RBS results analysis transforms raw backscattering energy spectra into accurate, standards-free measurements of elemental composition, film thickness, and depth profiles in thin films and advanced materials. This guide explains how Rutherford Backscattering Spectrometry data is acquired, interpreted, and applied across semiconductor, coating, and materials research industries. Understanding the fundamentals of RBS spectrum analysis enables engineers and scientists to extract quantitative depth information with nanometer-scale resolution.
Technical Overview
Rutherford Backscattering Spectrometry (RBS) is a quantitative ion beam analysis technique that uses a collimated beam of light ions, typically helium ions (He+ or He2+) accelerated to energies between 1.0 and 3.0 MeV, directed onto a sample surface. A small fraction of these ions undergo elastic nuclear scattering with atoms in the sample and recoil backward, where a semiconductor detector records their energies at a known scattering angle, commonly 160 to 170 degrees. The resulting energy spectrum contains three embedded pieces of information: the mass of the scattering atom, the depth at which the scattering event occurred, and the number of atoms per unit area.
The physical basis of RBS results analysis rests on three well-understood relationships. First, the kinematic factor K determines the fraction of incident energy retained by a backscattered ion after colliding with an atom of a given mass, allowing element identification from peak positions in the spectrum. Second, the Rutherford scattering cross section, which scales with the square of the atomic number of the target atom, governs the scattering probability and enables absolute quantification without reference standards. Third, the continuous energy loss of ions traveling through the sample, described by stopping powers, converts energy differences into depth scales, making RBS a inherently depth-sensitive technique.
Because scattering yields depend on known physical constants rather than instrument-dependent calibrations, RBS is widely regarded as one of the most accurate non-destructive methods for measuring elemental areal density, achieving typical accuracy of 1 to 5 percent for elements heavier than the projectile ion. Depth resolution on the order of 5 to 20 nanometers near the surface makes RBS particularly powerful for thin film characterization.
Testing Samples
Semiconductor Thin Films and Wafers
Patterned or blanket semiconductor wafers, high-k dielectric stacks, gate oxides, metal silicides, and diffusion barrier layers represent the largest category of RBS samples. These specimens benefit from RBS because the technique provides absolute dose and thickness values needed for process qualification.
Multilayer Stacks and Optical Coatings
Alternating layers of metals, oxides, and nitrides used in optical filters, mirrors, and magnetic devices can be profiled layer by layer. RBS resolves individual layers when their thickness exceeds roughly 10 nanometers and their constituent elements differ sufficiently in mass.
Implanted and Treated Materials
Ion-implanted silicon, surface-hardened metals, plasma-treated polymers with conductive coatings, and ion-mixed interfaces are analyzed to verify implanted fluence, peak concentration, and projected range.
Bulk and Ceramic Samples
Although RBS is fundamentally a near-surface technique with a probing depth of approximately 1 to 2 micrometers for standard 2 MeV helium beams, bulk ceramics, glasses, alloys, and geological specimens are routinely examined for surface composition and contamination layers.
Sample Requirements
- Samples must be vacuum compatible and free of volatile contaminants.
- A minimum beam spot region of roughly 1 to 5 millimeters is required, though micro-beam systems can reduce this substantially.
- Flat, smooth surfaces improve depth resolution; roughness broadens spectral features.
- Typical sample dimensions range from small fragments of a few millimeters to full wafer pieces of several centimeters.
- Insulating samples can be measured using electron flood guns or electron-suppressed geometries to control charging.
Testing Items
RBS results analysis delivers several distinct quantitative outputs from a single measurement:
- Elemental identification of all elements heavier than the incident ion, effectively carbon and above for helium beams.
- Areal density in atoms per square centimeter, convertible to film thickness using assumed or independently measured density.
- Quantitative composition and stoichiometry, expressed in atomic percent, without requiring calibration standards.
- Depth concentration profiles extending up to approximately 1 to 2 micrometers with nanometer-level resolution.
- Ion implantation dose verification with accuracy near 2 to 3 percent.
- Interfacial mixing, interdiffusion, and reaction layer characterization.
- Crystalline quality and lattice damage assessment when combined with ion channeling along major crystallographic axes.
- Detection of trace heavy elements in light matrices, with sensitivity improving as the square of atomic number, reaching parts-per-million levels for very heavy elements in silicon.
Hydrogen and helium cannot be detected by conventional backscattering geometry because their masses are below that of the probe ion; these light elements require elastic recoil detection (ERD or ERDA) configurations.
Testing Methods
Data Acquisition
A collimated helium ion beam, typically 2.0 MeV He2+ produced by a tandem or single-ended accelerator, rasters over a defined spot on the sample inside a high-vacuum chamber. Backscattered ions are collected by a silicon surface barrier or PIPS detector positioned at a fixed scattering angle, and pulse-height electronics convert detected ion energies into a histogram of counts versus channel number. Typical accumulated charge of 10 to 50 microcoulombs provides adequate statistics within 10 to 30 minutes. A reference spectrum from a bulk standard such as Si, Au, or TiO2 calibrates the energy scale, detector solid angle, and product of beam charge.
