Day 1 – Sunday April 02 2017
08:30
Short Course C-Wohlrabe
08:30 |
Statistical Planning and Evaluation of Experiments for Quality and Reliability in Electronics |
10:30
Coffee break
10:45
12:45
Lunch
13:45
Short Course C-Bailey
13:45 |
Co-design/simulation for advanced packaging – current status and future challenge |
15:45
Coffee break
16:00
18:00
End of courses
Day 2 – Monday April 03 2017
09:00
Welcome by Dr. Robert Franke, Dresden Economic Development Department
09:10
Industry keynotes (in Bellevue)
09:10 |
Predictive Modeling and New Reliability Frontier, Milena Vujosevic, Ph.D., Intel Corporation, Santa Clara, USA |
09:40 |
Reliability Calculations in Semiconductors – a Short History of Accomplishments, Craig Hillman, DfR Solutions, USA |
10:10 |
Presentation by Dr. Norbert Thyssen being cancelled, replaced by Achievement Award ceremony |
10:40
Break
11:00
S. 1 Technical Keynotes (in Bellevue)
12:30
Lunch
14:00
S. 2 Package level reliability characterisation (in Bellevue)
S. 3 Modelling and Optimisation of MEMS structures (in Leipzig)
S. 4 Thermal Behavioral Modeling (in Elbterrasse)
15:50
Break
16:20
S. 5 Advanced Experiments (in Bellevue)
S. 6 Thermal Design and Analysis (in Leipzig)
S. 7 Solid State Lighting I (in Elbterrasse)
17:50
End of 1st day technical sessions
18:00
Leave hotel lobby for dinner downtown at 18:30
Day 3 – Tuesday April 04 2017
09:00
S. 8 IC level stress simulation and characterisation (in Bellevue)
S. 9 Molecular Dynamics (in Leipzig)
S. 10 Thermal Characterization (in Elbterrasse)
10:10
Break
S. 11 Posters : Multi-Physics Issues in Microelectronics
S. 12 Posters : Thermal and Thermo-Mechanical Issues in Microelectronics
12:15
Lunch
13:30
S. 13 Solid State Lighting II (in Elbterrasse)
S. 14 MEMS, Sensors, Model Order Reduction (in Bellevue)
S. 15 Simulation and test applied to NEW technologies (in Leipzig)
15:20
Break
15:40
S. 16 Prognostics & Health Monitoring (in Bellevue)
S. 17 Characterization, experiments and modelling (in Leipzig)
S. 18 Thermo-mechanical Analysis (in Elbterrasse)
17:10
Exhibitor and Sponsor special session (in Bellevue)
17:10 |
AMIC – Jürgen Keller |
17:18 |
AMITRONICS – Lukas Berbuer |
17:26 |
CADFEM |
17:34 |
CWM / FRT Thomas Fries |
17:42 |
DANTEC – Roland Wahler |
17:50 |
Dynardo – Roland Niemeier |
17:58 |
EPoSS – Petra Weiler |
18:06 |
NANOTEST – Mohamed Abo Ras |
18:15
Cocktail party at venue
20:00
EuroSimE Organisers and Technical Committees meeting
Day 4 – Wednesday April 05 2017
09:00
S. 19 New Developments in Thermal Management and Solid State Lighting (in Bellevue)
10:00
Break
10:30
S. 20 Solder joint fatigue characterisation and simulation (in Bellevue)
S. 21 3D-Printing & Power Electronics Applications (in Leipzig)
12:20
Lunch, Awards (Sven Rzepka)
13:30
Electronics Packaging Community readiness for Heterogeneous Integration (HI) Challenges in 2020 and beyond (in Bellevue)
13:30 |
Dr Christopher Bailey, Greenwich University: co-design Challenges, real integration, adding value |
13:45 |
Dr XueJun Fan, Lamar University: co-design Challenges, realistic integration to determine real performance vs lifetime |
14:00 |
Dr Jürgen Wolf, Fraunhofer Institute: research to push HI in the next 10 years |
14:15 |
Dr Jean Trewhella, Global Foundries: manufacturing challenges to introduce innovative HI packaging |
14:30 |
Dr Milena Vujosevic, Intel: new devices reliability, device level specification, upgrade new standards |
15:00
End of conference
Session 1 |
Technical Keynotes |
11:00 Bellevue |
Monday April 03 2017 |
Chaired by C. Bailey, P. Rodgers |
11:00 |
30mn |
Keynote presentation – Air-coupled PMUT at 100 kHz with PZT Active Layer and Residual Stresses: Multiphysics Model and Experimental Validation Gianluca Massimino 1, Luca D’Alessandro 1, Francesco Procopio 2, Raffaele Ardito 1, Marco Ferrera 2, Alberto Corigliano 1 1 Politecnico di Milano 2 STMicroelectronics abstract In this paper a complete Multiphysics modelling via the Finite Element Method (FEM) of an air-coupled Piezoelectric Micromachined Ultrasonic Transducer (PMUT) is described, with its experimental validation related to the mechanical and acoustic responses.The numerical model takes into account the presence of fabrication induced residual stresses, which determine a non-linear initial deformed configuration of the diaphragm and a substantial frequency shift associated with the fundamental eigenmode of the vibrating system.The complete simulation of the device’s behaviour is obtained considering multiple coupling between different fields: electro-mechanical coupling for the piezoelectric model, thermo-acoustic-structural interaction and thermo-acoustic-pressure interaction for the waves propagation in the surrounding fluid.The model gives a realistic estimation of the fundamental frequency and of the PMUT’s quality factor through the adoption of large deformation analyses and by means of a proper modelling of the air, considering its thermo-viscous properties, that induce the power dissipation in the so-called boundary layer at the fluid-structure interface.The results of the numerical multi-physics model are compared with experimental ones in terms of the initial static pre-deflection, of the membrane center vertical displacement frequency spectrum and of the sound intensity at 3.5 cm on the vertical direction of the axisymmetric axis of the diaphragm. |
11:30 |
30mn |
Keynote presentation – Electronics Cooling: The Frontrunner in the race for Frontloaded CFD John Parry, Mentor Graphics, East Molesey, UK abstract CFD has become a widely accepted branch of CAE, now routinely used in the development of products. In many industries, such as aerospace and automotive, the use of CFD is often consigned to late design activities to confirm product performance is acceptable, and to perform minor optimizations, for example door mirror shape, looking at the impact on vortices in its wake, and their buffeting on the window, which adds to cabin noise. A challenge for all companies is to find ways to do two things. First, find ways of moving simulation up the design flow, so that it can make a greater contribution to the product creation process, and in doing so improve product performance, reduce design time and cost, and reduce the cost of the final product. Second, find ways to democratize the use of simulation so that its benefits can be brought to a wider audience than analyst specialists.Many approaches are being explored to address either or both of these issues by a number of vendors. In electronics cooling, due to a mix of factors, CFD rapidly migrated from being done in parallel to physical prototyping post design, to widespread use from concept design onwards today. This presentation looks at some of the factors that distinguish this application others, the technologies that are being brought to bear, and how these factors are now influencing the wider simulation community in niche applications like automotive lighting which is being revolutionized by the use of LEDs, and why this is so. |
12:00 |
30mn |
Keynote presentation – Board Level Reliability Assessment of Consumer Components for Automotive Use by Simulation and Sophisticated Optical Deformation Analyses R. Dudek 1, M. Hildebrand 1, S. Rzepka 1, J. Beintner 1, R.Döring 2, L. Scheiter 2, B. Seiler 2, Th. Fries 3, R. W. Ortmann 4 1 Fraunhofer ENAS, Micro Materials Center, Germany 2 CWM GmbH, Chemnitz, Germany 3 FRT GmbH, Bergisch-Gladbach, Germany 4 Continental Automotive France SAS, Toulouse, France abstract The development of automotive electronics (AE) towards autonomous driving applications generates various challenges,in particular also on the reliable functionality. In various cases dedicated automotive grade components are lackingand consumer components have to be used instead, which hardly fulfil automotive standards. Some of the reliabilitychallenges are therefore thermo-mechanical in nature. Some example issues, which are related to consumer electronics(CE) packaging, in particular MEMS-packaging, are given in the paper. Main focus is laid on the development of FEsimulationbased evaluation methodologies accompanied by experimental characterization methods, in particular withregard to solder fatigue. It is shown that secondary effects such as, for example, intrinsic warpage of the componentand of the circuit board, that are system related effects, can play an important role in AE application. A newly developedoptical multi-sensor metrology method is presented for the thermo-mechanical deformation measurement of electroniccomponents and systems for different size and resolution ranges. Thermally induced intrinsic warpage of circuitboards and components were analyzed both by means of the method. It was found that significant intrinsic deformationsand warpages can occur and should be considered directly in AE system design and indirectly when evaluation stressrisks, e.g. for solder fatigue. |
Session 2 |
Package level reliability characterisation |
14:00 Bellevue |
Monday April 03 2017 |
Chaired by A. Dasgupta, B. Vandevelde |
14:00 |
30mn |
Keynote presentation – Effects of residual stresses on cracking and delamination risks of an avionics MEMS pressure sensor Auersperg, Juergen 1, Auerswald, Ellen 2, Collet, Christian 3, Dean, Thierry 3, Vogel, Dietmar 2, Winkler, Thomas 4, Rzepka, Sven 2 1 Fraunhofer ENAS, Chemnitz, Germany and Berliner Nanotest and Design GmbH, Berlin/Chemnitz, Germany 2 Fraunhofer ENAS, Chemnitz, Germany 3 THALES RESEARCH 4 Berliner Nanotest and Design GmbH, Berlin/Chemnitz, Germany abstract Silicon based pressure sensors often take advantage of piezo-resistive gages which are normally embedded by multiple silicon oxide and silicon nitride layers where gold lines form a Wheatstone bridge. As a result of manufacturing – stepwise deposition of multiple layers – significant layer residual stresses occur in the GPa range in tension and compression. But also anodic bonding of the silicon MEMS device on usually glassy substrates results in additional initial stresses. Especially in avionics MEMS applications such stresses by far exceed the stresses arising under sensor operation and determine the major risks for cracking and delamination. Furthermore, those stresses could lead to a signal drift of the overall sensor over a long period of time – another important trustworthiness risk. |
14:30 |
20mn |
Analysing delamination in ASIC packages Pećanac, G., Silber, C., Kosbi, K., Vollmer, L., Wiedenmann, M., von Bargen, T., Fischer, A., Robert Bosch GmbH, 72762 Reutlingen, Germany |
14:50 |
20mn |
Characterization of epoxy based highly filled die attach materials in microelectronics I. Maus 1, C. Liebl 2, M. Fink 1, D.-K. Vu 2, M. Hartung 1, H. Preu 1, K.M.B. Jansen 3, B. Michel 4, B. Wunderle 5, L. Weiss 2 1 Infineon Technologies AG, Wernerwerkstr. 2, 93049 Regensburg, Germany 2 Infineon Technologies AG, Am Campeon 1-12, 85579 Neubiberg, Germany 3 TU Delft, Industrial Design Engineering, The Netherlands 4 Fraunhofer ENAS, Chemnitz Germany 5 TU Chemnitz, Germany, Fraunhofer ENAS, Chemnitz Germany |
15:10 |
20mn |
