Rutherford Backscattering Spectrometry (RBS) is a quantitative, non-destructive ion beam analysis technique used to determine the elemental composition, thickness, and depth distribution of thin films and surface layers with nanometer-scale depth resolution. By bombarding a sample with MeV helium or hydrogen ions and analyzing the energy spectrum of the backscattered particles, RBS delivers standards-free quantification of elements from beryllium to uranium. This article explains the physical principles, suitable sample types, measurable testing items, standard procedures, and industrial applications of RBS material characterization for engineers and researchers evaluating services from accredited third-party laboratories.
Technical Overview
Rutherford Backscattering Spectrometry (RBS) is one of the most widely applied ion beam analysis (IBA) techniques for surface and thin film characterization. A collimated beam of light ions, most commonly 1.0–3.0 MeV helium ions (He⁺ or He²⁺) or occasionally protons, is directed onto a specimen under high vacuum. A small fraction of these ions undergoes elastic scattering at large angles through Coulomb interaction with target nuclei, following the classic Rutherford scattering model. The energy spectrum of the backscattered ions is recorded by a silicon surface-barrier or passivated implanted planar silicon (PIPS) detector positioned at a scattering angle of typically 160°–175°.
The analytical power of RBS rests on three physical relationships. First, the kinematic factor describes the fractional energy retained by a backscattered ion after colliding with a target nucleus; because this factor depends on the mass ratio between the projectile and the target atom, the measured energy directly identifies the mass and therefore the element of the scattering atom. Second, the Rutherford scattering cross-section, which scales with the square of the atomic number of the target element, is known from first principles, allowing absolute quantitative determination of elemental amounts without calibration standards. Third, incident ions continuously lose energy through electronic stopping while traveling into and out of the sample, so the energy broadening of spectral features encodes depth information and enables concentration depth profiling.
Typical performance figures for RBS include a depth resolution of approximately 5–20 nm near the surface, a probing depth of roughly 1–2 µm for helium beams (and up to tens of micrometers for proton beams, at reduced resolution), and detection limits on the order of 10¹³–10¹⁴ atoms/cm² for heavy elements on light substrates. Sensitivity scales with Z², which means heavy elements can be detected at sub-monolayer levels in light matrices, while light elements, particularly hydrogen, are poorly detected. Quantitative composition results are typically accurate to within 1–5% in relative terms.
Several closely related ion beam methods extend the capability of standard RBS. Channeling RBS (c-RBS) exploits the guiding of ions along crystallographic axes of single crystals to assess lattice damage, crystalline quality, and lattice site occupation of impurities. Elastic Recoil Detection Analysis (ERDA or ERD) uses a heavier or comparable-mass projectile to detect hydrogen, deuterium, and other light species. Nuclear Reaction Analysis (NRA) employs specific nuclear reactions to quantify light elements such as lithium, boron, carbon, nitrogen, oxygen, and fluorine with high sensitivity. These techniques are frequently combined in modern IBA laboratories for comprehensive characterization.
Testing Samples
RBS is suitable for a broad range of solid materials, provided they are vacuum compatible and reasonably flat. The technique is especially powerful for layered structures, films thinner than a few micrometers, and any application requiring absolute, standards-free quantification. Common sample categories submitted to accredited third-party laboratories include the following.
Semiconductor Thin Films and Multilayer Structures
Gate dielectrics such as HfO₂, ZrO₂, and other high-k oxides, metal silicides, diffusion barrier layers such as Ta and TaN, metallization stacks, low-k dielectrics, and epitaxial layers are routinely analyzed to verify stoichiometry, thickness, and interfacial integrity. Ion-implanted silicon wafers are measured to confirm implanted dose and projected range.
Optical and Protective Coatings
Multilayer optical stacks, anti-reflection coatings, transparent conductive oxides such as ITO, nitride and carbide hard coatings, and diamond-like carbon films are analyzed for layer thickness, composition, and interdiffusion. RBS can resolve individual layers in complex multilayer mirrors used for extreme ultraviolet optics.
Energy and Functional Materials
Thin film photovoltaic absorbers such as CIGS, CdTe, and perovskites require precise stoichiometric control, which RBS provides quantitatively. Battery electrode coatings, solid electrolyte films, and fuel cell layers are measured for composition gradients and interfacial reactions after cycling.
