SEM-EDS Testing Explained: Element Mapping and Point

SEM-EDS testing combines scanning electron microscopy with energy-dispersive X-ray spectroscopy to measure the elemental composition of solid surfaces at micrometre scale. It works on metals, ceramics, coatings, welds, corrosion products, polymers and electronic components. Two modes are central: point analysis quantifies composition at a chosen spot, and element mapping shows how elements are distributed across a region.

Principle & mechanism

The analytical signal comes from electron–atom interaction. When the primary electron beam strikes the specimen, inner-shell electrons of the atoms may be ejected. This leaves a vacancy. An electron from an outer shell then fills it, and the energy difference between the two shells is released as a characteristic X-ray photon. Each element emits photons at energies specific to its electron shell structure. The measured X-ray energy therefore identifies the element, and the photon intensity relates to its concentration. A single excitation event cannot displace atoms heavier than the incident beam permits, so the lightest elements are the hardest to quantify. Hydrogen and helium produce no characteristic X-rays at all. Low-energy lines of carbon, nitrogen and oxygen overlap or absorb readily, so they need careful interpretation.

Working principle & instrument setup

A complete system has three parts: the electron column of the scanning electron microscope, the EDS detector, and processing software. The column generates a focused electron beam, usually at accelerating voltages between roughly 5 and 30 kV, which is rastered across the specimen surface. The EDS detector, typically a silicon drift detector cooled by Peltier elements, sits close to the specimen. It captures X-rays emitted from the interaction volume. A thin detector window determines how much of the low-energy spectrum is transmitted. Pulse processing converts each detected photon into an energy channel, building a spectrum of counts versus energy. The operator sets accelerating voltage, beam current, working distance, dead time and acquisition time. These settings depend on the elements of interest and the required image or map quality.

Sample types & surface preparation

Specimens must be vacuum-stable and electrically conductive, or made conductive by coating. Acceptable materials include metals and alloys, fracture surfaces, welds and solder joints, ceramics, glasses, corrosion deposits, powders, and hard phases in polished microsections. Non-conductive samples such as polymers, minerals and composites are sputter coated with a thin gold, platinum or carbon layer. Preparation quality directly governs analytical accuracy. Polished cross-sections yield quantitative results, whereas rough fracture surfaces limit analysis to semi-quantitative comparison. Samples should be free of oil, grease and loose particles. Contamination introduced during cutting or mounting appears as spurious carbon, chlorine or silicon peaks. Size is limited by the chamber, so consult the laboratory in advance about dimensions and orientation.

Element mapping & point analysis modes

Point analysis collects a full spectrum at a stationary beam position, for example on an inclusion, a particle, a coating layer or a suspect phase. It reports semi-quantitative to quantitative elemental composition, commonly normalised to 100 weight percent. Line scans record composition changes along a traverse. These are useful for diffusion profiles, plating layers and weld interfaces. Element mapping rastered the beam and is now typically performed by spectral imaging, in which a spectrum is stored at every pixel. Distribution maps of selected elements are reconstructed afterwards. Maps reveal segregation, corrosion products, foreign particles and layer structures that point analysis alone would miss. Common practice is to combine both modes: an overview map locates features, and point or line analysis quantifies them. As described in our article Scanning Electron Microscopy (SEM), imaging modes such as secondary and backscattered electrons complement this spectral information.

Detection

limits & analytical performance
Detection sensitivity depends on element, matrix, acquisition time and accelerating voltage. It is commonly quoted in the order of 0.1 to a few weight percent for routine analysis. This is considerably poorer than bulk techniques such as ICP-OES or XRF for trace work. Quantification compares measured peak intensities with standards or with standardless models built from theoretical X-ray yields. Accuracy is best for flat, polished, homogeneous samples. Typical analytical depth spans roughly one micrometre, set by the beam energy and the specimen density, so results represent a near-surface volume rather than the bulk. Overlapping peaks, such as those among transition metals, must be deconvoluted by the software. Light-element quantification carries larger uncertainty. Reported results should state voltage, mode, whether coating corrections were applied, and the normalisation basis.

Applications & co-testable techniques

Typical uses include identification of inclusions and second phases in steels, characterisation of plating and conversion layers, corrosion product identification, contaminant and foreign-particle analysis, solder-joint and semiconductor defect inspection, and mineral or ceramic phase studies. In failure investigation, SEM-EDS is rarely applied alone. It is routinely paired with metallographic examination, hardness testing and image analysis. Our article Rolling Bearing Failure Analysis: Root Cause Investigation with Metallography and SEM Testing illustrates how microstructure and local chemistry are combined to trace a root cause. Additional co-testable techniques include electron backscatter diffraction for crystallographic information, wavelength-dispersive spectroscopy for better light-element and trace sensitivity, X-ray photoelectron spectroscopy for surface films a few nanometres thick, and Raman or FTIR spectroscopy when organic species must also be identified.