11.6 Spectroscopy

Principles of spectroscopy

Spectroscopy is the study of how electromagnetic radiation interacts with matter. In geochemistry, it's one of the most widely used approaches for determining what geological materials are made of, from major element compositions down to trace-level concentrations. These techniques give you chemical and structural information about rocks, minerals, fluids, and gases, which in turn helps reconstruct processes like magma differentiation, metamorphism, and weathering.

Electromagnetic spectrum

The electromagnetic spectrum covers all wavelengths of radiation, from long-wavelength radio waves to short-wavelength gamma rays. The regions, in order of decreasing wavelength, are: radio, microwave, infrared, visible, ultraviolet, X-ray, and gamma ray. Visible light sits in a narrow window of roughly 380–700 nm.

Each region interacts with matter differently, and that's why so many distinct spectroscopic techniques exist. Infrared radiation probes molecular vibrations, X-rays excite inner-shell electrons, and so on. The choice of spectral region determines what kind of information you can extract from a sample.

Light-matter interactions

Four main types of interaction matter for spectroscopy:

• Absorption occurs when atoms or molecules take in photons at specific wavelengths, jumping to higher energy states. This is the basis of techniques like AAS and UV-Vis spectroscopy.
• Emission is the reverse: excited atoms or molecules release energy as photons when they drop back to lower energy states. ICP-AES relies on this process.
• Scattering happens when light is deflected by particles or molecules. Rayleigh scattering is elastic (no energy change), while Raman scattering is inelastic (the scattered photon gains or loses energy, revealing vibrational information).
• Refraction describes the bending of light as it passes between media with different optical densities. It's less central to analytical spectroscopy but matters for instrument optics.

Beer-Lambert law

The Beer-Lambert law is the foundation of quantitative absorption spectroscopy. It relates how much light a sample absorbs to the concentration of the absorbing species:

$A = \varepsilon b c$

• $A$ = absorbance (dimensionless)
• $\varepsilon$ = molar attenuation (absorptivity) coefficient, specific to the substance and wavelength (units: L·mol⁻¹·cm⁻¹)
• $b$ = path length through the sample (cm)
• $c$ = concentration of the absorbing species (mol/L)

This equation assumes a linear relationship between absorbance and concentration. That linearity breaks down at high concentrations due to molecular interactions and instrumental effects, so calibration curves should always be checked for deviations.

Types of spectroscopy

Different spectroscopic techniques exploit different light-matter interactions, and the choice of method depends on what you need to know: elemental composition, molecular structure, oxidation state, or something else. Here's a rundown of the most important techniques in geochemistry.

Atomic absorption spectroscopy

Atomic absorption spectroscopy (AAS) measures the absorption of light by free atoms in the gas phase. The sample is atomized using either a flame or a graphite furnace, and a light source (typically a hollow cathode lamp) emits radiation at the characteristic wavelength of the target element. The amount of light absorbed is proportional to the element's concentration in the sample.

AAS is highly sensitive, capable of detecting trace elements down to the parts-per-billion (ppb) range, especially with graphite furnace atomization. The main limitation is that it analyzes only one element at a time, which makes multi-element surveys time-consuming compared to techniques like ICP-AES.

Atomic emission spectroscopy

Atomic emission spectroscopy (AES) works in the opposite direction from AAS: instead of measuring absorbed light, it measures light emitted by atoms that have been thermally or electrically excited. Each element produces a characteristic set of emission lines, allowing both identification and quantification.

The most common form in geochemistry is Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES), which uses an argon plasma at temperatures around 6,000–10,000 K to excite the sample. A major advantage over AAS is simultaneous multi-element analysis, making it efficient for routine geochemical surveys.

X-ray fluorescence spectroscopy

X-ray fluorescence (XRF) spectroscopy bombards a sample with high-energy X-rays, which eject inner-shell electrons from atoms. As outer-shell electrons fill the vacancies, they emit characteristic X-rays whose energies identify the element and whose intensities indicate concentration.

