RBS thickness testing (Rutherford Backscattering Spectrometry) is a non-destructive ion beam analysis technique that provides absolute, calibration-free measurement of thin film thickness, composition, and depth profiles. By bombarding a sample with megaelectron-volt helium ions and analyzing the energy spectrum of the backscattered particles, RBS achieves nanometer-scale depth resolution for films ranging from a few nanometers to several micrometers. This article explains the measurement principle, sample requirements, testing workflow, and typical applications of RBS thickness measurement.
Technical Overview
Rutherford Backscattering Spectrometry (RBS) is a quantitative surface and thin film analysis technique based on the elastic scattering of energetic ions by atomic nuclei in a solid. In a typical experiment, a collimated beam of helium ions (He2+ or He+) is accelerated to an energy between 1.5 and 3.0 MeV and directed onto the sample surface inside a high-vacuum chamber. A small fraction of these ions undergoes elastic Coulomb scattering with target nuclei and is ejected backward, where a silicon surface-barrier detector positioned at a scattering angle of approximately 160 to 170 degrees records their energy spectrum.
The thickness measurement capability of RBS relies on two physical phenomena. First, the kinematic factor, which describes the fraction of energy retained by an ion after elastic collision with a nucleus, depends only on the masses of the projectile and target atom and on the scattering geometry, allowing unambiguous identification of the elements present. Second, an ion traveling into and out of the sample loses energy continuously through electronic stopping, so the energy width of the signal from a given element directly encodes its depth distribution. Combining these two effects converts the measured energy spectrum into an absolute depth scale expressed as areal density in atoms per square centimeter.
- Absolute quantification without reference standards, traceable to fundamental physics
- Non-destructive analysis under ultra-high vacuum conditions
- Depth resolution of approximately 5 to 20 nm near the surface
- Analysis depth of roughly 1 to 2 micrometers for standard helium beams
- Detection limits of approximately 0.1 to 1 atomic percent for heavy elements in a light matrix
- Simultaneous determination of thickness, composition, and stoichiometry in a single measurement
Testing Samples
RBS thickness testing is applicable to a wide range of inorganic thin films deposited on smooth, flat, and vacuum-compatible substrates. Samples are typically cut or mounted to dimensions compatible with the sample holder, commonly between 5 x 5 mm and 20 x 20 mm, although full wafers up to 200 mm or 300 mm can be accommodated in specialized chambers. Surface roughness should ideally be below a few nanometers to preserve depth resolution.
Semiconductor and Dielectric Films
Gate oxides, silicon dioxide, silicon nitride, and high-k dielectrics such as hafnium oxide, zirconium oxide, and aluminum oxide are routinely measured. RBS is particularly valuable for verifying the equivalent oxide thickness and stoichiometry of advanced gate stacks in complementary metal-oxide-semiconductor (CMOS) process development.
Metal Films and Diffusion Barriers
Sputtered or evaporated metal layers including titanium, tantalum, tungsten, copper, platinum, and gold, as well as barrier layers such as titanium nitride and tantalum nitride, are characterized for thickness, purity, and interfacial reaction. The technique is well suited to verifying seed and adhesion layers used in copper metallization schemes.
Optical and Functional Coatings
Transparent conductive oxides such as indium tin oxide (ITO), antireflection coatings, hard protective layers such as diamond-like carbon, and nitride or oxide coatings on glass, sapphire, and stainless steel substrates can be analyzed for layer thickness and elemental composition.
Multilayer and Interface Structures
Periodic multilayers, superlattices, giant magnetoresistance (GMR) stacks, and thin-film battery or photovoltaic structures can be profiled layer by layer. RBS reveals individual layer thicknesses, interfacial mixing, and diffusion that occur during deposition or subsequent thermal processing.
Testing Items
A standard RBS thickness test delivers the following quantitative results:
- Film thickness, reported in nanometers or as areal density in atoms per square centimeter
- Elemental composition and atomic concentration ratios, including stoichiometry of compounds
- Depth profiles showing the distribution of each element as a function of depth
- Interface abruptness and evaluation of interdiffusion between adjacent layers
- Detection of heavy-element contamination or unexpected interfacial reaction products
- Verification of deposited dose for surface treatments and implanted species
Testing Methods
Measurement Principle
When a projectile ion of mass M1 and incident energy E0 scatters elastically from a stationary target nucleus of mass M2, the energy E1 of the backscattered ion is determined by the kinematic factor K, which is a function of the mass ratio and the laboratory scattering angle. Because heavier target atoms retain more energy in the scattered projectile, signals from heavy elements appear at higher energies than signals from light elements at the same depth. As ions traverse the film, they lose energy at a rate given by the stopping power, which is well tabulated for helium ions in all elements. The energy difference between particles scattered from the surface and from the film-substrate interface therefore corresponds directly to the areal density of the film, and division by the assumed or measured mass density yields the physical thickness in nanometers.