Modelling of Thermal Aging of Moulding Compound by using an Equivalent Layer Assumption Bingbing Zhang 1, Alexander Lion 1, Michael Johlitz 1, Leo Ernst 2, K.M.B. Jansen 3, Duc−Khoi Vu 4, Laurens Weiss 4 1 Institute of Mechanics, Universität der Bundeswehr München, Germany 2 Emeritus Professor of Delft University of Technology, Delft, Netherlands 3 Department of Design Engineering, Delft University of Technology, Netherlands 4 Infineon Technologies AG, Neubiberg, Germany abstract Currently, the use of electronic components for automotive and aerospace applications is developing quickly. More and more components will be exposed to harsh environments, such as high temperature and high moisture. In general, this high temperature is always above the glass transition temperature (Tg) of the encapsulation material, being Epoxy Molding Compound (EMC). EMC exposed to high temperature could induce reliability problems of components due to changes of its material properties accompanied with volume shrinkage. Therefore, the characterization and modelling of the aging process in EMCs during high-temperature conditions has become an important issue. In our previous work [1], the characterization methods to obtain the material properties as function of aging time were discussed and introduced. The present work focuses on a new and efficient method to model the impact of the aging process of EMCs on the warpage and the stress state of a package using FEM simulation. Here, an “equivalent layer” model, which includes a fully oxidized layer and an unaged core, is applied to simplify the modelling of the thermal aging effects. The current thickness of the “equivalent oxidized layer” is obtained by combining the experimental results and numerical analyses of properly chosen samples. At the end of the paper the aging shrinkage is estimated by using the equivalent thickness concept |
15:30 |
20mn |
Simulation of delamination initiation and subsequent propagation using cohesive zones Maofen Zhang 1, Daoguo Yang 1, Leo Ernst 2, Bingbing Zhang 3 1 Guilin University of Electronic Technology, China 2 Emeritus Professor of Delft University of Technology, Delft, The Netherlands 3 Institute of Mechanics, Universität der Bundeswehr München, Germany abstract Delamination is one of the main problems in electronic packaging due to the relatively weak adhesion strength of interfaces. Delamination failure includes two processes: crack initiation and crack propagation. Various experimental setups and theories have been built to study these processes. Among these studies the cohesive zone method being implemented in various finite element packages is becoming a popular tool for crack propagation modelling. In our previous work [1-4] the efficient use of cohesive zone modelling of the delamination propagation process of pre-cracked samples was discussed. Here also an appropriate method to establish the cohesive zone parameters was given. Among this a fitting method to establish the critical energy release rate through cohesive zone modelling was presented.Although the modelling of delamination propagation for packages with assumed pre-cracks at various spots now is quite feasible, in the thermal-mechanical designing of packages the initiation of delamination (without assumed pre-cracks) is much more important. Therefore, the present work primarily focuses on applying the cohesive zone method for the modelling of the initiation of delamination, followed by the subsequent delamination propagation. The paper will primarily investigate the crack initiation (without pre-crack) for a chosen interface. The influences of mesh size, number of loading increments and the critical stress values are investigated and discussed in detail. The computing time is considered and compared for various settings, such as to overcome convergence problems. The present study will help to make proper choices for future correct and economically feasible simulations of delamination initiation and subsequent propagation. |
Session 3 |
Modelling and Optimisation of MEMS structures |
14:00 Leipzig |
Monday April 03 2017 |
Chaired by D. Andersson, N. Iwamoto |
14:00 |
30mn |
Keynote presentation – A model of the electric field in a one-dimensional micro-mirror array and electromechanical simulations and optimization in a single cell Duy Duc Nguyen 1, Nguyen Nhat Binh Trinh 1, Michel Lenczner 1, Frédéric Zamkotsian 2, Scott Cogan 1 1 FEMTO-ST, University of Bourgogne Franche-Comté, CNRS, UTBM, Besançon, France 2 LAM-CNRS, Marseille, France abstract This paper reports recent progress in modeling and simulation of a one-dimensional Micro-Mirror Array actuated by an electrostatic force. We present results obtained through numerical simulations of a single cell: the analysis and the optimization of the pull-in voltage and the analysis of the bounces of the mirror in contact with the base when it is subjected to a voltage exceeding the pull-in voltage. For the array, a model has been derived for the electrostatic field using a multi-scale modeling technique. The model is detailed together with simulation results. |
14:30 |
20mn |
Design and Simulation of a 2-Bit Distributed S-Band MEMS Phase Shifter Noor Amalina Ramli, Tughrul Arslan, University of Edinburgh, Edinburgh, United Kingdom abstract In this paper, a wideband distributed coplanar waveguide (CPW) phase shifter has been designed for S-band operation. The phase shifter is developed on a high impedance CPW transmission line periodically loaded with microelectro-mechanical system (MEMS) shunt switches. To reduce the insertion loss due to the transmission line, a 525 μm high resistivity silicon substrate (ρ=10 kΩ∙cm) and 2 μm thick aluminium center conductor are used. A large metal-insulator-metal (MIM) capacitor is implemented on the CPW center conductor between the two bits to separate the DC actuation voltage during operation. There are 41 MEMS shunt switches utilised in the design. The simulated reflection coefficients are less than −10 dB from 2 to 4 GHz for all 4 states (0⁰, 90⁰, 180⁰ and 270⁰). The simulated average insertion loss at 2.45 GHz is 1.69 dB. The size of the phase shifter has been optimized by employing 60⁰ mitered CPW bends to result in overall size of 11.62 mm × 28.35 mm. The simulated pull-in voltage of the fixed-fixed beams implemented in the design is 9.68 V. The true time delay characteristic of the phase shifter makes it suitable for use in smart antenna applications, radar systems and ultra-wideband (UWB) microwave imaging. |
14:50 |
20mn |
Model Evaluation and Improvement for Commercially Available Silicon Carbide Power MOSFETs Andrii Stefanskyi, Lukasz Starzak, Andrzej Napieralski, Lodz University of Technology, Lodz, Poland abstract Silicon carbide MOSFETs coming from every manufacturer selling on the market have been simulated using models provided by the respective manufacturers. Simulation results have been compared to datasheet characteristics. Discrepancies were identified and their possible causes have been investigated. This has been complemented with an analysis of each model structure.For model improvement, several approaches have been considered: modification of model parameter values (parameter re-extraction); modification of model circuit equations (e.g. additional coefficients, changes to implemented laws); using a completely different model structure. |
15:10 |
20mn |
Multi-physics based system simulations for magnetic sensors Helmut Köck 1, Gernot Binder 1, Frank Heinrichs 1, Gregor Wautischer 2, Florian Bruckner 2, Dieter Süss 2 1 Infineon Technologies Austria AG, Siemensstrasse 2, 9500 Villach, Austria 2 Christian Doppler Laboratory of Advanced Magnetic Sensing and Materials,Institute of Solid State Physics, Vienna University abstract Today’s magnetic sensor applications demand a detailed understanding of all relevant system components. This work presents a holistic simulation methodology to account for the close interaction between ferromagnetic target objects, magnetic circuit designs and specific sensor characteristics of relevant Hall and xMR technologies. The core element of this simulation approach is a proprietary developed finite element / boundary element based magnetic simulation environment capable to accurately calculate isotropic as well as anisotropic permanent magnets in a ferromagnetic target wheel setup. Available simulation features are superior in terms of field accuracy and computational cost compared to commercially available tools. Incorporating the magnetic simulation core in a holistic simulation design flow offers the integration of application driven key parameters like target wheel rotation or air gap variation and ultimately translates the results to electrical signals. Target wheels, package details including magnet design, chip layout and sensor specific properties are parametrized and investigated in detail.The numerical results of a target wheel optimization based on the multi-physics simulation approach presented in this work are eventually compared to actual measurements. The presented results clearly highlight the potential to support future product developments in the field of magnetic sensors. |
15:30 |
20mn |
Investigation and optimization of microfluidic flow-through chambers for homogeneous reaction space Péter Pálovics, Márta Rencz, Budapest University of Technology and Economics, Department of Electron Devices, Budapest, Hungary H-1117 abstract Many microfluidic applications use flow-through chambers in order to store [1], measure [2] or react [3] chemicals. These chambers are usually used with paused flow, but the novel solutions enable continuous flow inside the chamber, which assures much better throughput. In [4] Computational Fluid Dynamics simulations were presented for a given channel-chamber structure, where the chambers have a volume of V=1.1 µl. Our paper presents a case-study with CFD in which the flow is examined in flow-through chambers with different geometries. The goal of our study is to find a simple channel-chamber geometry which has lower diversity of flow velocity in the chamber. In this study we compare our results with the original case presented in [4].[1] Paul J Hung, Philip J Lee, Poorya Sabounchi, Nima Aghdam, Robert Lin, and Luke P Lee. A novel high aspect ratio microfluidic design to provide a stable and uniform microenvironment for cell growth in a high throughput mammalian cell culture array. Lab on a Chip, 5(1):44–48, 2005.[2] Ferenc Ender, Diána Weiser, András Vitéz, Gábor Sallai, Márton Németh, and László Poppe. In-situ measurement of magnetic nanoparticle quantity in a microfluidic device. Microsystem Technologies, pages 1–12, 2015.[3] Dominique M Roberge, Bertin Zimmermann, Fabio Rainone, Michael Gottsponer, Markus Eyholzer, and Norbert Kockmann. Microreactor technology and continuous processes in the fine chemical and pharmaceutical industry: is the revolution underway? Organic Process Research |
Session 4 |
Thermal Behavioral Modeling |
14:00 Elbterrasse |
Monday April 03 2017 |
Chaired by J. Parry, M. Janicki |
14:00 |
30mn |