XRF is non-destructive, which makes it valuable for analyzing irreplaceable specimens. It's widely used for bulk rock analysis and can detect elements from sodium (Na) through uranium (U). Handheld XRF analyzers have become standard tools in mineral exploration and environmental site assessment.

Infrared spectroscopy

Infrared (IR) spectroscopy probes molecular vibrations and rotations by passing infrared radiation through a sample. Different functional groups and bond types absorb at characteristic frequencies, producing an absorption spectrum that acts as a molecular fingerprint.

Two common variants are Fourier Transform Infrared Spectroscopy (FTIR) and Near-Infrared Spectroscopy (NIR). In geochemistry, IR spectroscopy is particularly useful for identifying minerals (especially hydrous phases like clays and amphiboles), characterizing organic matter, and measuring water content in glasses and nominally anhydrous minerals.

Raman spectroscopy

Raman spectroscopy is based on the inelastic scattering of monochromatic laser light. When photons interact with molecular vibrations, a small fraction are scattered at shifted frequencies. These Raman shifts provide information about molecular vibrations and crystal structures.

Raman and IR spectroscopy are complementary: some vibrational modes are Raman-active but IR-inactive, and vice versa (this follows from symmetry selection rules). Raman spectroscopy is non-destructive and requires minimal sample preparation, making it excellent for in situ analysis of minerals, fluid inclusions, and polymorphs. It's also useful for studying phase transitions under varying pressure and temperature.

Spectroscopic instrumentation

Getting reliable data from spectroscopy depends heavily on the quality and proper operation of the instruments involved. Understanding the key components helps you troubleshoot problems and choose the right setup for your analysis.

Spectrometers

A spectrometer separates incoming light into its component wavelengths and measures their intensities. The core components are:

• Light source (lamp, laser, or synchrotron beam)
• Wavelength selector (monochromator using a prism or diffraction grating, or an interferometer in Fourier transform instruments)
• Detector (converts light to an electrical signal)

Dispersive spectrometers use gratings or prisms to physically spread light by wavelength. Fourier transform spectrometers collect all wavelengths simultaneously and use mathematical transformation to produce the spectrum, offering better signal-to-noise ratios and faster acquisition. Spectral resolution refers to the instrument's ability to distinguish closely spaced spectral lines, and it varies with the optical design.

Detectors

Detectors convert photon intensity into a measurable electrical signal. Common types include:

• Photomultiplier tubes (PMTs): very sensitive, good for low-light applications, but measure one wavelength at a time
• Charge-coupled devices (CCDs): array detectors that capture a range of wavelengths simultaneously, widely used in modern spectrometers
• Photodiode arrays: similar to CCDs in concept, offering fast readout for multi-channel detection

Detector choice depends on the spectral range, required sensitivity, and dynamic range. Many detectors are cooled (thermoelectrically or with liquid nitrogen) to reduce thermal noise and improve the signal-to-noise ratio.

Sample preparation techniques

Proper sample preparation is critical for accurate, reproducible results. The approach depends on both the sample type and the technique:

• Solids: may need grinding to a fine powder, pressing into pellets (for XRF), or fusion with a flux (lithium borate) to create homogeneous glass disks
• Liquids: often require dilution to bring concentrations into the instrument's linear range, along with filtration and sometimes acid digestion
• Gases: may need concentration (e.g., cryogenic trapping) or separation before analysis

Contamination control matters at every step. Clean labware, reagent blanks, and careful handling prevent introducing artifacts that could compromise your data.

Applications in geochemistry

Spectroscopy underpins a huge range of geochemical investigations. The techniques described above are applied to problems spanning petrology, mineralogy, environmental science, and more.

Elemental analysis

Determining elemental concentrations is one of the most common applications. ICP-AES and XRF provide quantitative data on major elements (Si, Al, Fe, Mg, Ca, Na, K), minor elements, and trace elements in rocks, soils, and sediments. This data feeds into geochemical classification schemes, tectonic discrimination diagrams, and models of magmatic differentiation and crustal evolution. Geochemical fingerprinting of rock units relies on characteristic element ratios that can be matched across regions.

