Chemical Analysis of Minerals
Knowing a mineral’s chemical composition is essential to understanding its nature - composition determines crystal structure, physical properties, and stability in geological environments. The standard analytical tools fall into two broad categories: bulk dissolution methods that yield an average composition for all material dissolved, and beam methods that can target individual spots, grains, or compositional zones with spatial precision. The most widely used techniques for mineralogy are wet chemical methods, electron probe microanalysis (EPMA), scanning electron microscopy (SEM), X-ray fluorescence (XRF) spectroscopy, and mass spectrometry. An expanding array of additional techniques includes atomic absorption spectrometry, Raman spectroscopy, nuclear magnetic resonance spectroscopy, neutron activation and scattering analysis, optical emission spectroscopy, plasma emission spectroscopy, and synchrotron radiation analysis. Many commercial and research laboratories provide chemical analyses at modest cost. [1]
Wet Chemical Methods
Wet chemical analysis begins by dissolving or digesting the entire sample in a solvent - for geological materials, commonly an acid such as HCl, HNO₃, or HF. Once dissolved, various reagents are added to enable gravimetric, colorimetric, titration, or related analyses for each element of interest. Although these techniques can be highly accurate, they are no longer commonly used for geological materials because the procedures are slow, demand great skill, and - critically - yield only an average composition for all the material that was dissolved. They cannot resolve compositional variations within a single zoned grain, and they cannot distinguish between the target mineral and any fine-scale inclusions it may contain. [1]
Electron Probe Microanalysis (EPMA)
Electron probe microanalysis (EPMA) is performed on an instrument called an electron microprobe (or electron probe microanalyzer). It is one of the most widely used tools for mineral chemical analysis, and requires only a very small sample. A single spot on a mineral grain can be analyzed, or the electron beam can be rastered over the surface to produce a map of elemental distribution across the sample. [1]
Sample Preparation
The most common sample formats are standard petrographic thin sections (without coverslip) or mineral fragments mounted in an epoxy plug. The surface must be flat and carefully polished to minimize surface irregularities that would otherwise affect the emitted X-ray intensities. Samples are coated with a thin layer of a conductive material - commonly carbon or gold - so that the electrons striking the sample are conducted away rather than building up charge. [1]
Instrument Geometry
The electron microprobe and the scanning electron microscope share the same basic geometry (see Scanning Electron Microscopy below). Both are large cathode ray tubes. A tungsten filament is strongly heated to emit electrons, which are accelerated toward the anode by a voltage of several tens of kilovolts. A hole in the anode allows the high-energy electrons to pass through and strike the sample. Electromagnetic lenses focus the beam to a fine spot. Air interferes with the electron beam, so the instrument is evacuated before use. The exact position of the beam on the sample can be monitored with an optical microscope built into the instrument or with other imaging detectors. Most instruments accept multiple samples simultaneously and are fully automated - an operator can queue up many analysis spots, which the instrument then measures overnight. [1]
How Chemical Analysis Works: X-Ray Fluorescence Spectrometry
Chemical analysis in the electron microprobe is accomplished by X-ray fluorescence spectrometry. When the focused electron beam strikes the sample, some electrons lose energy by interacting with atoms in the sample, re-emitting some of it as a continuous X-ray spectrum. Other incident electrons dislodge electrons from the inner shells of target atoms; outer-shell electrons immediately drop in to fill these vacancies, emitting the characteristic X-ray spectrum whose wavelengths are specific to each element. K, L, and M peaks are produced when electrons fall into the K, L, and M shells respectively. Each element emits its own characteristic spectrum with an intensity proportional to the amount of that element present - this proportionality is the basis of quantitative analysis. The instrument is calibrated by comparison with standards of known composition, with corrections applied for X-ray absorption by other elements in the sample. A fundamental limitation is that neither detector type can distinguish different oxidation states (e.g., Fe²⁺ vs. Fe³⁺) or different isotopes of the same element. [1]
Wavelength-Dispersive Spectrometry (WDS)
A wavelength-dispersive spectrometer (WDS) uses a diffracting crystal (of known d-spacing) positioned at the Bragg angle θ required to diffract the characteristic Kα wavelength of the element being measured. The detector, positioned at 2θ, counts the diffracted X-ray intensity and sends it to a computer. By scanning the crystal and detector through a range of θ and 2θ, all X-ray peaks within the accessible wavelength range can be detected, with peak wavelengths identified from the Bragg equation and the known crystal d-spacing. A routine full WDS scan takes about half an hour. Electron microprobes are typically equipped with several WDS simultaneously, allowing multiple elements to be analyzed at once; different spectrometers carry different diffracting crystals to cover the full wavelength range needed to detect all elements. Some WDS spectrometers can detect elements as light as beryllium (atomic number 4), and detection limits for some elements can reach as low as ~30 parts per million. WDS analysis totals often do not sum to exactly 100% - this is normal, as all chemical analyses carry some error even when reported to three decimal places. [1]
Energy-Dispersive Spectrometry (EDS)
An energy-dispersive spectrometer (EDS) uses a single solid-state detector that captures all X-rays coming from the sample simultaneously and sorts them by energy level (E, keV). The energy-wavelength relationship E = hc/λ is used to convert between the two scales. A computer parses the entire energy spectrum and identifies the characteristic peaks of each element present. EDS is much faster than WDS - a basic scan takes less than half a minute - making it ideal for quickly confirming a mineral identity or identifying the major elements present in thin section. However, EDS has important limitations: many spectrometers do not perform well for elements lighter than sodium (atomic number 11); the light-element peaks cluster in the 0-1 keV range and can overlap significantly; and when oxygen is present, its strong Kα peak at 0.525 keV tends to overwhelm adjacent peaks. EDS results are considerably less accurate than WDS and are generally not suitable for minor or trace element analysis. The contrast between the two methods is illustrated clearly by comparing scans on molybdenite: EDS merges the sulfur and molybdenum peaks into a single broad, undifferentiated feature, while WDS resolves each peak in the characteristic spectrum distinctly. [1]
X-Ray Element Mapping and Additional Imaging Capabilities
Both WDS and EDS can be used for X-ray element mapping. The spectrometer is set to detect a specific element, and the electron beam is rastered systematically across the sample to build a pixel-by-pixel image in which brightness represents elemental concentration. The primary output is always grayscale, but image-editing software is routinely used to apply false color to highlight compositional variation. Electron microprobes are also typically equipped with backscatter electron detectors, which are especially useful for distinguishing different minerals (denser minerals appear brighter) and for quickly locating small accessory mineral grains such as zircon and monazite that have high density and stand out from the surrounding silicate matrix. Some microprobes also carry visible light detectors to capture cathodoluminescence - visible-wavelength emission caused by the electron beam - which provides insight into crystal growth history, replacement, and deformation at a scale not accessible by other means. [1]
Scanning Electron Microscopy (SEM)
A scanning electron microscope (SEM) shares the same basic geometry as the electron microprobe - both are essentially focused electron beam instruments - but the SEM is designed primarily for imaging rather than quantitative chemical analysis. SEM can produce images by detecting secondary electrons or backscattered electrons, which reveal surface topography, mineral morphology, and density contrasts at very high magnification and depth of field. [1]
Secondary electrons - low-energy electrons dislodged from the outermost atoms of the sample - are the most common signal used for SEM imaging. The brightness of any point in the image reflects how many secondary electrons from that point can reach the side-mounted secondary electron detector. Surfaces tilted toward the detector appear brighter; recesses and holes appear dark because secondary electrons from these locations cannot escape. This geometry makes secondary electron images extremely effective at showing surface topography and the three-dimensional shapes of mineral grains. [1]