Metals, Alloys, and Surface-Modified Layers
Surface-treated metals, nitrided or carburized layers, oxidation scales, corrosion products, and metallic contamination on wafers or glass are characterized for thickness and elemental depth distribution. Cultural heritage artifacts such as glazes, glass, gilding, and pigments can be examined non-destructively.
Sample Requirements
- Sample size typically ranging from a few millimeters to full wafers, depending on the beamline configuration
- Vacuum compatibility; the specimen must tolerate pressures of approximately 10⁻⁶ to 10⁻⁷ mbar
- A reasonably flat and laterally homogeneous analysis area of at least 1–2 mm, since a conventional beam spot is on the order of 1 mm
- Conductive samples are preferred; insulating samples can often be measured using electron flood guns, thin conductive coatings, or reduced beam currents to manage charging
- Total layer thickness within the accessible depth range, generally up to about 1–2 µm for helium beam analysis
- Sample history documentation, such as deposition conditions or implantation energy, which improves the accuracy of spectrum simulation
Testing Items
RBS material characterization covers a well-defined set of quantitative measurement deliverables. Depending on the sample structure and the analytical objective, typical testing items include:
- Absolute quantitative elemental composition, expressed in atomic percent or weight percent, without requiring reference standards
- Film thickness and areal density, reported in atoms/cm² and converted to linear thickness using assumed or measured density
- Elemental depth concentration profiles, revealing composition gradients and buried interfaces
- Stoichiometry verification of compound films, such as the metal-to-oxygen or metal-to-nitrogen ratio
- Interfacial diffusion, mixing, and reaction layer analysis in multilayer stacks and annealed structures
- Ion implantation dose verification and projected range confirmation in semiconductor process control
- Channeling measurements on single crystals, including minimum yield (χmin) determination for crystalline quality, damage depth profiles, and lattice site location of impurities
- Surface and interface contamination by heavy elements, such as trace metal residues on wafers
- Hydrogen and deuterium quantification performed by the complementary ERDA technique
Testing Methods
A professional RBS measurement follows a structured workflow designed to ensure quantitative accuracy and traceable reporting. Accredited third-party laboratories typically perform the following sequence of steps in accordance with recognized surface analysis terminology and practices, including the vocabulary defined in ISO 18115 and ASTM E673.
- Method selection and review of sample structure: the laboratory and client define beam species (He⁺, He²⁺, or H⁺), beam energy, detector geometry, and whether complementary ERDA, NRA, or PIXE measurements are required
- Sample preparation and mounting: specimens are cleaned if necessary, mounted on a holder, and loaded into the scattering chamber, which is evacuated to high vacuum
- Beam and detector calibration: the accelerator energy and detector electronics are calibrated using reference materials such as bismuth-implanted silicon or thin gold films on light substrates
- Data acquisition: spectra are collected while integrating a defined ion charge, typically 10–50 µC per spectrum; multiple detection geometries, such as normal incidence and grazing exit configurations, may be combined to improve depth resolution and resolve overlapping elemental signals
- Spectrum simulation and fitting: measured spectra are modeled using established analysis software such as SIMNRA, RUMP, or NDF (IBA DataFurnace), applying certified stopping powers and cross-section databases to extract composition and depth profiles
- Channeling analysis when required: crystalline samples are aligned with a goniometer along a major crystallographic axis to record aligned and random spectra for damage and strain evaluation
- Reporting and uncertainty assessment: results are delivered with layer compositions, thicknesses, areal densities, detection limits, and estimated measurement uncertainties, commonly within 1–5% relative for composition and 2–5% for areal density
Applications
RBS serves as a reference metrology technique across many industries because it provides absolute quantification that is difficult to achieve with other methods. Representative application areas include:
- Semiconductor and microelectronics manufacturing: qualification of high-k dielectrics, metal gates, silicides, barrier layers, and implanted dopant dose monitoring during process development
- Optics and photonics: thickness and composition control of multilayer interference coatings, EUV mirrors, and waveguide structures
- Photovoltaics: stoichiometry and depth profiling of CIGS, CdTe, CZTS, and perovskite absorber layers, including interdiffusion studies after annealing
- Batteries and energy storage: analysis of cathode and anode coatings, solid electrolyte interfaces, and crosstalk between stacked layers in solid-state cells
- Hard and tribological coatings: composition and gradient profiling of nitrides, carbides, and doped diamond-like carbon films
- Surface engineering and corrosion science: measurement of oxidation kinetics, scale thickness, and elemental redistribution in alloys
- Cultural heritage and archaeometry: non-destructive characterization of ancient glass, glazes, gilded surfaces, and metal artifacts
- Fundamental research: ion beam mixing, diffusion studies, thin film growth mechanisms, and radiation damage investigations in materials physics
FAQ
What is Rutherford Backscattering Spectrometry used for?