Mineral identification

IR and Raman spectroscopy are the go-to techniques for identifying mineral species and distinguishing polymorphs (minerals with the same composition but different crystal structures, like calcite vs. aragonite). These methods are fast and often non-destructive. X-ray diffraction (XRD) complements spectroscopic identification by providing detailed crystal structure data. Together, these tools support studies of metamorphic mineral assemblages, diagenetic changes in sediments, and alteration products in hydrothermal systems.

Trace element detection

Trace elements (present at ppm to ppb levels) carry disproportionate geochemical significance. Rare earth element (REE) patterns, for example, fingerprint magmatic sources and differentiation processes. Techniques like ICP-MS (inductively coupled plasma mass spectrometry) and GFAAS (graphite furnace atomic absorption spectroscopy) provide the sensitivity needed for these low concentrations. Trace element data also supports environmental monitoring of potentially toxic metals like Pb, Cd, and As in soils and water.

Isotope ratio determination

Isotope ratios reveal information that elemental concentrations alone cannot. Radiogenic isotopes (e.g., $^{87}\text{Sr}/^{86}\text{Sr}$, $^{143}\text{Nd}/^{144}\text{Nd}$) are used for geochronology and tracing mantle sources. Stable isotopes (e.g., $\delta^{18}\text{O}$, $\delta^{13}\text{C}$) track fluid-rock interactions, paleoclimate conditions, and biogeochemical cycling. Mass spectrometry is the primary tool, though some specialized spectroscopic methods (like cavity ring-down spectroscopy for light stable isotopes) are increasingly used.

Interpretation of spectra

Raw spectra are just plots of intensity versus wavelength or energy. Turning them into useful geochemical information requires careful interpretation.

Peak identification

Each peak in a spectrum corresponds to a specific electronic transition, molecular vibration, or elemental emission/absorption line. Identifying peaks involves:

1. Noting the peak position (wavelength, wavenumber, or energy)
2. Comparing against reference spectral libraries and databases
3. Considering peak shape and intensity, which can indicate concentration, coordination environment, or crystallinity
4. Applying selection rules to determine which transitions are allowed

Automated peak-matching software helps with complex spectra, but you should always verify assignments manually, especially in multi-phase geological samples.

Quantitative analysis

To convert peak intensities into concentrations, you need calibration. The two main approaches are:

• Calibration curves: Measure a series of standards with known concentrations, plot intensity vs. concentration, and use the resulting curve to determine unknowns.
• Standard addition: Add known amounts of the analyte to the sample itself, which helps account for matrix effects.

Internal standards (an element or compound added at a known concentration) improve precision by correcting for instrument drift and matrix variability. Always estimate and report your analytical uncertainties.

Spectral interferences

In complex geological samples, spectral features from different species can overlap. For example, in ICP-AES, emission lines of different elements may sit at nearly the same wavelength. Strategies to deal with interferences include:

• Selecting alternative analytical lines that are free of overlap
• Using spectral deconvolution algorithms to mathematically separate overlapping peaks
• Adjusting sample preparation to remove or reduce the interfering species
• Applying inter-element correction factors during data processing

Data processing techniques

Modern spectroscopy generates large datasets that benefit from computational processing:

• Baseline correction removes background signal that can distort peak measurements
• Smoothing and noise reduction improve spectral quality without losing real features
• Multivariate statistics (principal component analysis, cluster analysis) help identify patterns in complex datasets with many variables
• Spectral deconvolution separates overlapping peaks for more accurate quantification
• Machine learning algorithms are increasingly applied for automated classification and interpretation, particularly with large hyperspectral datasets

Advanced spectroscopic methods

These techniques push beyond the capabilities of standard laboratory instruments, offering higher spatial resolution, greater sensitivity, or unique analytical capabilities.