Instrumentation and Setup
A complete RBS system consists of a tandem electrostatic accelerator that generates the monoenergetic helium ion beam, magnetic or electrostatic analyzers that define the beam energy, focusing and scanning elements, a scattering chamber maintained at a pressure below 10-6 torr, a precision goniometer for sample positioning, and one or more silicon surface-barrier or passivated implanted planar silicon (PIPS) detectors. The incident beam is typically normal to the sample with a beam spot of 1 to 3 mm in diameter, and the accumulated charge is measured with a Faraday cup or current integrator to determine the total number of incident ions.
Standard Testing Workflow
- Visual and microscopic inspection of the sample, followed by mounting on the sample holder with conductive attachment where possible
- Energy calibration of the detection system using a reference standard such as a gold or bismuth film on silicon
- Pump-down of the chamber to high vacuum and alignment of the sample in the beam path
- Data acquisition with a typical incident charge of 10 to 60 microcoulombs until adequate statistics are obtained
- Spectral simulation and fitting using dedicated software such as SIMNRA, RUMP, or NDF, applying certified stopping powers and cross-section data
- Reporting of layer thicknesses, compositions, and uncertainties, including the density assumption used for unit conversion
Data Analysis and Simulation
Modern RBS analysis is performed by iterative simulation, in which a model of the layered structure is constructed and its theoretical spectrum is compared with the measured data. Layer thicknesses, compositions, and interfacial grading are adjusted until the simulated spectrum matches the experimental spectrum across the full energy range. This approach enables accurate analysis of complex multilayers where individual elemental signals overlap, and it allows rigorous estimation of confidence intervals on every fitted parameter.
Accuracy, Precision, and Limitations
Thickness results by RBS are generally accurate to within 1 to 5 percent, with the areal density being the most direct and assumption-free result. Converting areal density to linear thickness requires knowledge of the film density, and uncertainty in density is often the dominant error source. Depth resolution degrades gradually with depth due to energy straggling of the ions. Sensitivity to light elements such as boron, nitrogen, and oxygen in a heavy-element matrix is limited, and in such cases the technique may be complemented by elastic recoil detection analysis (ERDA), nuclear reaction analysis (NRA), or X-ray reflectometry.
Applications
RBS thickness testing serves critical roles across the semiconductor, data storage, optics, coating, and materials research industries:
- Qualification and process control of high-k dielectric gate stacks and metal gate layers in advanced CMOS manufacturing
- Verification of diffusion barrier thickness and integrity in copper interconnect metallization
- Thickness and compositional certification of ITO transparent electrodes for displays, touch panels, and photovoltaic devices
- Characterization of optical interference coatings, multilayer mirrors, and anti-reflective stacks
- Evaluation of interfacial reactions, alloy formation, and diffusion after annealing in thin film metallization studies
- Quality assurance of electroplated, sputtered, and evaporated layers in precision components and MEMS devices
- Academic research on thin film growth kinetics, solid-state reactions, and ion beam modification of materials
FAQ
What thickness range can RBS measure?
RBS can measure films from roughly 1 nanometer up to approximately 1 to 2 micrometers with standard helium beams. Very thin layers are reported as areal density with sub-monolayer sensitivity, while thicker layers can be probed using higher-energy beams or heavier projectiles such as lithium ions.
Is RBS a destructive test?
No. RBS is essentially non-destructive. The helium ion dose is low enough that it does not alter film properties or leave contamination, so the same sample can be retested or subjected to subsequent process steps and re-measured.
How accurate is RBS compared to other thickness measurement methods?
RBS is widely regarded as a reference method because results are derived from first-principles physics without calibration standards. Compared with optical techniques such as ellipsometry or X-ray reflectometry, RBS has somewhat lower depth resolution but provides absolute elemental identification and composition simultaneously, which makes it the preferred method for certifying unknown or complex films.
What sample conditions are required?
Samples must be vacuum compatible, reasonably flat, and large enough for the millimeter-scale beam spot. Surface roughness below a few nanometers preserves depth resolution. Insulating samples can be measured, although charge compensation or electron flood guns may be required to prevent charging artifacts.
Can RBS measure light elements such as oxygen and nitrogen?
Yes, oxygen, nitrogen, carbon, and even boron can be detected, but the sensitivity depends strongly on the matrix. Light elements are readily quantified in light-element films such as oxides on silicon, whereas their signals are weak in heavy-element matrices. For enhanced light-element sensitivity, ERDA or NRA is recommended.
Why is my RBS thickness report given in atoms per square centimeter?
Areal density is the direct physical quantity measured by RBS and is independent of any density assumption. To express the result in nanometers, the laboratory applies the nominal or independently measured film density. Reporting both values ensures transparency regarding the conversion and its associated uncertainty.