This paper investigates the effects of three ageing factors (chemical, humidity, and temperature) and their interactions on the physical properties and degradation of silicone sealant used in microelectronic applications. The thermal degradation of silicone sealants was investigated by exposing samples to temperatures in the range of 150 up to 175 °C. Also, a set of samples were aged at 40 °C in a salt spray set-up with 100 % humidity in a salty atmosphere. Results showed detectable changes in the FTIR spectra of aged specimen as compared with the as-received sample. In all accelerated testing conditions, peak intensities decreased with ageing time, inferring that that the surface characteristics of the sealant is affected by ageing. Shear test results showed that with increasing the ageing time, the maximum shear stress in most cases has decreased in all ageing conditions. Also, it appears that samples with longer ageing times have experienced more elongation before failure. Results also show that salt spraying of specimens is associated with a decrease in the mechanical properties of the sealant, indicating the deleterious implications of ionic contaminations for the mechanical properties of samples.

This study introduces a training protocol utilizing Convolutional Neural Networks (CNNs) and Confocal Scanning Acoustic Microscopy (CSAM) imaging techniques to classify Power Quad Flat No-leads (PQFN) package delamination. The investigation involves empty PQFN packages with varied substrate metallizations subjected to thermal cycling. Four delamination classes were labeled: Die-pad delamination (Class-A), Bond-pad delamination (Class-B), both Die-pad and Bond-pad delamination (Class-C), and No delamination (Class-D). Due to data imbalance, additional randomness was introduced for distribution balancing. Residual Networks (ResNet-18) based CNN model was selected for classification. Five-fold cross-validation assessed overfitting performance concerning input data size, image resolution, and batch size. The ResNet-18 prediction performance was evaluated using precision and recall metrics, with the model achieving average precision and recall scores of 0.86/1 and 0.83/1, respectively. Additionally, a comparison of delamination among different substrate metallizations was presented with Ag and NiPdAu indicating significant delamination compared to bare Cu substrate. This study pioneers the integration of CNNs with CSAM imaging for package defect detection and classification, laying the groundwork for future research to address the complex interplay of multiple failure mechanisms in functional packages.

Solder fatigue is a key failure mode in the electronic industry. Monitoring the actual degradation of the solder under real-time conditions in any application would be extremely beneficial. In this chapter, we describe the combination of experimental material characterization with numerical finite element (FE) simulations to obtain a prognostics and health monitoring (PHM) methodology for LED drivers used in outdoor lighting applications. Experimental characterization of a new type of solder is described. A FE model is created of a typical component in electronic drivers. The calculated damage level and the collected life data correlate together and form a model for predicting the lifetime of the drivers at certain user condition. The developed PHM methodology helps in identifying and reporting the failure of the driver in real time or can be used for predicting the actual remaining useful life (RUL).

A significant challenge in the implementation of health monitoring systems for estimating the health state of devices is the lack of accurate information about design details. This challenge is particularly prominent in the field of power electronics, where both IC designers and converter designers are often hesitant to share information about their designs. Addressing this issue, this paper introduces a novel AI-driven digital twin modeling methodology that enables the detection and classification of failures in power semiconductors, particularly Wide Band Gap semiconductors. By employing AI-based system identification techniques, this method offers a noninvasive approach to health monitoring of power switches with high resolution, even while operating under real conditions. The proposed method has been validated by simulating wire bond failure in a SiC power MOSFET using MATLAB SIMULINK, and the results demonstrate its accuracy.

Semiconductor devices are commonly encapsulated with Epoxy-based Moulding Compounds (EMC) to form an electronic package. EMC typically occupies a large volume within a package, and thus, governs its thermomechanical behaviour. When exposed to high temperatures (150°C and above), electronic packages predominantly show oxidation of the outer layer of EMC. Oxidized EMC exhibits notably different material properties, resulting in a modified deformation pattern of a thermally aged package under varying thermal loads. As the oxidation layer grows in thickness, its mechanical properties also evolve, indicating distinct phases of the oxidized material at different stages of thermal ageing. Reflecting these changes (i.e., the current state of degradation) into a Finite Element (FE) model-based analysis can provide better insights into failure prediction and component reliability. It requires updating the geometry and material behaviour as a function of ageing. This paper presents a systematic procedure to build a continuously updated physics-based Digital Twin of a thermally aged flip-chip package that can represent intermediate oxidation stages. First, experimental measurements are carried out to quantify the growth of the oxidation thickness at 150°C and a diffusion-dominant mathematical model is proposed. Then, an accurate geometry of the test package is prepared with a parametric outer layer from all exposed sides of EMC to represent the oxidized layer at different stages of thermal ageing. Next, the experimental characterization of a few partially oxidized EMC specimens is done, and analytical methods are utilized to extract the thermomechanical properties of the oxidized EMC at different stages of ageing. Experimental warpage data of aged test packages are utilized to verify the defined material-model parameters that represent curing shrinkage, thermal expansion, glass transition, and corresponding elasticity moduli of the oxidized EMC at select stages of ageing. Then, a workflow to establish continuity in the material model is presented. Finally, the developed Digital Twin is utilized for an FE analysis to study the change in the trend of out-of-plane package deformations as a function of several stages of EMC oxidation.

