Analysis Applications of TOF-SIMS for the Semiconductor Industry

Key semiconductor processes must undergo a long R&D and verification process usually. To accelerate the development of advanced processes, more precise and accurate materials analysis is essential, especially since semiconductor processes have crossed the nanotechnology threshold. In order to continue pushing past the boundaries of Moore’s Law—to achieve “More than Moore”, we continue to create new semiconductor device structures and materials like the FinFET (Fin Field-Effect-Transistor), GAAFET (Gate-All-Around FET), Nanosheet, nanochip transistors and other innovative components. The high-frequency and high-power electronic devices that have become popular in recent years have led to Compound Semiconductors such as gallium nitride (GaN), silicon carbide (SiC), and the Fourth Generation semiconductor gallium oxide (Ga₂O₃). The development of these components and processes relies heavily on extremely advanced analytical techniques.

Instruments frequently used for structural analysis include the Scanning Electron Microscope (SEM), which is able to observe fine surface structures, the Focused Ion Beam (FIB) Microscope, which is used to view sample cross-sections, and the Transmission Electron Microscope (TEM), which has atomic-level imaging resolution, and more. When it comes to the analysis of the ion concentration of doping elements in semiconductor processes, the tools available include the Spreading Resistance Profiler (SRP) and Secondary-Ion Mass Spectrometry (SIMS), etc., for observing the distribution and depth profile of ion implantations.

There are even more types of instruments for component analysis. These are categorized based on the analysis principles on which they operate and their areas of expertise. First, there are instruments that require detection sensitivity, such as Inductively Coupled Plasma Mass Spectrometry (ICP-MS) or Magnetic-Sector SIMS. Then there are instruments that emphasize spatial resolution, Such as Energy Dispersive X-ray Spectroscopy (EDS) equipped with TEM or Electron Energy Loss Spectroscopy (EELS), which have the great advantage of being capable of microanalysis on a scale of mm (10^-6m).

However, it isn’t easy to find one analytical technique that can take both sensitivity and a high resolution into consideration at the same, let alone identify which organic or inorganic compound you are dealing with. Under most circumstances, you need to rely on a combination of multiple instruments to complete the necessary qualitative and quantitative analysis of unknown components. ToF-SIMS surface analysis technology is, however, able to take into account both sensitivity and resolution needs at the same time. Therefore, it is widely used in the analysis of organic and inorganic materials, advanced semiconductor materials research, the improvement of advanced process development and packaging processes, and even biomedical research. It is a very versatile analytical tool. In this article, we will introduce the principles behind TOF-SIMS and discuss several practical application cases to help you better understand this analysis technology.

TOF-SIMS works by bombarding the surface of the sample with energized ions, thus blasting atoms out of the target area to generate secondary ions. The secondary ions are accelerated and introduced into a Time-of-Flight Mass Spectrometry system. All these ions acquire the same kinetic energy. However, due to the differences in ion mass, they fly at different speeds within the system, and the time they take to reach the detector varies. ToF-SIMS uses this to distinguish between ions with different charge-to-mass ratios and thus complete component analysis.

The number of secondary ions obtained (the signal strength) can be used to determine the concentration of each element, and the time of ion bombardment can be converted into the longitudinal depth of detection. In this way, the depth distribution of different elements in the sample (the depth profile) can be obtained. Since TOF-SIMS collects all the bombarded ions to calculate the signal strength, it has an excellent detection limit. It can measure the concentration of elements in solid materials at the level of less than one part per million.

Figure 1 shows the distribution of common component analyzers and their sensitivity and resolution capabilities. Compared to Magnetic-Sector SIMS, which requires a probe area of at least 50um x 50um, TOF-SIMS only needs a probe area of 20um x 20um, and its probe size (spatial resolution) can reach 50nm. Coupled with a detection sensitivity that can reach the level of parts per million (ppm), it can meet both detection sensitivity and spatial resolution needs. It can also analyze trace amounts of organic and inorganic contamination. As such, TOF-SIMS has become the preferred analysis tool for microanalysis.

Figure 1.　Distribution diagram of the sensitivity and resolution of various component analyzers

TOF-SIMS works by collecting all the ion fragments bombarded from the sample being analyzed. It can detect all the elements and isotopes on the periodic table as well as obtain the sample’s Molecular Information. Typically, TOF-SIMS analysis can be divided into three categories, which we will illustrate with a few practical analysis cases.

• Mass Spectrum Analysis
• Ion Image Analysis
• Depth Profile Analysis

Silicon oil has insulating properties such as resistance to corrosion, high temperatures, dust and moisture. As such, it is widely used in electronic processes. However, special cleaning agents are required to remove it. If silicon oil is left on the surface of a Printed Circuit Board (PCB), it can lead to subsequent process failures, such as Non-wetting or non-stick over-molding issues. TOF-SIMS has the ability to analyze the organic contamination in a micro area. So, you can use the Mass Spectrum to compare the surfaces of normal and abnormal PCBs (Figure 3). As you can see, the surface of the abnormal PCB is significantly different at a charge-to-mass ratio of 147. Then, through a comparison with the data in the database, it can be concluded that this signal represents the composition of silicon oil. In this way, it is possible to identify the types and possible sources of organic contamination. This is very similar to the FT-IR analysis approach.

Figure 3.　A TOF-SIMS mass spectrum diagram of a PCB surface contaminated by silicon oil

Through multiple semiconductor processes, the TFT Array panel will form final drive electrodes, such as S/D Electrodes, Pixel Electrodes, and Gate Electrodes. The process includes the use of metal materials such as aluminum, copper and rare metals like chromium, indium, molybdenum, titanium, and tantalum, etc. The TOF-SIMS’ Ion Image process Is equipped with clear ion imaging capabilities suitable for the direct observation of the distribution of TFT Array electrodes.

Figure 4 shows the TOF-SIMS ion image of the surface electrodes of a TFT panel. It covers aluminum (Al), titanium (Ti), molybdenum (Mo), silver (Ag), silicon (Si) and other elements. Select three of the ion signals with which to create an overlay mapping image. Use different colors to represent different ions. Then observe the distribution of the various metal films and semiconductor layers on the glass substrate. In this case, there is an obvious abnormality in the distribution of residual Mo+ over a large area.

Figure 4.　TOF-SIMS ion image of TFT electrode composition

Silicon oxynitride (SiON) has good optical properties. Its refractive index is between that of Si₃N₄ and SiO₂. As such, it is widely used in optical components. In addition to being used as an anti-reflection layer and passivation layer material in solar cells, it is commonly used in the oxide layer of the metal oxide semi-field effect transistor (MOSFET) in semiconductor components with a thickness of tens of nanometers or less.

Figure 5 shows a TOF-SIMS depth profile analysis. By using the Depth Profile mode to observe the distribution curves of oxygen (O) and nitrogen (N) signals in SiON films, the thickness of the silicon oxide and silicon oxynitride layers can be obtained. The test object should be an 11 nm SiO2/SiON/Si bilayer film. Since TOF-SIMS can achieve a maximum depth resolution of Å (10-10m), it has excellent analytical capabilities when it comes to low-energy ion implantations in advanced processes or ultra-thin epitaxial structures.

Figure 5.　TOF-SIMS depth profile of a SiON film

As semiconductor components continue to shrink in size, current leakages during operation have become more serious. Therefore, the ion implantation process needs to be adjusted. In recent years, semiconductor fabs have been using higher doses of dopants and lower energy implantation parameters to develop an Ultra-Shallow Junction Process in order to improve the efficiency of components. However, the implanted layers by Ultra-Shallow Junction Process usually are very shallow, making them very difficult to analyze. 

TOF-SIMS can control low primary ion incident energies within hundreds of electron volts. As such, it is more suitable for the analysis of Ultra-Shallow Junction Doping than, for instance, Quadrupole-SIMS. 

Figure 6 shows the implantation of a BF2 dopant into a silicon wafer at 4 keV incident energy. After analyzing the depth distribution map of boron (B) using TOF-SIMS, it can be observed that the boron signal has a distinct peak 2.7 nm from the surface. It should also be noted that Boron’s Doping Profile is very clearly complete. These are all important references when adjusting implantation parameters

Figure 6.　TOF-SIMS depth distribution map of BF2 ion-implantation ultra-shallow junction

3D sensors are gradually becoming more prevalent in our daily lives. It is an integral part of everything from the common smartphone’s facial recognition feature, Virtual Reality (VR), and Augmented Reality (AR) to the Metaverse. It is required for interacting with the virtual world ecosystem. It can also be found in drones, industrial robots, self-driving cars, security monitoring, remote medical care and other fields. 3D sensors are like human eyes, and the high-power VCSEL (Vertical Cavity Surface Emitting Laser) used to generate infrared light is a key component of the sensing system. The VCSEL generates tens of thousands of light spots at a time and projects them on the object being analyzed. The image sensor receives the reflective array from the light spots. These signals contain depth information that is converted by a processor into the profile of the surface of the object being examined. Like IC and LED components, 3D sensors are a key focus of development in the compound semiconductor industry.

Figure 7 (left) shows the analysis of principal VCSEL components using the TOF-SIMS Depth Profile function. You can clearly see the distribution, depth and thickness of the indium (In), aluminum (Al), gallium (Ga), and arsenic (As) from the sample surface to the GaAs substrate. You can also see the p-DBR (p-type Distributed Bragg Reflector), the MQW (Multiple-Quantum Well), the n-DBR (n-type Distributed Bragg Reflector), and the compositional variations of the Al_xGa_1-xAs structure.

Figure 7 (right) is a 3D image conversion result that combined with the TOF-SIMS depth profile and the Al and Ga 3D Overlay Image. The staggered distribution of red and green in the diagram details the spatial distribution of Al and Ga to within tens of nanometers per layer. After combining the spatial and depth resolution data, TOF-SIMS is able to delineate the three-dimensional image on the composition analysis, thus providing a more intuitive observation of the spatial distribution of substances.

Figure 7.　TOF-SIMS depth profile (left) and 3D image (right) of the principal component of the VCSEL structure

TOF-SIMS is able to analyze all kinds of conductors, semiconductors, and insulation materials. It also has a “full periodic table” mass spectrum of element analysis features and ppm level detection sensitivity. In addition, TOF-SIMS has a lateral spatial resolution of up to 50nm, and a depth analysis resolution of up to 0.1nm, making it ideal for analyzing ultra-shallow junctions, multilayer film structures, and trace amounts of organic and inorganic contamination. It can make up for the limitations of XPS and FT-IR analysis techniques.

TOF-SIMS also has the advantage of being able to take into account both sensitivity and resolution at the same time. Its many strengths have led TOF-SIMS to become more and more widely used in many types of analysis in recent years. Therefore, in order to meet the analysis needs of our many customers, MA-tek is introducing the latest M6 Plus model, which is expected to be compatible with the current magnetically biased and quadrupole mass spectrometers. This will help us provide the industry with more complete analysis services.