Fast charging is a key enabler for the widespread adoption of lithium-ion batteries (LIBs) in various applications ranging from electric vehicles to consumer electronics. However, high charging currents amplify internal heat generation that can accelerate degradation and potentially impact battery safety. Conventional sensing methods rely on external temperature measurements, which lack the resolution to capture real-time internal temperature dynamics. This study implements a dynamic electrochemical impedance spectroscopy technique to analyze the core temperature evolution of LIBs during fast charging. Based on impedance-based diagnostics, this approach provides continuous, non-invasive, and high-fidelity tracking of internal temperature under operational extremes. The accuracy and sensitivity of this diagnostic technique are demonstrated with commercial LIBs across a wide range of operating conditions. Furthermore, this work investigates the critical interplay between the impedance-derived thermal signatures and electrochemical degradation mechanisms such as lithium plating and solid electrolyte interphase growth. By integrating real-time thermo-electrochemical monitoring, this work advances non-invasive methodologies for enhancing the performance and safety of LIBs.

Abstract Automation of FIB-SEM systems is a major point of focus and an increasingly essential feature in modern tools for the nanoscale imaging and analysis of semiconductor devices. As semiconductor geometries gradually grow more complex, the need is rising to automate progressive FIB cross-sections in order to successfully reconstruct an accurate 3D model of die level features via FIB tomography. For most tools, automation features could be purchased as complementary software. However, a more cost-effective approach for labs with budget constraints, or for labs wanting to conduct a feasibility study of automated scripts, is to utilize a set of open-source software in order to perform automated progressive FIB cross-sections and 3D Tomography. In this paper, we introduce and analyze the use of PyAutoGUI (an open-source Python library), ImageJ (open-source image processing software) and 3D Slicer (open-source 3D visualization and analysis software) to create a fully open-source suite of tools for automating a progressive FIB cross-section and tomographic reconstruction. Aside from 3D Tomography, this approach could also be used to automate in-FIB deprocessing and EDX analysis, which we are actively investigating.

This study systematically evaluates the performance of Au-Si die-attach bonding through a combination of finite element analysis (FEA), cross-sectional material characterization, visual inspection, and mechanical testing. FEA thermal simulations demonstrate that replacing the resin thermoset die-attach with Au-Si bonding significantly enhances thermal performance across key metrics. In Device A, θJA decreased from 44 °C/W to 36 °C/W with either 14 μm or 7 μm Au-Si bond-line thickness (BLT). Device B exhibited a similar improvement, with θJA dropping from 26 °C/W to 20 °C/W. Notably, both Au-Si thicknesses delivered nearly identical thermal performance, indicating that thermal benefits plateau beyond a certain BLT. Mechanical FEA further reveals that Au-Si legs experience lower shear stresses than the resin thermoset, contributing to improved stress distribution. Die shear testing confirms robust mechanical integrity, with an average bond strength of 24.85 kg, which is significantly higher than the minimum specification of 2.5 kg. Cross-sectional SEM and EDX analyses confirm a dense die-attach layer with clean interfaces and Au-Si inter-diffusion. Despite these advantages, concerns arise from the higher normal stresses, approximately 250 MPa, observed in Au-Si joints compared to 90 MPa in the resin thermoset, which may increase the risk of interfacial delamination. Additionally, blister-like anomalies observed in Device B indicate silicon compatibility issues, highlighting the need for careful die-attach material selection, particularly in legacy die node designs.

Flip chip (FC) packaging technology with copper bump interconnects has been widely used for highly integrated semiconductors to meet specific industrial requirements, contributing to dramatic body-downsizing, fine pitch, and improved electrical and thermal performance. However, FC copper bump interconnects are susceptible to mechanical reliability issues like interfacial delamination and inter-layer dielectric (ILD) passivation film cracks, while under reliability testing due to the marginality of fracture strength with the brittle mechanical behavior of BEOL film layers. For example, in a flip-chip chip-scale-package (FCCSP) device, especially one with copper bumps, an ILD crack was observed under temperature cycling (TC) loading. The crack was consistent with the new failure mode (i.e., the unique inter-layer mechanical interaction due to aggressive stacking), thus necessitating new design considerations for future devices and other multi-layer substrate package technologies. This paper comprehensively investigates this new failure mode, the root cause, and the innovative mitigation approach. This study developed the finite-element-analysis (FEA) based silicon (Si)-die level and package-level models to understand the root cause of ILD/passivation film crack risk. Additionally, the metal density transition impact on the top-metal layer (METTOP) to package metal interaction created a unique crack initiation and propagation situation to evaluate.

Distributed measurement networks, from semiconductor testing arrays to environmental sensor grids, medical diagnostic systems, and agricultural monitoring stations, face a fundamental challenge: undetected site-to-site variations that silently corrupt data integrity. These variations create systematic biases between supposedly identical measurement units, which undermine scientific reproducibility and yield. The current site-to-site variation detection methods require extensive sampling or make rigid distributional assumptions, making them impractical for many applications. We introduce a novel framework that transforms measurement data into density-based feature vectors using Kernel Density Estimation, followed by anomaly detection with Isolation Forest. To automate the final classification, we then apply a novel probabilistic thresholding method using Gaussian Mixture Models, which removes the need for user-defined anomaly proportions. This approach identifies problematic measurement sites without predefined anomaly proportions or distributional constraints. Unlike traditional methods, our method works efficiently with limited samples and adapts to diverse measurement contexts. We demonstrate its effectiveness using semiconductor multisite testing as a case study, where our approach consistently outperforms state-of-the-art methods in detection accuracy and sample efficiency when validated against industrial testing environments.

Abstract In this work, we report the growth of 300 nm c-axis oriented barium titanate (BTO) films on (001)-strontium titanate (STO) substrates via off-axis radio frequency (RF) sputtering. The as-grown films exhibit exceptional crystallinity, with a rocking curve full width at half maximum (FWHM) of ~0.03°, and atomically smooth surface morphology with a root mean square (RMS) roughness of 0.23 nm, ensuring minimal optical losses. Transmission ellipsometric characterization of the film reveals a record-high electro-optic (EO) coefficient, r51 ~ 550 pm/V that exceeds by a factor of 3 compared to the previously reported values for RF-sputtered BTO. This high-quality, industry-compatible material paves the way for the development of efficient and compact EO modulators based on hybrid BTO-on-silicon waveguides.