Keynote presentation – Fin‐Tube and Plate Heat Exchangers – Evaluation of Transient Performance Ilja Belov 1, Andreas Nordh 2, Kent Salomonsson 1, Peter Leisner 3 1 Jönköping University, School of Engineering, Jönköping, Sweden 2 ZiGrid AB, Nora, Sweden 3 SP Technical Research Institute of Sweden, Borås, Sweden abstract A methodology for evaluation of transient performance of, and comparison between plate heat exchanger and plate-fin-and-tube heat exchanger was developed and realized, including experiment and 3-D simulation. Heat transfer from water to a gas medium was addressed. The heated gas volume was the same for both heat exchanger designs. This was achieved by placing the plate-fin-and-tube heat exchanger into enclosure. The volume average temperature of the gas as function of time was computed. Estimated material cost for the studied designs was at least seven times lower than for the stainless steel plate heat exchanger. The performance of the selected plate-fin-and-tube heat exchanger design was found comparable to the plate heat exchanger, when both fin and tube materials were set to Al, and the enclosure was a light-weight thermal insulator. Transient behavior of the studied heat exchangers should be of interest for micro-grid applications, but also for thermal management in electronic cabinets and data centers. |
14:30 |
20mn |
Characterization of Thermal Conductivity in Polymer Composite Heat Exchanger Parts Ismail Darawsheh, Antoine Diana, Peter Rodgers, Valerie Eveloy, Fahad Almaskari, Department of Mechanical Engineering, The Petroleum Institute, Abu Dhabi, United Arab Emirates abstract Fiber-reinforced, injection-molded polymer composite materials can provide heat exchanger heat transfer rates comparable to those of metallic materials. However, the relationship between fiber orientation and thermal conductivity, and its effects on the heat transfer rate need to be investigated. In this study, a methodology to determine the anisotropic thermal conductivity of an injection-molded commercially-available, thermally-enhanced polymer composite, based on numerical simulation combined with experimentation is presented. The injection molding process is numerically modeled to predict fiber orientation. The filler characteristics of injection-molded polymer composite parts are experimentally determined to derive the composite material thermal conductivity distribution using Nielsen semi-empirical model. This methodology is applied to a heat exchanger unit air channel geometry, that is virtually manufactured using either injection molding or a combination of injection molding and machining. The numerically predicted thermal conductivity values range from approximately 14 W/m.K to 16 W/m.K, depending on geometric location and manufacturing process. These values are underpredicted by up to 18% compared with laser flash measurements on physical prototypes manufactured using a combination of injection molding and machining, and are lower than the vendor-reported effective thermal conductivity (i.e., 19-21 W/m.K). |
14:50 |
20mn |
Thermal Design of Monocular Vision System used in Automotive Application Gamal Refai-Ahmed 1, Hoa Do 1, Arun P Raghupathy 2, Rubab Kadam 2, Jay Gillis 2 1 Xilinx Inc 2 Electronic Cooling Solutions Inc abstract Advanced driver assistance systems such as Forward Collision Warning or Lane Departure Warning on automobiles rely heavily on a vision-based system. One such system is a monocular camera that is mounted inside the cabin on the center of the windshield. This module has a temperature sensitive FPGA and Image Sensor. The module sees extreme temperature environments due to high solar radiation in the dashboard section of the vehicle. It is important to keep such a system under specified temperatures for reliable functioning and safety. This study presents a detailed methodology to analyze and design thermal management solutions for the automotive camera by factoring in the transient nature of its immediate surroundings. The transient environment that accounts for the temperature drop from the high initial ambient temperatures includes factors such as movement of the vehicle and/or turning on the air conditioner. Experiments help understand the boundary conditions required for the computational thermal analysis and appropriate experimental measurements are taken to define the relevant boundary conditions of the system. |
15:10 |
20mn |
Transient Thermal Simulation of High Power LED and its Challenges Sanchit Tandon 1, E Liu 2, Thomas Zahner 3, Sebastian Besold 3, Wolfgang Kalb 3, Gordon Elger 2 1 Technische Hochschule Ingolstadt, Germany and OSRAM Opto Semiconductors GmbH, Regensburg, Germany 2 Technische Hochschule Ingolstadt, Germany 3 OSRAM Opto Semiconductors GmbH, Regensburg, Germany abstract Transient thermal analysis (TTA) is widely used to measure the transient thermal impedance (Zth) and the thermal resistance of LEDs because reliability and lifetime of LED depends critically on junction temperature. To predict up-front in the product development process the lifetime of LED modules, calibrated finite element (FE) models are used. In this paper a FE-model for a family of high power LED is developed, i.e. different number of LED dies on ceramic sub-mounts of different sizes and calibrated to the Zth measurements. Based on the CAD data for the selected LED module (two LED dies on ceramic carrier), different modern FE tools (ANSYS, Comsol and Flo-EFD) are used for transient FE simulation and benchmarked. All tools deliver appropriate results when best practice FE modelling is applied i.e. mesh quality, correct boundary condition, material data and contact resistances. To model the Zth (t) measurement correctly, the suited approach of thermal boundary condition is investigated and a temperature boundary condition is proven as correct, practical and numerical efficient approach. The specific effect of heat generated in the converter of white LEDs on the transient impedance curve is revealed and investigated. Afterwards one FE-tool is coupled with the commercial optimizer OptiSLang. Based on available material data the FE model of the 2-chip LED module is calibrated to the experimental measured transient impedance curve. The calibrated model parameters are used to simulate the Zth (t) curves of another high power LED module of this family. It was found that the simulated curves matched the experimental Zth (t) curves of the LED modules. This validates the calibrated material properties for this entire LED family. |
15:30 |
20mn |
Thermal Simulation of Hybrid Circuits with Variable Heat Transfer Coefficient Tomasz Torzewicz, Agnieszka Samson, Tomasz Raszkowski, Artur Sobczak, Marcin Janicki, Mariusz Zubert, Andrzej Napieralski, Lodz University of Technology abstract This paper demonstrates, based on a practical example of a test hybrid circuit, the importance of proper modeling of the heat transfer coefficient dependence on the surface temperature rise and fluid velocity in air cooled electronic systems. Hybrid circuits usually have large surface area and consequently important temperature gradients could occur in them, hence the local values of the heat transfer coefficient might differ considerably. In order to show the significance of the problem, dynamic thermal responses of the test circuit were measured for various dissipated power levels and air velocities. The measurement results allowed then the generation of compact thermal models. Owing to the fact that these models take into account the variation of heat transfer coefficient value with cooling conditions, it was possible to increase significantly the accuracy of thermal simulations. |
Session 5 |
Advanced Experiments |
16:20 Bellevue |
Monday April 03 2017 |
Chaired by J. Zaal, V. Rochus |
16:20 |
30mn |
Keynote presentation – Rapid testing method for interface crack analysis of an adhesive bonded joint using an electrodynamic shaker C. Vernier 1, M. Dressler 1, H.-P. Seebich 1, B. Wunderle 2 1 Robert Bosch GmbH 2 Technische Universität Chemnitz |
16:50 |
20mn |
Vibration Investigation in Power Module Busbar Design Matt Packwood, Daohui Li, Xiapoing Dai, Steve Jones, Dynex Semiconductor, Lincoln, United Kingdom abstract The structural response of power module busbars under harmonic loads is considered using finite element method simulation. Simulation parameters and boundary conditions are considered in terms of their effect on the accuracy of simulation results with relation to real life application. Vibrational analysis was found to be highly sensitive to geometry preparation and constraint in terms of both mechanical stress magnitude and location. Overall the simulation process was determined to be a valuable tool in terms of reducing time/cost involved in iteratively designing power module busbars to withstand mechanical vibration with relation to life time application as well as ultrasonic weld process. |
17:10 |
20mn |
A Laser-Speckle-Photometry based Non-Destructive Method for Measuring Stresses Conditions in Direct-Copper-Bonded Ceramics for Power Electronic Application Stefan Muench, Mike Roellig, Ulana Cikalova, Beatrice Bendjus, Lili Chen, Shohag Sudip, Fraunhofer Institute for Ceramic Technologies and Systems IKTS, Dresden, Germany |
17:30 |
20mn |
Advanced study about the determination of elastic-plastic properties of thin films by micro bending tests Uwe Zschenderlein 1, Karthik Suresh 1, Mario Baum 2, Marie Weißbach 2, Bernhard Wunderle 1 1 University of Technology Chemnitz, Germany 2 Fraunhofer Institute ENAS, Chemnitz, Germany |
Session 6 |
Thermal Design and Analysis |
16:20 Leipzig |
Monday April 03 2017 |
Chaired by I. Belov, A.P. Raghupathy |
16:20 |
30mn |
Keynote presentation – The numerical analysis of heat transfer at nanoscale using full and reduced DPL models Tomasz Raszkowski, Agnieszka Samson, Mariusz Zubert, Marcin Janicki, Andrzej Napieralski, Lodz University of Technology, Lodz, Poland abstract This paper includes the analyses related to the thermal model order reduction. The simplified, one-dimensional structure is investigated. The reduction technique based on the moment matching is employed. For this purpose, the Krylov-subspace-based model order reduction method is used. The generation of the reduced Dual-Phase-Lag heat transfer model is carried out using the Arnoldi algorithm. The temperature distributions obtained using both the full- and reduced-order Dual-Phase-Lag models are analyzed and carefully compared. Moreover, the analysis of the level of the relative error of outputs generated using reduced models in relation to results yielded using full-order model is included. Finally, the computation times are compared and conclusions are included. |
16:50 |
20mn |
