Working Principles and Architecture of the Electron Probe (EPMA)

Overview An electron probe microanalyzer (EPMA) determines elemental composition by exciting characteristic X-rays from a sample using a focused electron beam. The instrument combines an electron-optical column, X-ray spectrometers, and data acquisition/display systems, and is often integrated with a scanning electron microscope (SEM) to enable correlated imaging and microanalysis. Analytical principle According to Moseley’s law, each element emits characteristic X-rays with a specific wavelength. By detecting these wavelengths (or their corresponding photon energies), EPMA performs qualitative identification of elements. Quantitative analysis is achieved by comparing the X-ray intensity of an element in the unknown sample with that from a standard, typically using measured diffracted intensities. A fundamental requirement is that the incident electron energy must exceed the critical ionization energy of the element’s target inner shell to generate the characteristic X-rays. Instrument architecture An EPMA consists of: (1) the electron-optical system (column), (2) X-ray spectrometers, and (3) information recording and display electronics. The electron-optical section is similar to that of an SEM, and many commercial systems integrate both. 1) Electron-optical system The column must deliver an electron beam with sufficiently high accelerating voltage, adequate beam current, and a minimum probe diameter at the impact point to efficiently excite X-rays. Typical implementations use a tungsten thermionic electron gun with two to three condenser lenses. Compared with conventional SEM operation, EPMA works at higher probe currents to increase X-ray signal: beam current is typically in the 10^-9–10^-7 A range, with accelerating voltages of 10–30 kV and probe diameters around 0.5 μm. A notable distinction from standard SEM configurations is the inclusion of a coaxial reflected-light optical microscope used to select and verify the analysis point. A fluorescent target (e.g., ZrO2) can be irradiated to visualize the beam position, which is then aligned to the microscope crosshair and to the correct geometry for the X-ray spectrometers. The optical microscope typically offers 100–500× magnification and allows simultaneous optical observation and X-ray analysis. 2) X-ray spectrometers Electron bombardment generates characteristic X-rays whose wavelengths and energies are element-specific. Two detection strategies are used: - Wavelength Dispersive Spectrometry (WDS), referred to as the wavelength spectrometer - Energy Dispersive Spectrometry (EDS), referred to as the energy spectrometer (1) Wavelength spectrometer (WDS) WDS identifies elements by measuring X-ray wavelength via Bragg diffraction using a dispersive crystal with lattice spacing d. When incident X-rays of wavelength λ meet the Bragg condition (2d sinθ = λ), they are diffracted. For any fixed incidence angle θ, only one wavelength satisfies the condition. By scanning the detector over 2θ, the instrument maps wavelengths to specific elements and acquires intensities. To increase diffracted intensity, focusing geometry with a curved crystal is used because X-rays cannot be focused by lenses. In a focusing setup, the X-ray source point on the sample surface, the diffracting crystal, and the detector lie on a common focusing circle. The crystal is bent so that the diffracting planes have a radius of curvature equal to 2R and the surface is polished to match the focusing circle. With the detector slit positioned at the focal point, it receives the strongly diffracted, single-wavelength X-rays from the entire crystal surface. Common WDS dispersive crystals provide defined energy/wavelength coverage ranges. The diffracted X-rays are typically detected with proportional counters: incident photons ionize the fill gas and generate electrical pulses. Signals are amplified (preamp and main amplifier) to approximately 0–10 V pulse heights and then processed by a pulse height analyzer. (2) Energy spectrometer (EDS) In EDS, X-ray photons pass through a beryllium window into a Si(Li) solid-state detector. Each photon deposits energy in the silicon, creating electron–hole pairs. The number of pairs is proportional to photon energy; for example, Fe Kα generates about 1685 pairs and Cu Kα about 2110 pairs. The resulting charge pulse produces a voltage on a fixed capacitance (e.g., 1 μμF): approximately 0.27 mV for Fe Kα and 0.34 mV for Cu Kα. The very small pulses are amplified by a low-noise FET preamplifier and a main amplifier, then sent to a multichannel pulse height analyzer (MCA). The MCA digitizes pulse heights and maps them to channel addresses, establishing a correspondence between channel number and photon energy. A common energy range is 0.2–20.48 keV. With 1024 total channels, the energy dispersion is about 20 eV per channel. Lower-energy photons map to lower channel numbers and higher-energy photons to higher channels. Counting statistics in each channel yield the intensity of elemental lines, which supports qualitative identification and relative quantitative information. The output is displayed as an X-ray energy spectrum on an XY recorder or CRT. (3) WDS vs. EDS: performance comparison - Detection efficiency: EDS generally has higher detection efficiency. The Si(Li) detector subtends a larger solid angle relative to the source, and WDS loses intensity in the diffraction process. - Spatial analysis capability: EDS can operate at lower beam currents due to higher efficiency, enabling smaller probe diameters and improved spatial resolution. In microprobe/analytical SEM modes, EDS can analyze micro- to submicrometer regions, whereas WDS spatial resolution is typically in the micrometer range. - Energy (spectral) resolution: EDS has a typical energy resolution of about 149 eV. WDS achieves a wavelength resolution around 0.5 nm, corresponding to roughly 5–10 eV energy resolution, i.e., approximately an order of magnitude better than EDS in spectral discrimination. - Analysis speed: EDS captures the full spectrum at a point simultaneously and can deliver qualitative results within minutes. WDS scans discrete wavelengths sequentially, so comprehensive analyses may take hours. - Elemental range: WDS can measure all elements from Be to U. Standard Si(Li) EDS with a Be window absorbs very light-element X-rays and typically measures elements from Na upward. - Reliability and quantitative accuracy: EDS systems are structurally simpler with no mechanical scanning in the spectrometer, offering good stability and reproducibility. However, typical quantitative errors for WDS are lower (about 1–5%) than for EDS (about 2–10%). - Sample requirements: WDS requires a flat sample surface to satisfy focusing geometry. EDS is less sensitive to surface topography and is suitable for rougher surfaces. Conclusion WDS and EDS are complementary. Modern instruments often integrate both approaches with SEM imaging in a single platform, enabling rapid, correlative analysis of microstructure and composition across a broad range of materials.

This article was originally published on our Chinese website. You can view the original version below: