Semilab (Hungary)

Bonding technologies in heterogeneous wafer integration packaging processes requires defect identification methods beyond the capabilities of the standard ultrasonic defect inspection methods. In the present study, a special infrared defect inspection method with optional stress sensitive mode is discussed for void inspection. Artificial and unintentionally generated process induced voids were identified and qualitatively characterized in non-metallized bonded wafer. Additionally, process induced voids were also revealed in highly metallized device wafers and void position were correlated with that of observed in scanning acoustic microscopy. Stress imaging mode of the infrared method opened up an alternative characterization of defects based on their localized stress lobes.

Abstract Bandstructure engineering is a key route for thermoelectric performance enhancement. Here, 20–50% Seebeck (S) enhancement is reported for XNiCuySn half‐Heusler samples based on X = Ti. This novel electronic effect is attributed to the emergence of impurity bands of finite extent, due to the Cu dopants. Depending on the dispersion, extent, and offset with respect to the parent material, these bands are shown to enhance S to different degrees. Experimentally, this effect is controllable by the Ti content of the samples, with the addition of Zr/Hf gradually removing the enhancement. At the same time, the mobility remains largely intact, enabling power factors ≥3 mW m−1 K−2 near room temperature, increasing to ≥5 mW m−1 K−2 at high temperature. Combined with reduced thermal conductivity due to the Cu interstitials, this enables high average zT = 0.67–0.72 between 320 and 793 K for XNiCuySn compositions with ≥70% Ti. This work reveals the existence of a new route for electronic performance enhancement in n‐type XNiSn materials that are normally limited by their single carrier pocket. In principle, impurity bands can be applied to other materials and provide a new direction for further development.

Abstract This study analyzes the optical properties of TiO2 films grown via atomic layer deposition (ALD) using Tetrakis(dimethylamino)titanium with oxidizing agents such as H2O, H2O2, O3, and O2‐plasma. TiO2‐H2O exhibited Ti3+ states and oxygen vacancies characteristic of black TiO2, enhancing visible‐NIR light absorption. This effect increased with more ALD cycles due to the surface memory effect, leading to a decreased growth rate. In contrast, films grown with H2O2, O3, and O2‐plasma are more stoichiometric, has fewer defects, and show no visible‐NIR absorption due to the absence of Ti3+. TiO2 films deposited with H2O also show increased surface roughness and hydrophobicity, while films grown with other oxidizing agents exhibited higher roughness but decreased hydrophobicity. Ellipsometry and UV–vis spectroscopy confirmed that TiO2 films grown with H2O has an increased refractive index and extinction coefficient in the visible range, along with a bandgap widening due to the Moss‐Burstein effect. Conversely, films grown with other oxidizing agents showed a decreased bandgap with increasing thickness and an increased refractive index but a zero extinction coefficient below the bandgap. Mid‐IR ellipsometry measurements revealed the dielectric response, highlighting the critical role of the oxidizing agent in tailoring the properties of TiO2 films for energy and environmental applications.

Silicon epitaxy is an essential building block in the manufacturing of complementary metal-oxide semiconductor (CMOS) devices. Accurate determination of epitaxial layer thickness is indispensable for a uniform and reproducible process. In this paper, we compare thickness values of the transition zone in silicon epitaxial wafers obtained by two of Semilab’s production-compatible electrical and optical characterization techniques: Fourier-transform infrared (FTIR) reflectometry and spreading resistance profiling (SRP). We demonstrate a high correlation between transition zone thicknesses obtained from the optical modelling of FTIR reflectance spectra and SRP profiles. The dependance of transition zone thickness change on the high-temperature annealing steps is also examined. FTIR reflectometry thus offers a quick, contactless alternative for obtaining structural parameters of an epitaxial layer, and these values can be well matched to those given by SRP.

The benefits of the wide bandgap semiconductor silicon carbide (SiC) for use in high-voltage and low-loss power devices [1] are driven by fundamental material properties of high breakdown field and thermal conductivity. Ion implantation is a key doping technique in SiC device processing which requires efficient methods for dose monitoring after room temperature or elevated temperature implantation. In this work we apply photo-modulated reflectance (PMR) as an effective technique to monitor implantation dose and implantation temperature through electro-optic and thermo-optic effects influenced by residual defects in as-implanted SiC. Note, the PMR technique is widely used for implantation process monitoring in silicon device manufacturing. The present work on 4H-SiC is based on our previous experience with the PMR technique applied for Si wafers implanted with different dopants in a wide range of ion energies and ion doses [2 –4].

Abstract Tests of human brain circuit function typically require fixed equipment in lab environments. We have developed a smartphone-based platform for neurometric testing. This platform, which uses AI models like computer vision, is optimized for at-home use and produces reproducible, robust results on a battery of tests, including eyeblink conditioning, prepulse inhibition of acoustic startle response, and startle habituation. This approach provides a scalable, universal resource for quantitative assays of central nervous system function.

Room temperature photoluminescence imaging is assessed in industrial environment to detect slip line formation on 300 mm resistive silicon wafers used for Insulated Gate Bipolar Transistor (IGBT) technology and to characterize associated dislocations. The technique, with a combined use of macro and micro scale acquisitions, enables inline detection, characterization, and quantification of defects, allowing efficient process optimization and production monitoring.

This work reports on the application of the differential junction photovoltage (diff-JPV) technique to characterize conducting layers at the emitter side of silicon heterojunction (HJT) cells structures by measuring their sheet resistance (Rsh). The diff-JPV measuring principle developed for homojunction emitters can be directly adopted for HJT cell structures to accurately determine Rsh of the transparent conductive oxide (TCO) layer at the emitter side. This is the first application of diff-JPV for non-silicon layers and heterojunctions. Furthermore, the result of this simple and rapid measurement is not influenced by the resistivity of the substrate wafer, and exhibits excellent correlation to optimized four-point-probe measurement. The unique feature of the diff-JPV technique is its capability to characterize emitter type layers even with very large sheet resistance, which is hardy accessible by other contactless techniques. It was found that performing the diff- JPV measurement on wafers after the PE-CVD process to deposit the hydrogenated amorphous Si layer stack, the resulted Rsh values lie in the 80 – 240 kΩ/sq. range. The dependence of the sheet resistance on the substrate doping indicates that the measurement is related to the inversion layer formed under the emitter amorphous silicon layer stack. Since such sensitive detection and characterization of the inversion layers cannot be accomplished by other techniques, diff-JPV can be an important method for the further optimization of heterojunction interfaces.

Photo-modulated reflectance measurements provide a powerful, non-contact, non-destructive and inline-compatible method with low-cost operation for statistical process control of ion implantation steps on monitor and on product wafers. We present case studies describing how photo-modulated reflectivity measurements (PMR) can be used for ion implantation dose, fluence and tilt angle monitoring with excellent resolution, even for low-energy ion implantation processes. This is important because precise dopant and damage profile control is crucial in state-of-the art semiconductor processes utilizing shallow junctions and complex 3D doping profiles.