Laser ablation ICP-MS

LA-ICP-MS couples a focused laser beam with an ICP mass spectrometer. The laser ablates a tiny volume of solid sample (spot sizes down to ~1 μm), and the resulting aerosol is carried into the ICP-MS for elemental and isotopic analysis. This allows in situ measurement of trace elements and isotope ratios within individual mineral grains, growth zones, and inclusions, all without dissolving the sample. It's become indispensable for U-Pb zircon geochronology, trace element mapping, and studying compositional zoning.

Synchrotron-based techniques

Synchrotron facilities produce extremely intense, tunable X-ray beams that enable techniques not possible with conventional lab sources. Key methods include:

• X-ray absorption spectroscopy (XAS): determines oxidation states and local coordination environments of specific elements, even at low concentrations
• X-ray fluorescence microscopy (μ-XRF): maps elemental distributions at micron-scale resolution

Synchrotron beamlines also allow experiments under extreme conditions (high pressure and temperature), simulating deep Earth environments. Access requires beam time proposals at national or international facilities.

Hyperspectral imaging

Hyperspectral imaging collects a full spectrum for every pixel in an image, creating three-dimensional data cubes (two spatial dimensions plus one spectral dimension). In geochemistry, this is applied at scales ranging from thin sections in the lab to satellite-based remote sensing of entire geological formations. Applications include mapping mineral distributions across outcrops, identifying alteration zones in ore deposits, and characterizing surface compositions on other planets (e.g., Mars rovers carry hyperspectral instruments).

Limitations and challenges

No analytical technique is perfect. Understanding where spectroscopic methods fall short helps you design better experiments and interpret results more carefully.

Detection limits

The detection limit is the lowest concentration of an analyte that can be reliably distinguished from background noise. It varies by technique, instrument, and sample matrix. For example, ICP-MS can reach sub-ppb detection limits for many elements, while XRF typically has detection limits in the low ppm range. Factors that degrade detection limits include high background noise, spectral interferences, and poor sample preparation. Pre-concentration steps (like ion exchange or co-precipitation) can improve sensitivity when needed.

Matrix effects

The composition of the sample itself can influence the measured signal. In XRF, for instance, the absorption and enhancement of X-rays by surrounding elements can cause the signal for a given element to be higher or lower than expected. This is a significant issue in geological samples, which are chemically complex and variable.

Mitigation strategies include:

• Using matrix-matched calibration standards (standards with a composition similar to your unknowns)
• Applying internal standardization
• Using fundamental parameters methods that model X-ray physics to correct for matrix composition

Sample heterogeneity

Geological samples are often heterogeneous at multiple scales. A single rock chip might contain several mineral phases with very different compositions. If you're doing bulk analysis (like pressed-pellet XRF), you need to ensure the analyzed portion is representative of the whole sample. This typically means grinding to a fine powder and homogenizing thoroughly. For techniques like LA-ICP-MS, heterogeneity is actually an advantage: you can deliberately target individual phases or zones to study compositional variation.

Integration with other techniques

Spectroscopy rarely works in isolation. Combining it with other analytical methods gives you a more complete picture of your samples.

Spectroscopy vs. chromatography

Spectroscopy characterizes what's in a sample based on light-matter interactions. Chromatography physically separates the components of a mixture. The two are often coupled:

• GC-MS (gas chromatography-mass spectrometry) separates volatile organic compounds and then identifies them by mass spectrum. It's a workhorse in organic geochemistry for biomarker analysis.
• LC-ICP-MS (liquid chromatography-ICP-MS) separates dissolved species and then measures elemental or isotopic composition, useful for speciation studies (e.g., distinguishing Cr(III) from Cr(VI) in water).

Neither approach replaces the other. Chromatography excels at separating complex mixtures; spectroscopy excels at identifying and quantifying what's been separated.