This paper introduces an ontology-based Digital Twin (DT) architecture for the lighting industry, integrating simulation models, data analytics, and visualization to represent luminaires. The ontology standardizes luminaire components, facilitating interoperability with design tools. The calculated ontology-level metrics suggest mid-level complexity with Size Of Vocabulary (SOV) at 37, Edge-to-Node Ratio (ENR) at 0.865, Tree Impurity (TIP) at 0, and Entropy Of Graph (EOG) at 2.61. A use case explores the utility of the ontology in the design phase across two different geographical locations, assessing environmental adaptability. The ontology captures opto-thermo-electric interactions, providing insights into luminaire performance. Results from inflating the DT and conducting simulations align with existing literature, indicating a degradation of around 12% over 8 years on the radiant flux. This ontology, up to the authors’ knowledge, is the first formal definition for the lighting industry, aiming to encompass the entire luminaire lifecycle. The current focus is on design and operational phases, with potential future enhancements to include real-time monitoring for performance evaluation and predictive maintenance. This work contributes to luminaire analysis and supports the development of sustainable lighting solutions in the industry.

In this chapter, the past and present situations of the reliability domain are discussed. As of today, most industries are in the transfer from test-to-pass approaches to more advanced strategies. These strategies currently are to determine the reliability capability by applying (where possible) the test-to-failure concept, extending reliability qualification conformance tests beyond the required levels, and assessing any physical or electrical degradation of a product during those tests. New concepts are under investigation that focus on the physics of degradation. Progress in the area of reliability will never stop so as to reduce the amount (and cost) of product field failures.

Model order reduction techniques are developed and utilized to make numerical simulations more efficient. The use of Reduce Order Models (ROM) also enables data exchange with external parties without disclosing the sensitive information present in a Full-Order Model (FOM). It is crucial to optimize for both the efficiency and accuracy of a ROM to keep a minimal deviation from the FOM. The complexity of a ROM-based simulation depends on the definition of the ROM as well as its connection with the remaining FOM. This paper investigates the effect of different ROM-FOM interface definitions for a test case consisting of an electronic package-on-PCB assembly. A virtual Design of Experiments (DoE) was carried out with a total of 41 cases considering three different locations and up to four different constraint equations for the ROM-FOM interface. The effect on the accuracy and time-efficiency of the ROM-based thermomechanical simulations are compared with respect to the full Finite Element (FE) model. The deformable configuration for the interface generally showed the most accurate results, while the rigid configuration was the most efficient across the board. The beam configuration did not always follow an expected trend based on the order of elasticity values of the assigned materials. Based on the deformation results and the time associated with ROM generation and use-pass, multiple optimal solutions from the DoE are discussed.

The increasing awareness of environmental concerns and sustainability underlines the importance of energy-efficient systems, renewable energy technologies, electric vehicles, and smart grids. Hence, stringent constraints and safety regulations have been prompted to meet reliability standards in power electronics. This chapter provides a comprehensive outlook on the current state of power semiconductor devices, field-critical applications, dominant degradation mechanism (chip-related and package-related), and the emerging measurement techniques for reliability/condition monitoring. This chapter delves into the underlying physics behind each reliability measurement method reviewed. A comparative summary of cost, complexity, online monitoring capability, accuracy, and intrusiveness is provided to enable readers to make informed decisions about the measurement methods. This chapter emphasizes the significance of early fault detection through online monitoring, as it can effectively reduce system downtime for seamless non-interruptive operation.

This study explores a novel approach to monitor the spectral emission of LEDs by estimating the spectral power distribution from the spectral sensor responses during an accelerated aging experiment. Two methods for reconstructing the actual LED spectra from sensor responses are presented and tested, one solely requires sensor datasheet information and the other uses a full spectral characterization of the sensor's spectral sensitivities. The reconstruction results show that a spectral sensor can provide accurate spectral estimates even after severe LED degradation. Only for an LED that suffered a phosphor crack, affecting its spatial radiation characteristics, limited ability to estimate the true spectral power distribution without prior assumptions about the spectral changes must be reported. Overall, the use of a spectral sensor, even without detailed characterization of the sensor itself, allows for an accurate monitoring of the true emission of LEDs, with a maximum radiometric error of 0.73 %, a maximum colormetric error of 0.0017Δ u'v' and a maximum spectral nRMSE error of 0.0097 compared to a spectroradiometric measurement. This advance holds great promise for improving lighting technology, particularly in applications that require constant radiometric output and stable color.