Transfer function domination approximation with compensation for successive network reduction Márton Németh, Péter Pálovics, András Poppe, Budapest University of Technology and Economics Department of Electron Devices abstract In this paper we present a direct computational method for calculating thermal transfer impedances between two separate locations of a given physical structure aimed at the implementation into a field-solver based on the SUNRED (SUccessive Node REDuction) algorithm. The paper describes a further reduction method in complex frequency space based on Hankel singular values. We tested the method on a typical MCPCB assembled LED structure in 2D. With that model we were able to comparise the results from analytical calculations and from that approximations. The results show that the approximation method is very accurate, the calculated error were below 2 %. |
17:10 |
20mn |
Simulation investigations for the comparison of standard and highly robust AlCu thick metal tracks Raj Sekar Sethu 1, Verena Hein 2, Marco Erstling 2, Kirsten Weide-Zaage 3 1 X-FAB Semiconductor Foundries AG, Kuching, Malaysia 2 X-FAB Semiconductor Foundries AG, Erfurt, Germany 3 Gottfried Wilhelm Leibniz Universität Hannover, Germany abstract The metal layout design influences the reliability of the metallization in semiconductor products. An optimized design of the interconnect stack can help to reduce the incidence of dielectric and passivation cracking during Joule heating of the metallization in semiconductor back end of line (BEOL) structures. The elements of the metal stack have different material properties. Thermal stress from Joule heating can cause mismatch in thermal expansion between the materials. This can lead to high stress gradients. The paper shows the comparison of the standard design versus the Highly Robust (HiRo) metallization layout. The evaluation is done for an AlCu metallization with W-plugs in a 180 nm technology node with a metal stack with thick metal (~3 µm thick) on top. The simulation results show better protection against thermal stress caused by Joule heating for the HiRo-layout. |
17:30 |
20mn |
The Scope of Applicability of DPL Model to the Heat Transfer in Electron Devices Mariusz Zubert, Tomasz Raszkowski, Agnieszka Samson, Marcin Janicki, Andrzej Napieralski, Department of Microelectronics and Computer Science, Lodz University of Technology, Lodz, Poland abstract This paper presents the scope of applicability of Dual-Phase-Lag model in modern electronic structures. Moreover, the investigation of obligatory application of this model, instead of the classical approach based on Fourier-Kirchhoff model, to heat transfer modeling is included. Furthermore, the modified Fourier-Kirchhoff thermal model, containing the special time lag parameter, is also taken into consideration. In order to obtain the mentioned scope of applicability both analyzed thermal model, the three different classical power transistors i.e. unipolar, bipolar and insulated gate bipolar transistors (IGBT), have been considered. The received simulation results have been carefully compared and analyzed in detail. |
Session 7 |
Solid State Lighting I |
16:20 Elbterrasse |
Monday April 03 2017 |
Chaired by S. Wiese, M. Stiebing |
16:20 |
30mn |
Keynote presentation – Luminescence Mechanism Analysis on High Power Tunable Color Temperature Chip-on-Board White LED Modules Jiajie Fan 1, Chaoyi Xie 1, Cheng Qian 2, Xuejun Fan 3, Guoqi Zhang 4 1 College of Mechanical and Electrical Engineering, Hohai University, Changzhou, China 2 Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China 3 Department of Mechanical Engineering, Lamar University, Beaumont, TX, USA 4 EEMCS Faculty, Delft University of Technology, Delft, the Netherlands abstract Recently, with the increasing requirements on guaranteeing the color uniformity and improving the luminous efficacy and manufacturing efficiency, a wafer level chip scale packaging (WLCSP) technology has been developed by thermally impressing a thin multiple phosphor film on a LED wafer, then being segmented into individual LED chips. In this paper, a high power white LED Chip-on-Board (COB) module with high color rendering index (CRI, Ra>93) and tunable correlated color temperatures (CCTs) is prepared. In this COB module, the tunable color temperatures are achieved by using two types of white LED WLCSPs with different target CCTs of 3000K and 5000K, and twelve CSPs are mounted in series with the flip-chip soldering technology. The thermal dissipation effect and luminescence mechanism of this COB module are studied through the finite element analysis (FEA) and Monte-Carlo Ray-tracing simulations respectively. The results reveal that: firstly, the measured spectral power distribution (SPD) intensity of the prepared COB module are smaller than the arithmetic sum of SPD intensities of its each series connected CSPs, which may be attribute to the light absorption happened among CSPs. Secondly, the case temperature and driven current have the different effects on the optical and color performances of COB module (such as luminous flux, CCT and CRI). |
16:50 |
20mn |
Effects of Phosphor Powder Dispersion on Luminous Properties of a Chip Scale Package LED Cheng Qian 1, Liang Liang Luo 2, Jia Jie Fan 3, Xiao Qiang Li 1, Xue Jun Fan 4, Guo Qi Zhang 5 1 Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China 2 Changzhou Institute of Technology Research for Solid State Lighting, Changzhou, 213161, China 3 College of Mechanical and Electrical Engineering, Hohai University, Changzhou, 213022, China 4 Department of Mechanical Engineering, Lamar University, Beaumont, TX 77710, USA 5 EEMCS Faculty, Delft University of Technology, Delft, the Netherlands abstract As being used in a phosphor converted white LED (pc-white LED), the phosphor powders are always mixed with silicone for the color conversion of LED blue chip. Therefore, phosphor dispersion in the phosphor/silicone composite is regarded as a critical factor for optical characteristics of pc-white LEDs. In this paper, effects of the phosphor dispersion on the spectra, luminous efficacies and color coordinates of the pc-white LEDs are investigated from experimental measurements of a number of Chip Scale Package (CSP) LEDs with different phosphor dispersion tragedies. The measurement results indicate that the spectra become not significantly varied when the phosphor concentration above the chip goes up to a threshold, e.g. 12%. Moreover, both the luminous efficacy and one of color coordinates u’ of the CSP samples slightly increase and then reach a convergence by shifting more phosphor from the area around the chip to the area above the chip. Whereas, the color coordinate v’ remains more or less the same. This work is meaningful in the following two aspects: on one hand, it gives a new perspective on the effects of phosphor dispersion on the optical characteristics of pc-white LEDs; on the other hand, it provides a new design guiding rule for those pc-white LEDs with typical spectra. |
17:10 |
20mn |
Thermal Expansion Multi Physic Simulation Methodology For LED Headlamps H. Mechmeche 1, Y. Cheng 2, C. Roucoules 1, D. Vilette 3 1 Valeo Lighting Systems, Bobigny, France 2 Valeo Lighting Systems, Wuhan, China 3 Valeo Lighting Systems, Meslin l’Evêque, Belgium abstract Lighting technology in LED Headlamp is becoming more complex to achieve performance and give comfort to the driver. One technology named “the glare free high beam” is providing lighting where needed on the road with part of the beam hidden or switch off to avoid glaring the incoming drivers. Yet product must be designed to provide the proper light distribution. One difficulty in low beam is to insure the beam stability of the cut off in temperature. Simulation method coupling thermal and mechanical simulation to predict the impact of thermal dilatation has been used to simulate the beam stability of halogen product. Yet due the complexity of LED module, this method has been identified as not predictive. The target has been to develop a simulation method enabling to visualise the impact of deformation of optical system caused by different effect such as thermal dilatation on light distribution. The main difficulty was in the proper mapping of deformed meshed from mechanical simulation into optical simulation software.This paper is presenting the impact of mesh coming from another simulation on optical beam. With the proper mesh resolution of the deformed optical module, analysis of different deformation on optical beam can be done. Analysis of the impact of deformation was considered in two types of situation: a mesh coming from injection simulation and a mesh coming from mechanical simulation. This method of mapping between simulation software mechanical or injection is providing new perspective of analysis of beam stability. |
17:30 |
20mn |
SnS monolayer as gas sensors: insights from a first-principles investigation Fafei Hu 1, Chuanjian Tan 1, Huaiyu Ye 1, Xianping Chen 1, Guoqi Zhang 2 1 Chongqing University, Chongqing, China 2 Delft University of Technology, Delft, The Netherlands abstract Using the first-principle calculations with density functional theory (DFT), we investigate systematically the adsorption of small gas molecules (CO, NH3, SO2 and NO2) on monolayer SnS. The energetics, charge transfer are obtained. We determine the styles of molecule doping, and discuss the nature of interaction mechanism between gas molecules and SnS sheet. According to the calculated results, the adsorption of the specified molecules except SO2 gas on monolayer SnS is a physisorption process with moderate adsorption energy and charge transfer, while SO2 is chemsisorption.CO, SO2 and NO2 act as charge acceptors for the monolayer while NH3 which is found to be charge donors. The results show that the sensing performance of SnS is superior to other layered materials such as graphene and phosphorene. The optical properties results exhibit that WFs could be effectively adjusted by selectively adsorbed different gas molecules. These results suggest that monolayer SnS is a promising candidate for gas sensing applications. |
Session 8 |
IC level stress simulation and characterisation |
09:00 Bellevue |
Tuesday April 04 2017 |
Chaired by S. Kunath |
09:00 |
30mn |
