Non-destructive inspection and analysis techniques are crucial for quality assessment and defect analysis in various industries. They enable for screening and monitoring of parts and products without alteration or impact, facilitating the exploration of material interactions and defect formation. With increasing complexity in microelectronic technologies, high reliability, robustness and thus, successful failure analysis is essential. Machine learning (ML) approaches have been developed and evaluated for the analysis of acoustic echo signals and time-resolved thermal responses for assessing their ability for defect detection. In the present paper different ML architectures were evaluated, including 1D and 2D convolutional neural networks (CNNs) after transforming time domain data into the spectral- and wavelet domains. Results showed that 2D CNNs processing data in wavelet domain representation performed best, however at the expense of additional computational effort. Furthermore, ML-based analysis was explored for lock-in thermography to detect and locate defects in the axial dimension based on thermal emissions. While promising, further research is needed to fully realize its potential.

Through silicon via (TSV) is the most promising technology for vertical system integration and novel 3-D chip architectures. Reliability and quality assessment necessary for process development, root-cause analysis, and manufacturing require appropriate nondestructive testing techniques to achieve high processing yield and reliability. This paper presents two advanced approaches for nondestructive localization of defects in TSVs. The first method investigated here was lock-in thermal imaging by thermography and photoemission microscopy to inspect and localize electrical shorts within the 3-D geometry of the TSV insulation liner. In addition, filling defects in the TSVs have been inspected using acoustic microscopy in the gigahertz frequency band in time and spectral domains.

Microelectronic packaging is crucial for protecting, powering, and interconnecting semiconductor chips, playing a critical role in the functionality and reliability of electronic devices. With the growth in complexity and miniaturization of these products, the implementation of efficient inspection techniques becomes crucial in preventing failures that may result in device malfunctions. This review paper examines the progress made in utilizing Scanning Acoustic Microscopy (SAM) to assess the structural integrity of microelectronic systems within the broader field of Nondestructive Evaluation/Testing (NDE/T) methods. With an exclusive emphasis on SAM, we point out SAM technological advancements in multi-die stacking, Through Silicon Vias (TSV), and hybrid bonding inspection that improve inspection sensitivity and resolution required to be prepared for upcoming challenges accompanying 3D- and heterogeneous integration architectures. Some of these approaches compromise the depth of inspection for the benefit of lateral resolution, while others do not sacrifice the in-depth range of evaluation. These developments are of the utmost importance in addressing the substantial obstacles associated with examining microelectronic packages, facilitating the early detection of potential failures, and enhancing the reliability and robustness of semiconductor devices. Furthermore, our discussion consists of the fundamental principles and practical approaches of SAM. It also examines recent investigations that integrate SAM with machine learning concepts and the application of deep learning models in order to automate defect detection and characterization, thus substantially augmenting the efficiency of microelectronic package assessments.

The excitation and propagation of acoustic waves as employed in scanning acoustic microscopy follows highly complex phenomena which commonly cannot be visualized experimentally. The current paper describes the application of modelling and numerical simulation of these phenomena in structures like complex microelectronic samples or acoustic lenses. The design of acoustic lenses and their active elements have been optimized based on the numerical studies performed here. Simulation based visualization of the wave propagation and energy conversion in a Through Silicon Via with different defects at 1 GHz acoustic frequency supported the understanding and interpretation of the received acoustic signals.