Spectrum Interpretation Fundamentals
Interpreting an RBS spectrum relies on reading three features systematically:
- Peak and edge positions identify elements, because heavier target atoms produce backscattered ions retaining higher energies according to the kinematic factor.
- Peak widths correspond to layer thickness, since ions scattered from progressively deeper atoms lose additional energy traveling into and out of the film.
- Peak heights and integrated areas quantify atomic concentration, following the proportionality between yield, scattering cross section, detector solid angle, collected charge, and areal density.
For a film on a substrate, the surface energy approximation converts the measured energy width into areal density using stopping powers evaluated at the incident energy at the surface and the backscattered energy at exit. This calculation, refined with iterative procedures for thicker films, yields thickness values with typical uncertainty below 5 percent.
Simulation and Fitting
Modern RBS results analysis employs spectral simulation software such as SIMNRA or RUMP, in which the analyst constructs a trial layered structure with estimated compositions and thicknesses. The software calculates the theoretical spectrum using evaluated stopping powers, Rutherford or screened cross sections, energy straggling models, and instrumental resolution, then iteratively adjusts model parameters until simulated and measured spectra converge. This approach resolves overlapping peaks, decodes multilayer stacks, and quantifies gradual interfaces that simple edge analysis cannot handle.
Channeling RBS
For crystalline samples, aligning the beam with a crystal axis reduces backscattering yield from lattice sites by up to two orders of magnitude, leaving scattering from displaced atoms, interstitials, and impurities. Comparing aligned and random spectra quantifies crystal damage, amorphization depth, and lattice site occupation, which is essential for implant anneal optimization and epitaxial quality control.
Applications
RBS results analysis supports critical decisions across many industries and research fields:
- High-k dielectric development, verifying HfO2, ZrO2, and lanthanide oxide thickness, stoichiometry, and interfacial SiO2 growth in advanced gate stacks.
- Barrier and seed layer qualification for interconnect metallization, including TaN, TiN, and WN film dose measurement.
- Ion implantation process control, providing absolute fluence verification cross-checked against SIMS profiles.
- Silicide formation studies, tracking NiSi, CoSi2, and TiSi2 reaction stoichiometry and consumed silicon thickness.
- Optical and protective coating characterization for layer thickness, composition uniformity, and interdiffusion after thermal cycling.
- Battery electrode and solid electrolyte research, quantifying protective coatings on cathode particles and lithium-containing layer stoichiometry when using heavier probe ions.
- Cultural heritage and archaeometry, non-destructively measuring glass corrosion layers, gilding thickness, and pigment composition.
- Magnetic multilayer and spintronics research, where absolute areal densities of Fe, Co, Pt, and rare-earth layers calibrate magnetic moment per atom calculations.
The standards-free quantification of RBS frequently serves as a reference method against which SIMS, XPS, ellipsometry, and XRF measurements are cross-validated.
FAQ
What elements can RBS detect, and what are the detection limits?
RBS detects all elements heavier than the incident ion, meaning carbon through uranium for helium beams. Sensitivity scales with the square of atomic number, so heavy elements such as Au, Pt, or W can be measured down to parts-per-million levels in light matrices like silicon, while light elements near the substrate mass may require concentrations of at least 0.5 to 1 atomic percent.
Why does RBS struggle with light elements in heavy matrices?
Backscattered signal from a light element appears at low energy, overlapped by the strong yield from heavier matrix atoms, and the scattering cross section is inherently small for low-Z targets. Hydrogen and helium are invisible to conventional RBS entirely and require elastic recoil detection analysis.
What depth resolution and probing depth can be expected?
Near the surface, depth resolution is typically 5 to 20 nanometers depending on detector geometry, energy resolution, and sample roughness. The maximum probing depth for a 2 MeV helium beam is approximately 1 to 2 micrometers, limited by ion stopping; higher beam energies or heavier ions extend this range.
How accurate are RBS thickness and composition results?
Thickness and areal density results are generally accurate to 1 to 5 percent because yields derive from fundamental Rutherford cross sections and stopping power databases. Composition accuracy is best for heavy elements and degrades for overlapping or low-yield signals.
Is RBS destructive, and does it require special sample preparation?
RBS is considered non-destructive for most inorganic materials; ion doses are low enough to avoid measurable sputtering or damage, although beam exposure should be minimized for radiation-sensitive organics. Samples need only be vacuum compatible, and no coating, sectioning, or chemical preparation is required.
How does RBS compare with SIMS, XPS, and AES?
RBS offers absolute quantification without standards and excellent depth accuracy but modest depth resolution and poor light-element sensitivity. SIMS provides superior trace sensitivity and depth resolution yet requires standards for quantification, while XPS and AES offer surface chemical-state information with shallower sampling depths. The techniques are complementary, and RBS often serves as the calibration anchor in multi-technique studies.
Can RBS measure density?
RBS directly measures areal density in atoms per square centimeter. Converting to physical thickness requires an assumed or independently measured mass density; conversely, combining RBS areal density with an XRR or ellipsometry thickness yields the film density itself.