Papers by Keyword: Photoluminescence (PL)

Abstract: Suppressing the expansion of Single Shockley-type stacking faults (1SSFs) is critical for the growing demand of high-performance power devices. However, the underlying suppression mechanism has not yet been fully elucidated. Through proton ion implantation studies, we have established a fundamental approach by modeling this phenomenon. Carbon vacancy (V_c) generated by high-energy proton implantation are found to play a significant role in suppressing the expansion of 1SSFs.

Abstract: We have been developing the expansion–visualization–contraction (EVC) method as an inspection technique for 4H-SiC wafers, in which Shockley-type stacking faults (SSFs) are intentionally expanded by UV irradiation and subsequently visualized to identify converted dislocations that are not directly detectable by conventional PL inspection. In this study, we demonstrate a low-cost “operando” PL spectrum mapping approach for the EVC tool by using the 355-nm expansion laser as the PL excitation source and adding only a miniature spectrometer via an optical fiber, avoiding the need for an expensive hyperspectral camera.Two experiments were performed. In Experiment 1, proton-implanted and non-implanted regions on n-type 4H-SiC epilayers were compared using EVC screening and PL imaging. The proton-implanted regions exhibited narrower SSF widths, and a two-sample t-test yielded extremely small p-values, indicating a statistically significant suppression effect that remained after activation annealing. In Experiment 2, a thick epilayer wafer containing polytype inclusions was screened. PL spectrum mapping identified not only 1SSF-related emission (~420 nm) but also Frank-type components (~488 nm) and polytype-inclusion-related emission (~540 nm), revealing composite stacking faults expanded from inclusions. The results suggest that operando PL spectrum mapping can help distinguish stacking-fault types during EVC screening and potentially prevent unnecessary expansion of thermally uncontractable faults, thereby reducing yield loss.

Abstract: The yield of power electronic devices is influenced by many factors including crystal defects like stacking faults (SFs). There are different types of stacking faults but their influence on the finished device and its performance and the behavior of SF during processing is not fully understood yet. With our contribution, we shed light on the issue, showing four different optically characterized subtypes of SFs with different electrical behavior that can already be found after implantation and wafer annealing in photoluminescence (UVPL) imaging. This enables a distinction between different SF classes without the need for a finally processed device and the corresponding electrical characterization. The goal of this paper is to illustrate an alternative for subdividing SF types that would otherwise be detected as triangular defects without any distinction and to show the different effects those subclasses have on finished devices with non-destructive methods that can be used in between device manufacturing steps. These results will be used as basis for further studies to confirm the found classes and to compare them with research about the different crystal structures by spectral PL measurements. For better understanding of the effect on the finished device, the PL imaging data is correlated with I-V characteristics of trenched diodes and the defect types are evaluated on their effect on the I-V characteristic, identifying 3 defect types with detrimental influence on the reverse bias and blocking voltage while the forward bias characteristic and I-V characteristic of one type is not effected by the defects.

Abstract: Proton implantation has been reported as an effective approach for suppressing bipolar degradation in 4H-SiC; however, implantation inevitably introduces lattice damage and point defects. In this work, we investigate: (i) suppression of Shockley-type stacking fault (SSF) expansion in both the proton-implanted layer and the region beyond the implanted layer, and (ii) adverse effects associated with proton implantation. Half of an n-type 4H-SiC epitaxial wafer was implanted with protons (350 keV, 1×10¹³ cm⁻²) and annealed at 1600 °C for 30 min for dopant activation with carbon capping. SSF expansion was induced by UV laser irradiation, and photoluminescence (PL) imaging was used to quantify SSF width and observe cross sections. The proton-implanted region exhibited clearly reduced SSF expansion, with the expanded SSF width typically about 30 μm smaller than that in the non-implanted region; cross-sectional PL further confirmed that SSFs did not propagate into the near-surface implanted layer. Additional experiments with varied implantation depth and dose revealed a linear relationship between SSF width and active drift-layer thickness (defined as the drift-layer thickness minus the proton implantation depth), consistent with geometric expectations from the wafer off-cut. However, PL observations also showed anomalous SSF morphologies and evidence of dislocations, indicating that proton implantation can generate new SSF nucleation sites. Furthermore, the band-to-band PL peak intensity decreased after implantation and did not recover after the activation anneal, suggesting persistent lattice damage, including in proton-traversed regions. These results highlight a trade-off between SSF-suppression benefits and implantation-induced degradation.

Abstract: The performance and reliability of silicon carbide (SiC) devices are critically dependent on the quality of epitaxial layers which in turn are influenced by substrate properties. The accurate classification of epitaxial defects coming from substrate crystal defects and surface defects is critical since these can adversely affect device performance. In this paper, two new methods of defect characterization in substrates and epitaxial layers are presented utilizing photoluminescence (PL) spectrum and carrier lifetime. These methods can be used to study the evolution of defects from substrates to epi and to better predict Epi yields.