Keynote presentation – Correlation between Mechanical Material Properties and TSV-induced Stress in 3D-Integrated Silicon Microstructures M. Stiebing 1, D. Vogel 2, W. Steller 3, M. J. Wolf 3, U. Zschenderlein 1, B. Wunderle 4 1 Technische Universität Chemnitz, Chemnitz, Germany 2 Fraunhofer ENAS, Chemnitz, Germany 3 Fraunhofer IZM, Moritzburg, Germany 4 Technische Universität Chemnitz, Chemnitz; Fraunhofer ENAS, Chemnitz, Germany abstract Three-dimensional (3D) electronic systems enable higher integration densities compared to their 2D counterparts, a gain required to meet the demands of future exa-scale computing, cloud computing, big data systems, cognitive omputing, mobile devices and other emerging technologies. Through-silicon vias (TSVs) open a pathway to integrate electrical connections for signaling and power delivery through the silicon (Si) carrier used in 3Dstacked microstructures. As a limitation, TSVs induce locally thermomechanical stress in the Si lattice due to a mismatch in the coefficients of thermal expansion between Si and the TSV-filling metals and therefore enforce temperature related expansion and shrinkage during the annealing cycle. This temperature-induced crowding and relaxation of the Si lattice in proximity of the TSV (called ”keep-out-zone” forbidden for active device positioning) can cause a variety of issues ranging from stress-induced device performance degradation, interfacial delamination or interconnect failures due to cracking of the bond or even of the entire Si microstructures at stress hotspots upon assembly or operation. Additionally also the interconnect structures induce stress that will overlap with the TSV induced stress. In this paper, we investigate layout and material related stress profiles from a theoretically and experimentally point of view. Applied on a novel dual-shell Si interposer platform including liquid-cooling technology based on fluidic micro-channels aiming at high performance heat removal (i.e. as dissipated from HPC-processors on top of the evice). Additionally while using water as coolant certain sealing structures have to be integrated to circumvent electrical shorts and electrolysis. Initially we investigate experimentally the mechanical properties of the electro- lated copper used for the fabrication of the actual devices. In a second step we did the simulation of strain the strain brought into the Si lattice. The process- and morphology-dependent stress strain curve is a crucial input parameter for the following simulation as deviations at small strain levels have a major impact on the outcomes. Tensile testing for 5, 15 and 20 μm thick Cu samples were performed with 15 μm thickness being the nominal diameter of he TSVs used in the later studies. This macroscopic material behavior is correlated with the microstructure of the material as determined by Scanning Electron Microscopy. Experimentally, the stress profile in Si around a certain interconnect and sealing structure is measured by fibDAC, a method for investigation of stress relaxation. A comparison of measurements and Finite Element Analysis (FEA) will be provided, revealing good qualitative and quantitative correlations. Finally, we estimate reliability risks, an assessment which is based on the combined thermo-mechanical stress determinations by experiment and theory. |
09:30 |
20mn |
Optimization of Through Crackstop Via using Finite Element Modelling Mohamed A. Rabie, Nick Polomofff, Global Foundries Inc., USA abstract ICs are separated by dicing the wafer along dicing lanes with a saw to form IC chips or dies which are then packaged to form the final product. Cracks form during wafer dicing due to mechanical forces resulting from dynamic vibrations of the die during dicing. Those cracks initiate in the weakest point of the die which is typically in the back end of line (BEoL) stack. Crackstops surrounding the die are used to prevent the edge cracks from propagating in the BEoL stack and damaging the die. A Through Crackstop Via has been proposed as an advantageous structure over conventional crackstops. The ability of TCV to prevent cracks from propagation is compared to conventional crackstop using a novel systematic simulation methodology in this work. The design of TCV is, then, optimized by varying the angle between the metal layers as well as the TCV width. The larger angle and width result in a better TCV. The improvement in the TCV saturates at a specific TCV width. |
09:50 |
20mn |
Strength Characterization of TEOS and FTEOS Interlayer Dielectric Materials to Compare Fracture Risk in Die Stack by 3D FE Simulation Approach Ilko Schmadlak 1, Nishant Lakhera 2, Jim Howell 2, Scott Kipperman 2, Evan Welsh 2 1 NXP Semiconductors, Munich, Germany 2 NXP Semiconductors, Austin, USA abstract In this study, two very common ILD materials, tetraethyl orthosilicate (TEOS) and fluorinated tetraethyl orthosilicate (FTEOS) were compared with respect to their fracture toughness in a wafer stack. A combination of mechanical testing, mechanical modeling and simulation were performed to identify fracture parameter of the particular bulk materials and material interfaces, translate them into fracture stresses and compare different BEOL structures with respect to their potential fracture risk. To estimate the critical fracture stress, a simple 2D Finite Element Analysis (FEA) was conducted. In a second step a much more complex 3D sub-modelling approach was used to investigate the specific differences of die locations with respect to fracture risk and considered the BEOL stack metal layout in greater detail. The simulations used stress-based as well as energy-based parameters for comparison. |
Session 9 |
Molecular Dynamics |
09:00 Leipzig |
Tuesday April 04 2017 |
Chaired by Huaiyu Ye, Haibo Fan |
09:00 |
30mn |
Keynote presentation – Interface Comparison Involved in Flexible Electronics Using Molecular Modeling N.E. Iwamoto, Carol Putman, Gregory Vernon, Rachel Cramm Horn, Aaron Bernreuther, Honeywell, USA abstract Although the area of flexible circuits is not new, current display and portable electronics consumer markets are driving device and packaging designs that will require increasingly higher robustness, especially as more devices are driven toward lightweight, wearable designs that enhance the personal IOT (Internet of Things) association. There will be an advantage for developers in flexible electronics to fully understand critical interface weaknesses that can in-turn help to push new material development. The use of molecular modeling to study interfacial failure has been applied to the flex circuit adhesive-to-metal interfaces in order to understand the relative weaknesses at these interfaces. Specifically copper/adhesive and chromium/adhesive interfaces have been modeled using the major polymer components of the adhesive and the metal oxides. This paper will discuss the results of these models and how polymer composition may be contributing to specific interface failure, and consistencies with observed failure. |
09:30 |
20mn |
Thermal conductivity of functionalized graphene-ploymer nanocomposite: A non-equilibrium molecular dynamics study Hongyu Tang 1, Huaiyu Ye 2, Xianping Chen 2, Xuejun Fan 3, Guoqi Zhang 4 1 Delft Institute of Microsystems and Nanoelectronics, Delft University of Technology, Delft 2628 CD, The NetherlandsChangzhou Institute of Technology Research for Solid State Lighting, Changzhou, 213161, China 2 Key Laboratory of Optoelectronic Technology 3 Department of Mechanical Engineering, Lamar University, Beaumont, Texas 77710, USA 4 Delft Institute of Microsystems and Nanoelectronics, Delft University of Technology, Delft 2628 CD, The Netherlands abstract Heat transfer across thermal interface material, such as graphene-polymer composite, is a critical issue for microelectronics thermal management. To improve its thermal performance, we use chemical functionalization on the graphene with hydrocarbon chains in this work. Molecular dynamics simulations are used to identify the thermal conductivity of monolayer graphene and graphene-polymer nanocomposites with and without grafted hydrocarbon chain. The influence of functionalization with hydrocarbon chains on the interfacial thermal conductance of graphene-polyethylene nanocomposites was investigated using a non-equilibrium molecular dynamics (NEMD) simulation. We also study the effects of the graft density (number of hydrocarbon chain) on the thermal conductivity of graphene and the nanocomposite. |
09:50 |
20mn |
The intriguing electronic and optical properties modulation in blue phosphorene/g-III-nitrides heterostructures Qun Yang 1, Chunjian Tan 1, Huaiyu Ye 1, Xianping Chen 1, Guoqi Zhang 2 1 Key Laboratory of Optoelectronic Technology, Chongqing University, Chongqing, China 2 Delft University of Technology, Delft, Netherlands abstract In this work, the structural, electronic and optical properties of blue phosphorene (BP) and graphene-like III-nitrides, denoted as g-XN (AlN and GaN) nanocomposites are investigated by the first-principles method. Our results unveil that the hybridized BP/g-XN bilayers exhibit a decreased band gap. We also find that the optically active states of the maximum valence and minimum conduction bands are localized on opposite monolayers, leading to electrons and holes spontaneously separated, which enhances the photocatalytic efficiency. More interestingly, despite of the indirect band gap nature of the BP and g-AlN monolayers, BP/g-AlN heterostructure in most energetic preferable pattern exhibits a moderate direct band gap. The BP/g-XN heterobilayers also exhibit a significant improved visible light and UV adsorption peak, comparable or even superior to pristine BP, and the superior optical properties is robust, independent of stacking pattern. Therefore, the g-XN layers can be an excellent solution to protect the BP layer from its degradation in ambient conditions. We predict that such effective electronic band gap engineering, together with intriguing optical properties, point toward the potential of BP/g-XN heterobialyers for applications in a variety of nanodevices. |
Session 10 |
Thermal Characterization |
09:00 Elbterrasse |
Tuesday April 04 2017 |
Chaired by P. Rodgers, T. Raszkowski |
09:00 |
30mn |
Keynote presentation – Investigations on the Temperature Distribution of Integrated Heater Configurations in a Lab-on-a-Chip System Petra Streit 1, Joerg Nestler 2, Robert Schulze 3, Alexey Shaporin 1, Thomas Otto 1 1 Fraunhofer ENAS, Chemnitz, Germany 2 BiFlow Systems GmbH, Chemnitz, Germany 3 Chemnitz University of Technology, Chemnitz, Germany |
09:30 |
20mn |
IR Imaging of Laser Structures for Thermal Control of Photonics Integrated Circuits (PICs) Niamh Richardson, Jeff Punch, Eric Dalton, Marian Carroll, CONNECT, Stokes Laboratories University of Limerick, Castletroy, Limerick, Ireland abstract The Thermally Integrated Smart Photonics Systems (TIPS) H2020 project aims to develop a solution to meet the significant demands of data traffic growth, by designing a scalable, thermally-enabled, 3D integrated optoelectronic platform. Micro-thermoelectric coolers (TECs) and micro-fluidics will be integrated with optoelectronic devices to precisely control device temperature, and thus device wavelength. To understand the thermal-hydraulic behaviour of micro-scale heat exchangers, a range of exchanger geometries will be characterised hydraulically (via manometry and velocimetry) and thermally (via infra-red imaging). This paper will discuss the optical and thermal characterisation of existing active laser devices using infra-red imaging in order to obtain a baseline thermal resistance for the development of the heat exchangers. Thermographs of existing active devices were recorded to determine the thermal characteristics of the laser structure and the spatial temperature variation across the laser surface. The increase in temperature of an active laser as a function of dissipated power was found to be linear at 45⁰C/W. Repeatability and laser-to-laser variation tests showed good agreement. The spatial temperature variations in the x- and y- directions were ±8⁰C and ±3⁰C of the mean temperature, respectively. An understanding of the thermal characteristics of existing laser devices will allow for appropriate testing of the viability of microfluidic heat exchangers as coolers for micro-TECs. |