RBS is used to quantitatively determine the elemental composition, thickness, and depth distribution of elements in the near-surface region of solid materials. It is the reference method of choice for thin film stoichiometry, multilayer metrology, implant dose verification, and quantitative depth profiling without calibration standards.
Which elements can RBS detect, and can it measure hydrogen?
When using the standard helium beam, RBS can detect elements heavier than the projectile, effectively covering elements from beryllium upward, with best performance for medium and heavy elements. Hydrogen cannot be detected by conventional RBS because the scattering kinematics forbid observing backscattered ions from lighter targets; hydrogen and deuterium are instead quantified with the complementary ERDA technique.
What are the detection limits of RBS?
Because the scattering cross-section scales with the square of the atomic number, RBS is highly sensitive to heavy elements in light matrices, reaching sub-monolayer detection on the order of 10¹³–10¹⁴ atoms/cm² for elements such as gold or bismuth on silicon. In practice, detection limits correspond to roughly 0.1–1 atomic percent for heavy elements in light matrices, while light elements in heavy matrices may only be detected at levels of several percent or higher.
How deep can RBS profile into a sample?
With standard 1–3 MeV helium beams, the accessible depth is approximately 1–2 µm. Proton beams can probe tens of micrometers, but with significantly degraded depth resolution. The near-surface depth resolution is typically 5–20 nm and gradually worsens with depth due to energy straggling.
Is RBS a destructive technique?
RBS is generally non-destructive for metals, semiconductors, ceramics, and inorganic films, making it suitable for valuable samples such as heritage artifacts and product wafers. Ionically or covalently bonded insulators and certain polymers may suffer beam-induced damage under high current, in which case the beam dose is minimized.
Why is RBS considered standards-free and quantitative?
The Rutherford scattering cross-section is calculable directly from fundamental physical constants, so the yield of backscattered ions per unit collected charge can be converted into absolute areal concentration without reference samples. Only the energy calibration of the system and the stopping power database must be accurate, which is why RBS results are frequently used to calibrate other techniques.
How does RBS compare with SIMS, XPS, and EDS?
RBS offers absolute quantification and excellent depth resolution over the top micrometer but limited sensitivity for trace elements. Secondary Ion Mass Spectrometry (SIMS) provides much lower detection limits down to ppm or ppb levels but requires reference standards and suffers from matrix effects. X-ray Photoelectron Spectroscopy (XPS) is surface-sensitive within the top 10 nm and provides chemical state information. Energy Dispersive Spectroscopy (EDS) offers fast element mapping but poorer light-element and quantitative accuracy for thin films. These techniques are complementary rather than interchangeable.
What is channeling RBS and when is it needed?
Channeling RBS aligns the ion beam with a crystallographic axis of a single crystal, dramatically reducing backscattering yield from the lattice. Comparing aligned and random spectra yields the minimum yield (χmin), a sensitive indicator of crystalline quality, ion implantation damage, epitaxial layer quality, and the lattice sites occupied by impurity atoms.
What should be considered when submitting samples for RBS analysis?
Clients should provide the nominal layer structure, expected elements, approximate thicknesses, and the analytical objective, since this information significantly constrains the simulation and improves the reliability of the fitted results. Samples must be vacuum compatible, flat, and laterally uniform over the beam spot area, and the total stack thickness should remain within the accessible depth range of the selected beam.