Abstract This manuscript addresses a challenging semiconductor wafer defect—manifesting as Ni-capped silicon spike structures at the wafer edge—whose origin and nature eluded standard inline inspection tools due to its subsurface location. To resolve this, a cross-functional team from contamination-free manufacturing (CFM), etch, and failure analysis (FA) collaborated to develop an innovative etch-assisted hybrid inspection methodology. This approach physically exposed buried anomalies, enabling advanced imaging and elemental analysis. The team integrated conventional CFM with targeted dry etching and high-resolution lab-based techniques—including Scanning Electron Microscopy (SEM), Focused Ion Beam (FIB), Transmission Electron Microscopy (TEM), Scanning Transmission Electron Microscopy (STEM), and Energy-Dispersive X-ray Spectroscopy (EDS)—to confirm nickel diffusion from the wafer backside to the frontside, forming metallic spikes beneath critical device structures. This interdisciplinary strategy not only pinpointed the root cause—backside-to-frontside Ni diffusion—but also underscored the importance of adaptive workflows and crossdomain collaboration in identifying complex, latent defects in modern semiconductor manufacturing. The methodology offers broader implications for developing robust contamination control strategies and enhancing defect detection capabilities.

Abstract Bitmaps are a useful tool that can make SRAM yield learning vehicles preferable to their logic counterparts. Being able to see the exact failing bits enables a unique ability to separate systematic signatures, perform quick failure analysis, and even investigate parametric sensitivities. This paper details the successful use of bitmapping, combined with automatic pattern recognition, to isolate signatures for a missing spacer defect and help identify and resolve their two unique root causes.

ABSTRACT Azoles are widely employed as copper corrosion inhibitors in the semiconductor industry but pose environmental concerns due to their poor biodegradability and toxicity toward nitrifying microorganisms. Their occurrence in effluents highlights the need for effective treatment strategies. Here, four advanced oxidation processes (AOPs) – Fenton, UV/Fenton, UV/H2O2, and UV/persulfate processes – were systematically evaluated for the degradation of 1,2,4-triazole (TZ), a representative azole, in both lab-prepared and actual semiconductor wastewater collected immediately downstream of chemical mechanical planarization (CMP) manufacturing processes. While all three UV-assisted processes achieved complete degradation of 0.72 mM TZ from lab-prepared wastewater within 30 min, the UV/persulfate process exhibited the fastest kinetics (pseudo-first-order rate constant of 0.38 min−1). However, in this semiconductor wastewater with 0.33 mM TZ, the UV/Fenton process exhibited the fastest kinetics and the highest TOC removal rate, underscoring the strong influence of matrix constituents. Given the high volume of this semiconductor wastewater, the potential of implementing a pre-concentration step was evaluated. Adsorption studies revealed that Fe(II)-zeolite exhibited the highest affinity for TZ, though its overall capacity was too low for practical pre-concentration. These findings provide practical guidance for optimising oxidation-based treatment strategies specifically for azole-containing semiconductor wastewater.

Electrical discharge plasmas rapidly degrade short-chain(SC) per-and polyfluoroalkyl substances (PFAS) in the presence of sacrificialsurfactants. These surfactants facilitate the transport of PFAS tothe plasma-liquid interface through electrostatic and hydrophobicinteractions, where PFAS and surfactants are ultimately degraded.This study investigates the degradation of perfluorobutanesulfonate(PFBS) by nonthermal plasma, both in the absence and presence of quaternaryalkyl trimethylammonium surfactants: octyl-, dodecyl-, and hexadecyltrimethylammonium bromide (C8TAB, C12TAB, and C16TAB, respectively).Advanced analytical techniques, including targeted, suspect, and nontargetedliquid chromatography-tandem mass spectrometry (LC-MS/MS), gas chromatography(GC-MS headspace), and ion chromatography (IC), were employed to identifythe degradation byproducts of PFBS and the surfactants. Suspect andnontargeted analyses (NTA) revealed the formation of shorter-chainperfluorocarboxylic acids (PFCAs), products from H/F and OH/F exchangereactions, fluorinated ketones, fluorinated alcohols, unsaturatedfluorinated compounds, and shorter-chain perfluorosulfonic acids (PFSAs);however, the last has been ascribed to contamination in the PFBS reagent.Notably, the degradation of PFBS, both in the absence and presenceof surfactants, produced a nearly identical set of byproducts. Basedon these newly identified byproducts, a series of degradation pathwayshas been proposed, involving solvated electrons and OH radicals asthe primary reactive species. This study provides critical insightsinto the complex mechanisms and pathways of PFBS degradation duringnonthermal plasma treatment. The findings have significant implicationsfor optimizing plasma technologies and other PFAS treatment methods,with the proposed pathways expected to be relevant for PFAS degradationby technologies that utilize oxidative and nonoxidative species.

Abstract Classification-based learning has become a cornerstone of deep neural networks, particularly in few-shot learning, where accurate similarity metrics, such as Hamming distance, are critical. However, conventional architectures require retrieving class vectors from a physically separated memory for Hamming distance calculations, incurring significant energy penalties due to data movement. This inefficiency poses a challenge to scalability and overall system performance. In-memory computing, which eliminates data transfers between processing and memory units, is increasingly recognized as a promising solution to this von Neumann bottleneck. Analog content-addressable memory (CAM)-based systems address this issue by embedding class vectors directly within CAM cells. However, their reliance on sensing circuits, particularly analog-to-digital converters (ADCs), introduces scalability and reliability challenges. The limited sense margin of ADCs, combined with device variability, further constrains array size and performance. These issues are exacerbated with emerging non-volatile memory devices like ferroelectric field-effect transistors (FeFETs). In this work, we present an innovative FeFET-based digital Logic-in-Memory (LiM) XOR cell, fabricated using GlobalFoundries’ 28 nm SLPe technology, eliminating the need for ADCs. Our 2T FeFET-based XOR cell offers a fully digital, compact, and energy-efficient solution that is robust to device variability and scalable for large systems. Applied to Hamming distance calculations for 4096-bit class vectors, our design achieves a 23-fold reduction in energy consumption, a 3-fold decrease in latency, and a 14-fold reduction in silicon footprint compared to state-of-the-art solutions. Crucially, our FeFET-based architecture demonstrates an unprecedented efficiency of 2337 Gsamples/(s·W·mm2 ), a 300-fold improvement over conventional designs, offering a unique competitive advantage where energy efficiency, reliability, and performance trade-offs have long been a concern. Our efficiency gains, while maintaining the maturity of digital computing, align with the industry’s demand for energy-efficient, scalable, and reliable in-memory computing. Furthermore, digital LiM supports the broader goal of energy-efficient AI hardware without sacrificing reliability, making it highly appealing to researchers and industries focused on sustainable computing.