09:50 |
20mn |
Requirements Specification for Multi-Domain LED Compact Model Development in Delphi4LED Anton Alexeev 1, Robin Bornoff 2, Sangye Lungten 1, Genevieve Martin 3, András Poppe 4, Grigory Onushkin 3, Márta Rencz 5, Joan Yu 3 1 Eindhoven University of Technology, The Netherlands 2 Mentor Graphics, London, United Kingdom 3 Philips Lighting, Eindhoven, The Netherlands 4 Budapest University of Technology and Economics, Budapest, Hungary 5 Mentor Graphics Mechanical Analysis Division MicReD, Budapest, Hungary abstract Light-emitting diode (LED) technology has been rapidly developing due to high energy efficiency and longer lifetimes of LED luminaires. One of the main challenges in designing LED components is to manage the inter-twined relation between thermal, electrical, and optical performances. These dependencies are required to be well understood in order to operate LEDs efficiently and have accurate performance prediction at different levels of the product value chain. Delphi4LED project aims at developing standardised method to create multi-domain (thermal, electrical, and optical) compact models from the measurement data. To obtain highly reliable and representative data sets via characterisation and calibration, end-user requirements specification for diverse LED samples are needed. In this paper, we report the lists of LED parameters and components selected for measurements, simulation and calibration considering end-user needs. We also show some of the methodologies for compact thermal model development that are recommended in Delphi4LED, followed by definition of simulation benchmark problems. |
Session 11 |
Posters : Multi-Physics Issues in Microelectronics |
10:10 |
Tuesday April 04 2017 |
Chaired by B. Wunderle, K. Jansen |
|
|
Leveraging Accelerated Testing to Assess the Reliability of Two-Stage and Multi-Channel Drivers J. Lynn Davis 1, Curtis Perkins 1, Aaron Smith 1, Terry Clark 2, Karmann Mills 1 1 RTI International, Research Triangle Park, NC, USA 2 Finelite Inc., Union City, CA, USA abstract The next wave of LED lighting technology is likely to be tunable white lighting (TWL) devices which can adjust the colour of the emitted light between warm white (~ 2700 K) and cool white (~ 6500 K). This type of lighting system uses LED assemblies of two or more colours each controlled by separate driver channels that independently adjust the current levels to achieve the desired lighting colour. Drivers used in TWL devices are inherently more complex than those found in simple SSL devices, due to the number of electrical components in the driver required to achieve this level of control. The reliability of such lighting systems can only be studied using accelerated stress tests (AST) that accelerate the aging process to time frames that can be accommodated in laboratory testing. This paper describes AST methods and findings developed from AST data that provide insights into the lifetime of the main components of one-channel and multi-channel LED devices. The use of AST protocols to confirm product reliability is necessary to ensure that the technology can meet the performance and lifetime requirements of the intended application. |
|
|
Lifetime Prediction Based On Analytical Multi-Physics Simulation for Light-Emitting Diode (LED) Systems Xi Yang 1, Zili Wang 1, Yi Ren 1, Bo Sun 1, Cheng Qian 2 1 School of Reliability and Systems Engineering, Beihang University, China 2 State Key Laboratory of Solid-State Lighting, China abstract A novel method based on the analytical multi-physics simulation to predict the lifetime for LED systems is presented in this paper. LED is a typical multidisciplinary system that transfers electrical input into photonic output, and lifetime prediction is a must for new electronic device before releasing to market. The lifetime feature of LED systems is closely connected to the power, thermal, and optical parameters, but there is rare method concerning all these features to conduct lifetime prediction. In this paper, an original concept, Lumen Impulse (referred as LI), is proposed, which combines the lifetime and performance parameters together to indicate the accumulation of luminous flux in the time domain. With LI being regarded as an objective function, multidisciplinary design optimization (MDO) method is employed to solve the coupling formulas, the lifetime with higher precision can be obtained subsequently. The impacts of certain design parameters in LED systems on the lifetime and LI are discussed to present suggestions for the enhancement of lifetime performance of the LED systems. This research result will be helpful for manufactures and designers to predict the lifetime with higher accuracy and determine a most qualified operating point of LED systems. |
|
|
A Tool for Aided Multi-Scale Model Derivation and its Application to the Simulation of a Micro Mirror Array Walid Belkhir 1, Nicolas Ratier 1, Duy Duc Nguyen 1, Nguyen Nhat Binh Trinh 1, Michel Lenczner 1, Frédéric Zamkotsian 2 1 University of Bourgogne Franche-Comté, FEMTO-ST, Besancçon, France. 2 LAM-CNRS, Marseille, France. abstract Modeling the electric field in a matrix of micro-mirrors ispresented as the first application of the MEMSALab software package. Thelatter is dedicated to semi-automated derivation of multiscale models byasymptotic methods and will complement simulation software as finite elementsoftware. It is designed according to a principle of reusability which iscalled the extension-combination method developed with techniques derivedfrom the theory of rewriting. |
|
|
Considerations on Pre-Stress in a 3D- and Inkjet-Printed Capacitive Force/Pressure Sensor Lisa-Marie Faller, Hubert Zangl, Sensors and Actuators, Institute of Smart System Technologies, A-9020 Klagenfurt, Austria abstract In this work, we infer the influence of pre-stress in the design of a capacitive force/pressure sensor. For this purpose, experimental plans, so-called Designs of Experiments (DoEs), are employed. Based on these designs, not only the influence of design parameters can be inferred, but also a prediction model, which uses the design parameters as predictors, can be determined. Rapid prototyping technologies, such as the considered 3D- and inkjet-printing processes, provide major advantages such as flexibility in design. At the same time, such processes lead to inhomogeneous material properties. The presented methodology, consequently, aims at determining the variations in sensor read-out due to pre-stress influence. A combination of Response-Surface-Method DoEs for design variables and randomized DoE of noise variables (i.e. the varying material properties) is employed. We demonstrate the influence of pre-stress on the considered geometry as well as the setup of a suitable prediction model for the read-out. |
|
|
Adsorption of gases on monolayer GeSe: a first principle study Lian Liu 1, Qun Yang 1, Huaiyu Ye 1, Xianping Chen 1, Guoqi Zhang 2 1 Chongqing University, Chongqing, China 2 Delft University of Technology, Delft, The Netherlands abstract Graphene with many fascinating properties has attracted enormous attention. Recently, we found that Germanium selenide similar to black phosphorous. The results exhibit that NH3 adsorption with moderate adsorption energy is physisorbed on monolayer GeSe, whereas SO2 and NO2 adsorption are chemisorbed on monolayer GeSe. SO2 and NO2 adsorption configurations lead to the GeSe monolayer deformation. There is occur serious orbital hybridization after adsorption SO2 and NO2. The ionic bond be formed in NH3 adsorption instead of covalent band in SO2, NO2 adsorption. Our analysis reveals that monolayer GeSe is suitable for highly selectivity NH3 sensor. We also found that monolayer GeSe can be used to catalyze poisonous gases (SO2, NO2). |
|
|
Electrical and Optical Properties of NO and H2S Adsorption on Arsenic Phosphorus Yingying Zhang 1, kai Zheng 1, Xianping Chen 1, Guoqi Zhang 2, Lian Liu 1, Chunjian Tan 1, Qun Yang 1, Junke Jiang 1, Huaiyu Ye 1 1 Chongqing University, Chongqing, China 2 Delft University of Technology, Delft, Netherlands abstract With the continuous development of two dimension materials technology, found that can be used in more and more application fields. The application of new sensor based on new 2D materials with higher precision has sparked an upsurge research. Here, we study by first-principles calculations the adsorption of NO and H2S gas molecules on a monolayer arsenic–phosphorus. The simulation results testify that Si-doped AsP is sensitive to H2S gas molecule with an excellent charge transfer and a moderate adsorption energy. And then we calculate the work function of a AsP sheet to different small molecules, which means that the selective adsorption of these small molecules. Such selectivity and sensitivity to adsorption makes AsP a superior gas sensor that promises wide-ranging applications, and AsP doped with Si has great potential to be high performance catalyst. |
|
|
Packaging Effects on Q Factor of MEMS Resonator Kisoo Shin, Korea Advanced Nano Fab Center Do-Hwan Park, Justek Inc. Seungoh Han, Hoseo University abstract Quality factor of resonators has been one of the most important parameters because it determines sensitivity of the device, espacially in the case of vacuum-packaged MEMS resonators. Typical energy loss mechanisms of vacuum-packaged MEMS resonator are supporting loss and TED. Those mechanisms can be affected by package structure because the package structure consists of acoustic wave path as well as thermal dissipation path. Through severe numerical simulations, temperature-dependency of Q factor including thermal deformation of its package were obtained, where TED was the limiting mechanism and it decreased with the temperature increase. Paste used to adhere MEMS die to the package has also significant effects on Q factor because it works as one of thermal resistors for TED mechanism. Therefore, it’s required to optimize package structure, number and location of adhesive paste, and MEMS resonator itself in order to maximze Q factor. |
Session 12 |
Posters : Thermal and Thermo-Mechanical Issues in Microelectronics |
10:10 |
Tuesday April 04 2017 |
Chaired by B. Wunderle, K. Jansen |
|
|
Method for determination of Young’s Modulus and Coefficient of Thermal Expansion of thin films Sergey Ananiev, Manfred Schneegans, Infineon Technologies AG, Neubiberg, Germany |
|
|
Influence of Under Bump Metalization Dimensions on Passivation Nitride Stress Raj Sekar Sethu 1, Salil Hari Kulkarni 2, How Ung Ha 2, Kok Heng Soon 2 1 X-FAB Semiconductor Foundries AG, Kuching, Malaysia 2 Swinburne University of Technology Sarawak, Kuching, Malaysia abstract Cracks in die passivation adjacent to solder bumps due to thermal stress can lead to a multitude of failures including bond pad lifting. Modifying the solder reflow temperature-time profile is not trivial as this may impact the metallurgical properties of the solder ball. In this work, Under Bump Metalization (UBM) dimensions were optimized using finite element analysis (FEA) and response surface methodology (RSM) to obtain the lowest first principal stress for the Silicon Nitride (Si3N4) passivation layer. The optimized minimum stress value using FEA was well within the confidence interval predicted using RSM. |
