probe is scanned across the specimen and transmitted signals are detected as a function of probe position. Different detector geometries generate different contrast mechanisms. Annular dark field imaging can produce strong atomic number dependent contrast, while high angle annular dark field imaging can provide so called Z contrast that is particularly useful for observing heavier atomic columns against lighter backgrounds under suitable conditions.
STEM can be combined with EDS and electron energy loss spectroscopy to obtain chemical information at very high spatial resolution. EELS can measure energy losses associated with elemental excitations, electronic transitions, bonding and other interactions. Fine structure in an energy loss spectrum can contain information about oxidation state and local coordination. The combination of STEM imaging and spectroscopy therefore permits detailed correlation between structure and chemistry, although beam damage, specimen thickness, detector geometry and signal statistics remain important limitations.
Aberration correction has significantly expanded the capabilities of modern TEM and STEM. Electron lenses inherently suffer from aberrations that limit the smallest probe or the sharpness of an image. Correctors can compensate important lens aberrations and allow smaller probes and improved information transfer under suitable operating conditions. Modern instruments can therefore resolve extremely fine structural features and, in favourable specimens, individual atomic columns. High resolution does not eliminate the need for careful experimental design.
Atomic scale imaging requires stable specimens, precise alignment, suitable thickness, appropriate orientation and careful control of dose. Beam sensitive materials can change before sufficient information is collected. Drift and vibration can blur images, while lens instabilities can affect quantitative interpretation. Advanced detectors and computational methods can further improve sensitivity and enable four dimensional STEM, electron tomography and other specialised measurements. These developments have transformed TEM from a conventional imaging method into a highly integrated platform for nanoscale structural, chemical and crystallographic analysis.
TEM is used extensively in nanomaterials research, semiconductor technology, metallurgy, catalysis, battery science, polymers, ceramics, geology, biology and medicine. Nanoparticle morphology, particle size, crystallinity, surface layers and internal defects can be examined directly. Semiconductor interfaces and multilayer devices can be investigated at nanometre and atomic scales. Catalysts can be studied to determine particle structure, support interactions and active material distributions. Battery materials can be examined before and after cycling to identify phase transformations, cracks, structural degradation and interfacial layers.
In biology, electron microscopy can reveal cellular ultrastructure and specialised cryogenic methods can investigate biological macromolecules. Failure analysis can use site specific TEM specimens to identify nanoscale causes of device or material failure. The ability to combine imaging, diffraction and spectroscopy from a highly local region is particularly valuable when a material property depends on a small number of nanoscale features.
TEM has substantial limitations despite its exceptional resolution. Specimen preparation can be difficult, expensive and destructive. Thin sections may not represent bulk behaviour, and preparation can introduce artefacts. The electron beam can cause radiolysis, knock on damage, heating, sputtering, contamination and structural transformations. Interpretation of high resolution images can be model dependent, particularly when multiple crystallographic orientations or overlapping structures contribute to the recorded intensity. TEM instruments are also complex and expensive, requiring high vacuum, precise alignment and specialist expertise.
A single image represents a very small region, so statistically meaningful conclusions may require examination of many areas or complementary bulk measurements. TEM should therefore not be regarded as universally superior to other microscopy methods. Its principal advantage is the combination of very high spatial resolution with direct access to internal structure, diffraction and nanoscale analytical signals. Its limitations are best managed through careful preparation, low dose strategies, complementary characterisation and explicit consideration of representativeness and measurement uncertainty.