The introduction of silicon carbide(SiC) has reduced the superiority of traditional silicon-based power module pack-aging strategies. As packaging strategies become increasingly complex, classical thermal modelling tools often prove inadequate in balancing efficiency with accuracy. Integrating these tools with machine learning (ML) can significantly enhance their application potential. This discussion commences by addressing the pressing issues in thermal modelling of SiC modules, specifically the challenges associated with multiple heat sources and heat spreading. During the design stage, ML models can swiftly simulate the thermal response of various packaging strategies, aiding engineers in eliminating ineffective options. In the monitoring phase, the employment of a digital twin enables a deeper investigation into degradation phenomena. This article reviews the current status and explores the potential applications of ML in thermal modelling of SiC power modules.

Board level reliability can be of high interest for automotive electronic components when exposed to vibration-prone environments. However, the absence of an industry standard for board level vibration testing poses several challenges in establishing a well-characterized test setup. One of the challenges is that automotive applications can induce abnormal stresses on components that can lead to early failures in the field. Such loading conditions are not always covered in the current board level vibration test methods. This paper aims to correlate the stresses from automotive modules to board levels by measuring the printed circuit board (PCB) vibration spectrum. Firstly, the study compares and assesses several module board level vibration measurement units, such as LASER Doppler Vibrometer (LDV), strain gauges, and accelerometers. Experiments and simulations show that LDV enables good correlation with Micro-electro Mechanical Systems (MEMS) accelerometers. Secondly, the module-board interaction unveils insights into several module design features that impact the PCB vibration response and solder joint interconnect reliability. These findings underscore the necessity for the user to correctly validate the reliability of packages beyond board level testing, i.e., at the module level. This reliability test approach enables the translation of reliability test results from the lab to the field life of components once built in the final application equipment.

In this article, we provide a comprehensive review of defect formation at the atomic level in interfaces and gate oxides, focusing on two primary defect types: interface traps and oxide traps. We summarize the current theoretical models and experimental observations related to these intrinsic defects, as they critically impact device performance and reliability. By integrating theoretical insights with experimental data, this review provides a thorough understanding of the atomic-scale interactions that govern defect formation.

Industrial assembly lines are the heartbeat of modern manufacturing, where precision and efficiency are paramount. This paper introduces a novel hybrid Explainable artificial intelligence (XAI) approach to enhance monitoring and analysis in industrial assembly. By fusing the power of vision anomaly detection models with the clarity of the gradient tree boosting algorithm, this framework not only boosts defect detection accuracy but also provides transparent, actionable insights. This synergy transforms how operators and engineers interact with AI, fostering trust and enhancing operational excellence.

Silver sintering offers a promising landscape for Pb-free die attachment in electronics packaging. However, the sintered interface properties are highly process-dependent and deviate from bulk silver properties. Conventional measurement methods do not adequately capture the die-attach application geometry. Hence, this study introduces a novel methodology for characterizing thin bond-line interfaces with high-conductive materials. The transient heat flux impedance ΔZth(t, Δx) was measured between two thermally sensitive devices interconnected using pressureless Ag-sintering material. A correction factor was derived, based on thermal half-space principles, to account for non-uniform heat spreading over the die-attach interface. Experimental findings estimate an effective conductivity of ∼115W/mk for the pressureless Ag-sintered interface. The measurement results were validated by measuring a SAC305 soldered interface, which exhibited ∼55 W/mK, and a non-conductive epoxy interface of ∼2.5 W/mK. Voids on the die-attach layer, resulting from material processing, were identified to influence the interface thermal behavior. An uncertainty analysis was further discussed, emphasizing equipment tolerances, measurement sensitivity, and geometrical and thermal anisotropies. The article concludes with a comparative summary of the proposed methodology against conventional methods, highlighting differences in working principle, thickness range, measurement parameters, and their advantages and limitations.