|
|
Impact of Wire Material and Fluorine in Dielectric on Wire Pull Test Stress Raj Sekar Sethu 1, Christian Schirrmann 2, How Ung Ha 3, Kok Heng Soon 3 1 X-FAB Semiconductor Foundries AG, Kuching, Malaysia 2 X-FAB Semiconductor Foundries AG, Dresden, Germany 3 Swinburne University of Technology Sarawak, Kuching, Malaysia abstract The move towards low-k dielectrics and change of wire material from gold (Au) to copper (Cu) is being driven by the need for lower resistive-capacitive (RC) delay in circuits and cost. But this can increase the incidence of bond pad lifting. Finite element analysis (FEA) stress simulation prior to mass production can provide an insight to the probability of such failures and guide device manufacturers to focus their efforts on factors that are significant. In this work, it was discovered that the change in wire material has a higher impact to back end of line (BEOL) structure’s first principal stress compared to the change in the low-k dielectric material properties. |
|
|
Experimental Determination of the Young’s Modulus of various Electronic Packaging Materials F. Kraemer 1, M. Roellig 2, R. Metasch 2, S. Wiese 1, J. Al Ahmar 1, K. Meier 2 1 Saarland University, Saarbrucken 2 Fraunhofer IKTS, Dresden abstract The paper presents details about the adequate experimental determination of the Young’s modulus on miniaturized specimens for material used in electronic packaging. The difficulty to determine accurately the Young’s modulus is caused by the requirements of representative specimens for the area of electronic packaging. In many cases such specimens, e.g. solder balls, are connected with the issues of inhomogeneous stress distributions, small dimensions, or special gripping requirements, that create a number of challenges to conduct mechanical experiments. In addition there are problems that arise from the nonlinearities in the constitutive behaviour of the material to be characterized, such as creep deformation. Therefore any attempt to accurately determine the Young’s modulus needs a case to case consideration of the specific issues for the given specimen and material. |
|
|
Effect of voids on crack propagation in AuSn die attach for high-temperature power modules Faical Arabi 1, Loic Theolier 1, Toni Youssef 2, Mathieu Medina 3, Jean-Yves Deletage 1, Eric Woirgard 1 1 IMS Laboratory, University of Bordeaux, 351 Cours de la Libération, 33405 Talence Cedex – France 2 Safran Tech, Safran Paris-Saclay, Châteaufort, France 3 Serma Technologies, Pessac, France abstract Eutectic AuSn solder is increasingly used in high reliability and/or high temperature applications where conventional Pb-free solders exhibit insufficient strength, creep resistance, thermal conductivity, and resistance to corrosion. Excessive void presence and growth and irregular phase formation are critical factors governing the solder joint reliability as long as they act as crack initiation sites. As a whole, the thermal resistance of test vehicles assembled using AuSn solder has experienced significant variation during aging [-55/ 175]°C. On the other hand, inspection of as-prepared die-attachments by X-ray and SEM (observation of cross-section) showed that the initial voids sizes were increased and a propagation of transverse cracks inside the joint between voids had appeared. In this paper, the joint strength and fracture surface of AuSn solder power assemblies after thermal-aging testing were studied experimentally. Finite-element method (FEM) simulations were performed on the joint to study the influence of the voids size and position on the initiation and propagation of cracks. Simulation results were in good agreement with the experiments. Additional challenges, possibilities, and recommendations for how the reliability of a high-temperature AuSn bonding can be improved are also discussed. |
|
|
Nonlinear Modeling and Characterization of a Thermally Driven MEMS Actuator with a Folded Spring Reference Beam Shahabeddin Vamegh Estahbanati, Maher Bakri-Kassem, Rached Dhaouadi, American University of Sharjah, Sharjah, United Arab Emirates abstract This paper proposes a nonlinear model and reports the current-voltage (I-V) characteristics of a thermally driven V-shaped MEMS actuator with a folded spring (FS) reference beam. The nonlinear lumped element model is developed using circuit elements with temperature-dependent material properties. In addition, it is shown experimentally that the Joule heating occurring across the polysilicon layer is a nonlinear phenomenon. The equivalent resistance of the polysilicon layer is therefore a function of temperature and thus the resulting nonlinear I-V characteristics is obtained. The experimental I-V data are used to find the resistance-voltage data which are then fed into the model. The actuator with the FS is also compared to the actuator without it in terms of the I-V characteristics, power consumption, experimental tip displacement, and simulated temperature. |
|
|
FEM Simulation of Cracks in MLCC during Reflow Soldering Joseph Al Ahmar, Steffen Wiese, Saarland University abstract This paper presents results of FEM Simulation and a statistical analysis for cracks in Multilayer Ceramic Capacitors (MLCCs). The main objective of the analysis is to validate the results of bending experiments on MLCCs and mainly the influence of solder type on fracture susceptibility and fracture form in ceramic body. The model also considers the pre-stresses occurring in the capacitor during firing and soldering process and their corresponding failure mechanisms. |
|
|
A test device for in situ TEM investigations on failure behaviour of carbon nanotubes embedded in metals under tensile load Nathanael Jöhrmann 1, Steffen Hartmann 1, Kiran Jacob 1, Jens Bonitz 1, Kathrine E. MacArthur 2, Sascha Hermann 1, Stefan E. Schulz 3, Bernhard Wunderle 1 1 Technische Universität Chemnitz, Germany 2 Ernst Ruska-Centrum und Peter Grünberg Institut, Forschungszentrum Jülich GmbH, Germany 3 Fraunhofer ENAS, Chemnitz, Germany abstract This paper presents an in situ pull-out test device to characterize interfaces between single-walled carbon nanotubes (SWCNTs) and metals. After summarizing results of maximum stresses calculated from molecular dynamics simulations and obtained from in situ scanning electron microscope experiments the need for an in situ experimental method with atomic resolution to study the mechanics of SWCNT-metal interfaces in further detail is outlined. For that purpose, a silicon-based micromechanical test stage with a thermal actuator for pull-out tests inside a transmission electron microscope was developed and characterized. To measure actuator movements digital image correlation was used. First experiments showed a stable movement of the metal electrode in the focal plane of the electron microscope. On the other hand, image drift due to the heat impact of the thermal actuators was observed. Finite element simulations were applied to further investigate the cause of the drift and to evaluate different approaches to solve the issue. Finally, a successful drift compensation by preheating the test device and keeping power consumption constant during the pull out experiment is demonstrated. In the future the presented system may be also used and further developed for in situ characterization of other materials. |
|
|
Lifetime Modeling Based on Anodic Oxidation Failure for Packages with Internal Galvanic Isolation Rainer Schaller 1, Volker Strutz 1, Horst Theuss 1, Rainer Dudek 2, Sven Rzepka 2 1 Infineon Technologies AG, Regensburg, Germany 2 Fraunhofer ENAS, Chemnitz, Germany abstract In More-than-Moore technologies, the number and complexity of micro and nano devices, that are directly integrated into control units of power electronics and mechatronics systems, increase. These systems typically operate at working voltages in the range of 220-1000 VRMS. The galvanic isolation between the high voltage and the low voltage domains is usually ensured by the geometries of the electronic package. Consequently, detailed knowledge about the field behavior under operating and environmental conditions is essential for product development in order to ensure proper isolation quality throughout the product lifetime. The paper describes the development of a lifetime model for devices subjected to strong electrical fields. Anodic oxidation of silicon has been found to be the most important failure mode that occurs under the specific conditions of these devices. After studying the chemical and physical effects of the phenomenon, a reliability model is derived. It allows estimating the lifetime under field conditions based on tests which use the extrinsic accelerators voltage, humidity, and temperature. Furthermore, the impacts of operational and environmental conditions are taken into account. Finally, the accumulation of the degradation is considered by applying Miner’s rule for calculating the cumulative lifetime. The paper ends with an advice for the design of future qualification tests. |
|
|
Simulations of the impact of single-grained lead-free solder joints on the reliability of ball grid array components Andreas Lövberg, Per-Erik Tegehall, Göran Wetter, Klas Brinkfeldt, Dag Andersson, Swerea IVF AB, Mölndal, Sweden |
|
|
Fatigue crack growth analysis of interface between lead frame and molding compound under isothermal mechanical loading Erkan Bektas 1, Katrin Broermann 1, Goran Pećanac 1, Sven Rzepka 2, Christian Silber 1, Bernhard Wunderle 3 1 Robert Bosch GmbH, Reutlingen, Germany 2 Fraunhofer ENAS, Chemnitz, Germany 3 Technical University of Chemnitz, Chemnitz, Germany abstract Integrated circuit (IC) packages can delaminate under thermal cyclic loading because of elastic properties mismatch between the components. On the other hand, delamination has no pattern which points out the effect of variations of the design parameters on the response. Hence, a robustness estimation should be done to have more stable IC packages which show no important variations in the response with respect to uncertainties in the design parameters. In this paper, the importance of the robustness estimation for IC package design will be explained and a methodology on robust design estimations will be introduced. After that a short literature review will be given about robust design optimization. |
|
|
A Practical Application Of The Response Surface Methodology To Power Electronics Failure Prediction A. Renaud, E. Woirgard, Laboratoire IMS, Talence, France |
|
|