Complementary analytical methods

• Electron microscopy (SEM, TEM): provides high-resolution imaging and elemental mapping via energy-dispersive X-ray spectroscopy (EDS), often used alongside Raman or FTIR for correlated chemical and structural analysis
• X-ray diffraction (XRD): identifies crystal structures and mineral phases, complementing the molecular/vibrational information from IR and Raman spectroscopy
• Thermal analysis (TGA, DSC): reveals how minerals behave during heating (dehydration, decomposition, phase transitions), adding context to spectroscopic observations
• Isotope ratio mass spectrometry (IRMS): provides high-precision stable isotope measurements that complement spectroscopic isotope data

Using multiple techniques on the same sample set strengthens your interpretations and helps catch errors that any single method might miss.

Environmental applications

Spectroscopy is central to environmental geochemistry, where rapid, sensitive detection of contaminants is often required.

Water quality analysis

Several spectroscopic methods are routinely used in water analysis:

• UV-Vis spectroscopy quantifies dissolved species like nitrate ($\text{NO}_3^-$), phosphate, and organic contaminants based on their absorption characteristics
• AAS and ICP-AES measure dissolved metals and trace elements (e.g., Pb, Cu, Zn, As) at regulatory-relevant concentrations
• Fluorescence spectroscopy detects dissolved organic matter and can monitor algal blooms through chlorophyll fluorescence

Portable spectrophotometers allow field-based measurements, which is valuable for screening large numbers of samples before selecting a subset for more detailed lab analysis.

Soil contamination assessment

• Handheld XRF provides rapid, non-destructive screening of heavy metals (Pb, Zn, Cu, As) in soil, making it a standard tool for site investigations
• FTIR and Raman spectroscopy identify organic contaminants (hydrocarbons, pesticides) and their degradation products in soil matrices
• Laser-induced breakdown spectroscopy (LIBS) uses a focused laser pulse to ablate and excite soil material in situ, generating an emission spectrum for elemental analysis
• Hyperspectral imaging maps contaminant distributions across large areas from airborne or drone platforms

These data feed directly into risk assessments and remediation planning.

Atmospheric pollutant detection

• IR spectroscopy monitors greenhouse gases ($\text{CO}_2$, $\text{CH}_4$, $\text{N}_2\text{O}$) and industrial pollutants
• Differential optical absorption spectroscopy (DOAS) measures trace gases ($\text{NO}_2$, $\text{SO}_2$, $\text{O}_3$) in the atmosphere using their characteristic UV-Vis absorption features
• Lidar (Light Detection and Ranging) uses laser pulses to profile aerosol distributions and atmospheric structure
• Satellite-based spectrometers (e.g., TROPOMI on Sentinel-5P) provide global-scale monitoring of atmospheric composition

Portable devices for real-time air quality monitoring are becoming more common in urban and industrial settings.

Future trends in spectroscopy

Portable spectroscopic devices

Miniaturization continues to make spectroscopic analysis more accessible outside the laboratory. Handheld XRF analyzers are already standard in exploration geology and environmental work. Portable Raman and FTIR spectrometers now support field-based mineral identification with results comparable to benchtop instruments. Improvements in battery life, processing power, and wireless data transfer are making these tools increasingly practical for remote fieldwork.

Remote sensing applications

Spectroscopy is integral to satellite and airborne remote sensing for geological mapping. Hyperspectral sensors on aircraft and satellites identify surface mineralogy over large areas, supporting mineral exploration and hazard mapping. Drone-mounted spectrometers offer high spatial resolution for surveying difficult terrain. Planetary missions (like NASA's Mars rovers Curiosity and Perseverance) carry spectroscopic instruments that analyze rock and soil compositions on other worlds, applying the same principles used in terrestrial geochemistry.

Machine learning in spectral analysis

Machine learning is transforming how spectral data are processed and interpreted. Neural networks and other algorithms can classify minerals from spectral libraries faster and more consistently than manual methods. Deep learning approaches are being applied to hyperspectral image analysis for automated geological mapping. AI-assisted peak deconvolution handles overlapping features in complex spectra, and integration of spectroscopic data with other geospatial datasets enables predictive modeling of subsurface geology and resource distribution.