Secondary ion mass spectrometry (SIMS) is one of the most sensitive compositional analysis techniques available for solid materials, with detection limits reaching the parts-per-billion and in some cases parts-per-trillion range. It also detects all elements and their isotopes, including hydrogen, which the electron- and X-ray-based methods commonly used in semiconductor analysis cannot detect.
Several different instruments and analysis modes can be used to obtain different sets of information. Dynamic SIMS, or simply SIMS, is performed on a magnetic sector or quadrupole instrument and uses a continuous ion beam to profile composition as a function of depth. ToF-SIMS uses a time-of-flight analyzer and a pulsed beam to identify molecular as well as elemental species, either from the outermost atomic layers or, in dual-beam mode, across the depth of a film. The term static SIMS is often used interchangeably with ToF-SIMS, although it describes the pulsed beam regime used for surface analysis rather than the instrument itself.
The following sections outline the strengths of each technique and their ideal applications in the semiconductor industry, as well as limitations and alternative techniques for situations where neither SIMS nor ToF-SIMS provides the required information.
Similarities and differences between SIMS and ToF-SIMS
Both SIMS and ToF-SIMS work by bombarding the sample with a focused beam of primary ions under vacuum, ejecting secondary ions that are then separated by mass and counted. Both require solid, vacuum-compatible samples, and both are destructive, although surface-mode ToF-SIMS removes very little material.
The continuous beam used in SIMS, typically oxygen (O2+) for electropositive elements and cesium (Cs+) for electronegative ones, sputters material away at a steady rate. Composition recorded against sputtering time converts to a depth profile ranging from a few nanometers to several micrometers, with nanometer-scale depth resolution. Magnetic sector and quadrupole analyzers provide the dynamic range needed for quantitative trace analysis, but they transmit one mass at a time, so the species of interest must be selected before the measurement. Results are quantitative when calibrated against matrix-matched reference materials.
The time-of-flight analyzer used in ToF-SIMS records the full mass spectrum at every measurement point, across a significantly wider mass range than the analyzers used for SIMS. This allows molecular fragments and clusters to be identified rather than only the elements present, and means the data can be revisited afterwards for species that were not anticipated when the analysis was planned. The sub-micron lateral resolution also supports chemical mapping of the surface. The trade-off is quantification: ToF-SIMS results are qualitative or at best semi-quantitative.
Table 1: Comparison of SIMS and ToF-SIMS
SIMS | ToF-SIMS | |
Obtained information | Elemental and isotopic composition as a function of depth | Elemental, isotopic, and molecular composition of the surface (or as a function of depth in dual-beam mode) |
Mass analyzer | Magnetic sector or quadrupole | Time-of-flight |
Primary beam | Continuous, O2+ or Cs+ | Pulsed, usually a bismuth cluster source; separate sputter beam for depth profiling |
Information depth | Few nm to several µm | 1–2 nm in surface mode, through the film (up to ~500 nm) in dual-beam mode |
Detection limits | ppb, in favorable cases ppt | ppm to ppb |
Quantification | Quantitative with matrix-matched standards | Qualitative to semi-quantitative |
Lateral resolution | Micron scale | Sub-micron, with chemical imaging |
Typical semiconductor use | Dopant profiles, trace impurities, layer stacks | Surface contamination, defect identification, adhesion failures |
Ideal applications of SIMS
The combination of nanometer-scale depth resolution and ppb-level sensitivity makes SIMS suitable for a range of semiconductor applications:
Dopant and implant profiling: SIMS can be used to verify implant dose and energy for dopants such as boron, phosphorus, and arsenic in Si and aluminum and magnesium in SiC and GaN, and to quantify diffusion by comparing profiles taken before and after annealing. Because the total concentration of dopant atoms is measured rather than the electrically active fraction, comparing the results to a carrier concentration profile from spreading resistance profiling (SRP) or electrochemical capacitance-voltage (ECV) measurement reveals how much of the implanted dopant is electrically active.
Trace impurity analysis: SIMS quantifies trace impurities down to ppb levels, a capability that is particularly rare for light elements such as hydrogen, carbon, oxygen, and nitrogen. Background from residual gas in the vacuum system is also lower than in techniques that alternate between sputtering and analysis (e.g., XPS), as continuous sputtering leaves little time for gas to adsorb on the surface.
Isotopic tracing: Because SIMS distinguishes isotopes of the same element, an isotopically enriched reagent can be used to track which atoms a given process step contributes. For example, running a passivation or annealing step with deuterium instead of hydrogen shows how much hydrogen that step introduced, independently of the background.
Layer stacks and interfaces: A single profile through a CVD, ALD, or PVD stack resolves the thickness and composition of each layer, as well as interdiffusion and unintended interfacial layers. In high-k gate stacks, for example, this reveals the SiO2 or SiON interlayer forming between the dielectric and the substrate, and whether metal from the dielectric has migrated into it.
Failure analysis: Contamination introduced at a specific process step appears at a characteristic depth, so a SIMS profile can be used to track the problem in the process flow. Mobile ions such as sodium and potassium in dielectric layers are a common target, as they shift threshold voltages and cause device failure at trace concentrations below the reach of most other methods.
Ideal applications of ToF-SIMS
The combination of molecular-level chemical identification and monolayer surface sensitivity makes ToF-SIMS suitable for a different set of semiconductor applications:
Organic surface contamination: Photoresist and cleaning residues, plasticizers, fatty acids, silicones, and other adsorbed organics are identified by their fragment patterns rather than only by their elemental content. This kind of contamination is often only a few monolayers thick, which puts it below the reach of methods such as FTIR.
Airborne molecular contamination: Species outgassing from wafer carriers, cleanroom materials, and process chemicals can be detected directly on the wafer surface, rather than inferred from cleanroom air monitoring.
Defect and particle identification: The roughly 100 nm beam spot of ToF-SIMS makes it possible to determine the chemical identity of individual particles and other sub-micron defects located by optical or electron inspection.
Adhesion and bonding failures: When a film delaminates or a bond fails, the failure surface generally carries a thin organic layer that can be identified with ToF-SIMS to determine the responsible material or process step.
Molecular depth profiling: In dual-beam mode, ToF-SIMS profiles retain molecular information at every depth and require no advance selection of target species, enabling the identification of unknown contaminants.
Limitations and alternative techniques
SIMS cannot be used to obtain quantitative compositional information when matrix-matched reference standards are not available. In such situations, XPS can be used for (semi-)quantitative depth profiling, although there are some trade-offs: a significantly higher detection limit (~0.1–1 at%) and inability to detect hydrogen or helium. If these limitations are acceptable, XPS has the added advantage of providing chemical bonding information, which helps distinguish between the oxides or alloys of each element.
For metallic trace contamination analysis of wafer surfaces, VPD-ICP-MS is usually more practical than SIMS or ToF-SIMS, as it enables the quantification of dozens of elements simultaneously from the full surface. Similarly, GD-OES is typically preferable for profiling thick bulk materials, as it is very fast, covers large depths (up to ~150 µm), and does not require a high vacuum atmosphere.
One partner for all your semiconductor analysis needs
Measurlabs offers SIMS and ToF-SIMS for high-resolution compositional analysis of wafers, thin films, and other solid materials. More information about sample requirements, turnaround, and indicative pricing is available on the service pages:
We also offer a broad selection of other analyses for semiconductor materials and devices, including electrical characterization with methods such as spreading resistance profiling, Hall effect measurement, and four-point probe resistivity, as well as complementary compositional analyses with techniques such as XPS, GD-OES, and VPD-ICP-MS.
Use the form below to describe your testing needs, and one of our experts will get back to you in one business day.