Power MOSFET dies in the automotive industry are becoming larger (>5 × 5 mm) and thinner (<50 µm) to meet high-performance and lifetime requirements. Ensuring the mechanical robustness of these large ultrathin chips is crucial for reliable electronic devices and high-throughput packaging processes. The high aspect ratio and advanced chip designs incorporating trench technology present significant challenges in semiconductor assembly, packaging, and testing. This paper introduces an experimental front-end strategy aimed at strengthening the front side (FS) of large ultrathin dies using various die-top systems. Industry-equivalent 50 µm thick dummy power MOSFET dies were fabricated to evaluate the efficacy of different chip designs and materials, such as polyimide (PI), in mitigating fracture risks. Fabrication-induced stresses and warpage in the device layers were measured using a thin-film stress measurement tool. Additionally, the FS strength of the ultrathin dies was assessed using the three-point bending method, with the resulting data analyzed via two-parameter Weibull distribution plots. Results demonstrated that the deposition of 5 µm PI on the nitride die topside significantly increased die strength from 339 MPa to 760 MPa, with 5 µm PI proving more effective for die strengthening than 10 µm. The interaction between the metal-trench layer and the die was found to be critical to the robustness of ultrathin dies, influenced by the pattern and layout of the trenches. Die-top metallization designs, such as meandering patterns, showed promising improvements in die strength compared to standard designs. A proposed chip layout aims to maximize PI coverage for clip-bonded products on the die topside, leveraging its strengthening effect. The study also demonstrated that dummy reference chips can facilitate rapid and straightforward evaluation of extensive design experiments to identify robust chip designs.

Solder joint failure is one of the most common board-level failure modes in electronic components. It is crucial for a next-generation reliability assessment method to have an in-situ health monitoring system in place to evaluate the current state of degradation. This is achieved by specialized embedded sensors and processing the data on the edge. This study focuses on monitoring the mechanical degradation of package-to-PCB solder interconnects of a WLCSP using a high-resolution piezoresistive sensor. First, a measurement workflow was set up to optimize and significantly improve the sensor readout time. Then, utilizing a design of experiments, the test specimens were subjected to certain combinations of mechanical and thermal loads in a four-point bending setup. Temperature-coupled mechanical loading showed a greater impact on the resulting stress pattern compared to that of a superposition of the corresponding individual purely thermal and mechanical load configurations. Finally, the specimens were tested under a purely mechanical load until failure, and a correlation between the recorded stress pattern and the initiation and propagation of a crack was established.

Prognostic monitoring of power quad flat no-lead (PQFN) packages with four distinct silver pastes, each varying in material composition (pure-Ag and resin-reinforced hybridAg) and sintering processes (pressure-assisted and pressureless), was investigated in this study. The PQFN packages with silver sintered die-attach materials were subjected to thermal cycling tests (?55 ° C to 150 ° C), and the performance degradation was evaluated based on the following metrics: 1) electrical ON-state resistance RDSon monitored periodically at specific thermal cycling intervals and 2) transient thermal impedance Zth(t = 0.1 s) monitored online during thermal cycling. These measurements were further validated using acoustic microscopy imaging and cross-sectional inspection. The pressureless Ag-sintering material demonstrated comparable performance to pressure-assisted Agsintering, with a dense microstructure, and consistent electrical and stable thermal performance. Whereas the pressureless resinreinforced hybrid-Ag material exhibited degradation with a relative increase of 33% in RDSon, 38% in Zth(t = 0.1 s), and 67% delamination of the die-attach interface over 1000 cycles. These findings suggest that pressureless Ag-sintering may offer a viable alternative to pressure-assisted methods for lead (Pb)- free die-attachments, while resin-reinforced hybrid-Ag requires further development for improved thermomechanical reliability..

Low-temperature sintering technology of Ag nanoparticles is widely used in high-power electronic device packaging. Utilizing simulation methods to comprehend and control the microstructure and properties of Ag sintering materials has emerged as a prominent research area. This work uses the General Form PDE module in COMSOL to simplify the implementation of the phase field method. A two-particle model was established to explore the effects of different particle sizes and temperatures on the sintering neck of Ag nanoparticles. The two-particle model was expanded to multi-particle models and 3D models flexibly. This work presents the potential and limitations of these phase-field models, preparing for further multiphysics analysis and optimization of Ag sintering materials.

The ability to accurately predict the reliability and lifetime of electronics is of great importance to the industry. The failure of the solder joint is of particular interest for these predictions, because of their susceptibility to failure under thermo-mechanical stress. However, the experimental or even conventional simulation techniques employed to estimate the lifetime of a solder joint are often too expensive or time consuming to be of practical use. Therefore, this work introduces a physics-informed Long Short-Term Memory (LSTM) to predict the plastic strain in the critical area of the solder joint. The predicted values are in agreement with the values gained from finite elements, thereby demonstrating the advantage of applying the proposed methodology.