Abstract: CuO/TiO₂ nanocomposites were synthesized using an economical drop-casting method and subsequently irradiated with high-energy C⁺ ions at fluence levels of 1 × 10¹⁴, 1 × 10¹⁵, 1 × 10¹⁶, and 1 × 10¹⁷ ions cm⁻². While ion irradiation of metal oxide materials is well established, the systematic investigation of C⁺ ion effects on the structural and optical properties of CuO/TiO₂ nanocomposites under these specific fluence conditions has been limited. This study therefore contributes new insight into how controlled C⁺ irradiation can tailor the behavior of this composite. These un-irradiated and irradiated nanocomposites were characterized using various techniques such as Energy Dispersive X-Ray Spectroscopy (EDX), Raman Spectroscopy, Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM), Photoluminescence (PL) Spectroscopy and Diffuse Reflectance Spectroscopy (DRS) to analyze structural, morphological and optical properties of these nanocomposites. The Raman and EDX analysis confirmed the formation of pure CuO/TiO₂ nanocomposites. The SEM results represent the spherical morphology of these nanocomposites in aggregated form. PL spectra’s depicted the pure and C⁺ ions irradiated nanocomposites were the same before and after C⁺ irradiation in the Photoluminescence emission. DRS results indicated that band gap energy was decreased as the fluence rate of C⁺ ions increased up to 1 × 10¹⁷ ions cm^-2.

Abstract: We are currently developing an inspection system that will provide a low-cost means of screening prior to shipment by fully visualizing latent 1SSF (single Shockley stacking fault) defects originating from basal plane dislocations (BPDs) that cannot be detected by current defect inspection systems. The system will capture not only the defects that expand into right triangles under relatively low-level forward bias, but also the defects that expand into more serious bar-shaped 1SSFs under relatively high-level forward bias, with a particular focus on capturing TED (threading edge dislocation)-converted BPD at or below the buffer layer/substrate interface. Since these defects are known to cause forward voltage degradation during device operation, so-called "burn-in" (accelerated current stress) screening operation is currently utilized in some device manufacturers to avoid the shipping of the defective devices, but it is very time-consuming process which raises a total cost of production. The system we are developing, which can significantly reduce the screening time, has the potential to replace the "burn-in" operation.

Abstract: Correlation of X-ray topography and production line defect inspection tools has demonstrated the capability of in-line tools to differentiate between geometrically comparable basal plane slip bands (BPSB) and bar stacking faults (BSF) on 4H SiC wafers. BPSBs were found to propagate through epitaxial growth at high rates and with similar photoluminescence signatures to post-epitaxy BSFs. Molten KOH etching post-epitaxy provided evidence of distinguishing features between BPSBs and BSFs, suggesting that the defects were indeed correctly identified by in-line defect inspection tools pre-epitaxy.

Abstract: In the previous report [1], we proposed the S-EVC (Selective Expansion-Visualization-Contraction) method (Fig. 1) that effectively screens for malignant BPDs (basal plane dislocations) in the drift and buffer layers, which expand to SSFs (Shockley-type stacking faults), leading to forward voltage degradation. The method intentionally utilizes the REDG (recombination enhanced dislocation glide) mechanism by UV (ultraviolet) irradiation in wafer sorting to replace the so-called burn-in (accelerated current stress) process, which is time-consuming during mass production. In the report, triangular SSFs were examined to verify the effectiveness of the method, but they only occupy a much smaller area of the active region on the chip than bar shaped SSFs. In this study, to improve the S-EVC method to be more practical, we focused on the more serious bar shaped SSFs which have a non-negligible impact on electrical characteristics. The bar shaped SSFs are mostly expanded from TED (threading edge dislocation)-converted BPD at or below the substrate epitaxial layer interface. In PL (photoluminescence) observation by a 710 nm LPF (long-pass filter), the TED-converted BPD and the complete TED extended from the bottom of the substrate are observed as the same dark spot, but it was confirmed that both can be distinguished by the presence or absence of their SSF expansion by UV irradiation. In addition, in order to confirm the validity of the S-EVC method even on the virgin epi wafer, UV irradiation was performed on both the aluminum doped PN structured wafer and the virgin epi wafer, and the similar SSF expansion was observed. Meanwhile, the correlation between UV irradiation and forward voltage degradation was quantified using PiN diodes by comparing the glide velocity of 30°Si (g) core partials for bar shaped SSFs by UV irradiation stress with that by current stress.

Abstract: In this study, we report that the thermal treatment effects on the Raman and photoluminescence (PL) spectra of mono and few-layer MoS₂ films by annealing in the vacuum and air at 300°C, respectively. The MoS₂ film samples were prepared on silicon substrate by exfoliating from a bulk MoS₂ crystal with a micromechanical exfoliation. For characterization of structural properties of the MoS₂ films and identification of the Raman active modes, Raman spectrometer equipped with a He-Ne laser source and an optical microscope has been used. The results show that the vacuum annealing 7L MoS₂ decreases the Full Width at Half Maximum (FWHM) of the Raman active modes as E¹_2g, A_1g and the vacuum annealing 1L MoS₂ increases the PL intensity and peak energy, for 60% and 13.3meV, respectively also air annealing bilayer MoS₂ increased the PL intensity (I_A) and peak energy (E_A), respectively for 85% and 15.4 meV (300°C for 40 min). After thermal annealing (vacuum and air), we observe that the indirect bandgap of the few-layer MoS₂ was changed.