SEM images are always produced as grayscale; color in published images is added afterward using digital editing software. The magnification range of a conventional SEM extends from 5× to 300,000×, far beyond the 25× to 400× range of a petrographic microscope. [1]
Backscattered electrons (higher-energy electrons deflected back out of the sample by interaction with atomic nuclei) are detected by a separate backscatter detector mounted above the sample. Because high-atomic-weight elements scatter electrons more efficiently, minerals rich in heavy elements appear brighter in a backscatter image. A practical example: ilmenite (FeTiO3) appears distinctly brighter than adjacent quartz (SiO2) because iron and titanium are much heavier than silicon. [1]
Most SEMs are equipped with an energy-dispersive spectrometer (EDS) to provide semi-quantitative elemental data. EDS makes it easy to quickly identify the major elements in a mineral grain, but EDS analyses from a standard SEM are less accurate than those from an electron microprobe because surface irregularities in the sample affect how much of the characteristic X-ray spectrum actually reaches the detector. Some instruments are also fitted with wavelength-dispersive spectrometers, which improve sensitivity for trace elements. [1]
Reporting Chemical Analyses
Major Elements
Chemical analyses of major element composition are conventionally reported as weight percent (wt%) of each constituent. For minerals in which oxygen is the dominant anion - the majority of minerals - major element data are reported not as the pure element but as the weight percent of the corresponding metal oxide (e.g., SiO2, Al2O3, FeO). The amount of oxygen present is not directly measured; instead, the weight percent of each cation is measured and the oxygen required by the stoichiometry of the oxide is added in. A binary element such as arsenopyrite may instead be reported in elemental form: for example, 34.30% Fe, 46.01% As, and 19.69% S. [1]
Two different forms of water may be reported. H2O− represents actual molecular water adsorbed onto grain surfaces or trapped in voids - it is not part of the crystal structure and is detected by measuring weight loss when the sample is heated to 100-110°C. H2O+ represents hydrogen that is genuinely part of the crystal structure, though not necessarily in the form of water molecules; it is typically available only through wet chemical techniques and is not always reliable. [1]
Analytical totals rarely sum to exactly 100%. A total below 100% may indicate undetected elements; a total above 100% points to compensating errors. Shared standards among laboratories help calibrate instruments and allow results from different facilities to be compared directly. [1]
Trace Elements
Trace element concentrations are reported in parts per million (ppm) of the element, or occasionally of its oxide. Trace elements either substitute in small amounts for a major cation in the crystal structure, or occupy sites not normally filled in the ideal structure - in both cases they constitute structural defects. Rare earth element (REE) data are routinely normalized by dividing observed concentrations by the corresponding elemental abundances in an average chondrite (a primitive meteorite inferred to represent the bulk chemistry of the solar system), making it easy to compare REE patterns across different samples and localities. [1]
Related Topics
Scanning Electron Microscopy (SEM)
A scanning electron microscope (SEM) is functionally very similar to an electron microprobe but is designed primarily for imaging rather than quantitative chemical analysis. The SEM excels at...
Silicate Minerals: Structure and Classification
Silicate minerals are by far the most abundant group of minerals in the Earth's crust, and their diversity and economic importance make them the dominant focus of systematic mineralogy. The common...
Structural Defects
The idealised model of a crystalline structure assumes a perfectly regular, infinitely repeating pattern of atoms or ions that continues without interruption throughout an entire crystal. Real...
Bragg's Equation
Bragg's equation (nλ = 2d sinϑ) describes the conditions under which X-rays diffracted by parallel planes of atoms in a crystal interfere constructively, forming a detectable reflection. It is the fundamental relationship linking diffraction angle, interplanar spacing, and X-ray wavelength.
References & Citations
- 1.Introduction to Mineralogy Nesse, W. D.
Master UPSC Geology Optional
Ex-ONGC Geologist & Rank Holder
Learn the exact analytical answer-writing patterns needed for UPSC Optional from an AIR 2 & AIR 25 holder.
Offline in Delhi