Fracture Probability of MLCC in Dependence of Solder Fillet Height Joseph Al Ahmar, Erik Wiss, Steffen Wiese, Saarland University abstract This paper presents results of flex cracking experiments on Multilayer Ceramic Capacitors (MLCCs). The influence of solder fillet height, solder type and MLCC dimensions on the crack susceptibility of the component has been investigated. Components mounted with lead free solders have shown a higher resistivity to flex cracking. Thermal shock cracks during cool down phase after soldering were also detected especially for capacitors with higher solder fillet. |
|
|
Finite element representation of complex printed circuit boards for thermo-mechanical reliability assessment Jan Albrecht, Fraunhofer ENAS, Chemnitz, Germany |
|
|
First principle design of CdS/germanene heterostructures with tunable electronic and transport properties KaiZheng 1, Huaiyu Ye 2, Guoqi Zhang 3, Yingying Zhang 2, Lian liu 2, Junke Jiang 2, Qun Yang 2, Chunjian Tan 2, Xianping Chen 2 1 Key Laboratory of Optoelectronic Technology 2 Key Laboratory of Optoelectronic Technology 3 Delft Institute of Microsystems and Nanoelectronics, Delft University of Technology, Delft 2628 CD, The Netherlands abstract In this paper, wesystematically researched electronic and transportproperties of the interface between the CdS monolayerand germanene monolayer, by means of first-principlescalculations based on the DFT. CdS/germaneneheterostructures exibit desirable electronic, optical andtransport properties. More specifically, we find that theband structures, light absorption, conductivity ofCdS/germanene can be widely tuned by applyingmechanical deformation. These remarkablecharacteristics are highly desirable to apply on solar cell,photodiode, phototriode, strain sensor, field effecttransistor et al. ectronic and optoelectronic devices. |
Session 13 |
Solid State Lighting II |
13:30 Elbterrasse |
Tuesday April 04 2017 |
Chaired by L. Davis, C. Qian |
13:30 |
30mn |
Keynote presentation – In-Situ Characterization of Moisture Absorption and Hygroscopic Swelling of Silicone/Phosphor Composite Film and Epoxy Mold Compound in LED Packaging Ibrahim Khalilullah 1, Talukder Reza 1, Liangbiao Chen 1, A. K. M. Monayem H. Mazumder 1, Jiajie Fan 2, Cheng Qian 3, Guoqi Zhang 4, Xuejun Fan 5 1 Lamar University, Beaumont, USA 2 Hohai University, Changzhou; State Key Laboratory of Solid State Lighting, Beijing, China 3 State Key Laboratory of Solid State Lighting, Beijing, China 4 State Key Laboratory of Solid State Lighting, Beijing, China; Delft University of Technology, Delft, the Netherlands 5 Lamar University, Beaumont, USA; State Key Laboratory of Solid State Lighting, Beijing, China abstract Epoxy mold compound (EMC) and optical grade silicone based materials have been widely used in manufacturing Light Emitting Diode (LED). Silicones are being used as encapsulant and lenses for next-generation LED packaging designs, such as chip on board (COB) LEDs. Silicones serve several roles including protective lenses, stress relieving encapsulant, mechanical protection and light path materials. Even though silicone-based materials are typically hydrophobic, they are not resistant to moisture absorption, which may greatly affect the key attributes of silicones such as high transparency, high refractive index, stable thermo-mechanical properties, and tunable hardness. On the other hand, EMCs are used in LEDs as packaging material because of their low cost and high productivity, whereas their moisture absorption is long known. In this study, a new generation of silicone/phosphor composite films and a commonly-used EMC material for COB LEDs are studied regarding their moisture absorption and hygroscopic swelling behaviors by in-situ measurements. Moisture sorption tests were performed using TGA Q5000SA at different humidity and temperature conditions, such as 30˚C & 30% RH, 30˚C & 60% RH, 30˚C & 80% RH, 60˚C & 30% RH, 60˚C & 60% RH, 60˚ C & 80% RH, 80˚C & 30% RH, 80˚C & 60% RH and 80˚C & 80% RH. Hygroscopic swelling was determined using DMA-RH Q800 at 60˚C. Material properties including diffusivity, activation energy, saturated moisture concentration and the coefficient of hygroscopic swelling are extracted from the experimental data. Based on the results, comprehensive discussions are made on the characteristics of moisture absorption and hygroscopic swelling for the two different materials. |
14:00 |
20mn |
Transient Thermal Analysis and Step-Stress Accelerated Ageing Testing for Phosphor-converted White LED Chip Scale Packages Jiayi Fang, Changzhou Institute of Technology Research for Solid State Lighting, Changzhou, China |
14:20 |
20mn |
A PoF and Statistics Combined Reliability Prediction for LED Arrays in Lamps Bo Sun, Delft University of Technology, Delft, the Netherlands Xuejun Fan, Lamar University, Beaumont, Texas, USA Jiajie Fan, Hohai University, Changzhou, Jiangsu, China Cheng Qian, Institute of Semiconductors, Chinese Academy of Sciences, Haidian, Beijing, China Guoqi Zhang, Delft University of Technology, Delft, the Netherland abstract In this work, a physics-of-failure (PoF) reliability prediction methodology is combined with statistical models to consider the interaction between the lumen depreciation and catastrophic failures of LEDs. The current in each LED may redistribute when the catastrophic failure occurs in one of LEDs in an array, thus affecting the operation conditions of the entire LED array. A physics-of-failure based reliability prediction methodology is combined with statistical models to consider the interaction between the lumen depreciation and the catastrophic failure. Electronic-thermal simulations are utilized to obtain operation conditions, including temperature and current. Meanwhile, statistical models are applied to calculate possibilities of the catastrophic failure in different operation conditions. |
14:40 |
20mn |
Understanding the transport phenomena leading to tarnishing of the reflecting silver layer causing reduced light output of LEDs A. Herrmann 1, S.J.F. Erich 2, L.G.J. v.d. Ven 1, W.D. van Driel 3, M. van Soestbergen 4, A. Mavinkurve 4, F. De Buyl 5, O.C.G. Adan 2 1 Eindhoven University of Technology, Eindhoven, the Netherlands 2 Eindhoven University of Technology, Eindhoven, the Netherlands 3 Philips Lighting, Eindhoven, the Netherlands 4 NXP Semiconductors, Nijmegen, the Netherlands 5 Dow Corning Europe s.a., 7180 Seneffe, Belgium abstract Tarnishing of the reflective silver layer in LED packages is an important failure mechanism. To be able to improve package integrity, it is necessary to know the transport path of corrosive species to the silver surface. In this work, we show results on water permeation and content in optical grade silicone. We compare the results on diffusivity obtained by wet cup and dynamic vapor sorption (DVS) measurements. Furthermore, we introduce a setup for studying the predominating transport path of corrosive species in LEDs. |
15:00 |
20mn |
Theoretical and Experimental Study of Thermal Resistance & Temperature Distribution in High-Power AlGaInN LED Arrays A.E. Chernyakov 1, A.V. Aladov 1, A.L. Zakgeim 1, M.N. Mizerov 2, V.M. Ustinov 2, V.I. Smirnov 3, V.A. Sergeev 3 1 SHM R 2 ITMO University, St. Petersburg, Russia 3 Kotel’nikov UBIRE RAS, Ulyanovsk, Russia abstract The driving current, chip area of high-power AlGaInN light emitting diodes (LEDs) and the level of integration of LED arrays are continuously increased to provide ever higher output light flux. The new developments require more attention to pay to the thermal management of LEDs, commonly assessed in terms of the thermal resistance. Temperature distribution in a LED array and its effect on the chip thermal resistance has been studied both theoretically and experimentally. The thermal measurements were performed with the help of a temperature-sensitive parameter – forward voltage drop on the p-n junction under the action of heating current. Two methods of heat exciting were used: step-like оr harmonically pulse-width modulated heating current. Analysis of forward voltage relaxation at transient thermal processes allows determination of thermal impedance components corresponding to the structural elements of the LEDs and arrays.The temperature distribution in the LED array predicted by coupled simulations of the heat transfer agrees well with the experimental measured temperature mapping by the IR-microscopy. |
Session 14 |
MEMS, Sensors, Model Order Reduction |
13:30 Bellevue |
Tuesday April 04 2017 |
Chaired by A. Corigliano, M. Röllig |
13:30 |
30mn |
Keynote presentation – Modelling and Design of Micro-Opto-Mechanical Pressure Sensors in the Presence of Residual Stresses V. Rochus 1, R. Jansen 1, R. Haouari 2, B.Figeys 1, V. Mukund 1, F. Verhaegen 2, J. Goyvaerts 1, P. Neutens 1, J. O’ Callaghan 1, A. Stassen 1, S. Lenci 1, X. Rottenberg 1 1 imec 2 imec/KUL abstract This paper presents the modelling of a Micro-Opto-Mechanical Pressure Sensor (MOMPS) manufactured in a SiN integrated photonic platform. Implemented using a single wavelength Mach Zehnder Interferometer (MZI), it relies on a single readout detector that could be integrated directly on CMOS. Current photonic processes exhibit high residual stresses in the sensitive membrane. In this paper we will focus the analytical model taking the residual stresses in the membrane and the losses in the optical circuit into account. Compressive stress will be considered and the analytical model will be compared with experimental measurements. |
14:00 |
20mn |
Uncertainty quantification in polysilicon MEMS through on-chip testing and reduced-order modelling Ramin Mirzazadeh 1, Saeed Eftekhar Azam 1, Eelco Jansen 2, Stefano Mariani 1 1 Department of Civil and Environmental Engineering, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, Italy 2 Institute of Structural Analysis, Leibniz Universität Hannover, Appelstrasse 9A, 30167 Hannover, Germany abstract In this paper, micro-scale uncertainties affecting the behaviour of microelectromechanical systems (MEMS) are investigated through a mixed numerical/experimental approach. An on-chip test device has been designed and fabricated using standard MEMS fabrication techniques, to deform a (microstructured) polysilicon beam. To interpret the experimental data and also the relevant scatterings in the system response, a high fidelity, parametric finite element (FE) model of the device is developed in ANSYS. Uncertainties in the parameters governing the polysilicon mechanical properties and the geometry of the movable structure are estimated through an inverse analysis. To systematically quantify the uncertainty levels within the realm of a cost-effective statistical analysis, a model order reduction technique based on a synergy of proper orthogonal decomposition (POD) and Kriging interpolation is proposed. The resulting reduced order model is finally fed into a transitional Markov chain Monte Carlo (TMCMC) algorithm for the estimation of the unknown parameters. |
14:20 |
20mn |
Model Order Reduction as an Integral Part of e-Mobility Development Process Thomas Iberer, CADFEM GmbH, Grafing b. München, Germany abstract Ever increasing power, efficiency and robustness requirements of electronic systems demand rising degree of co-operation between development partners. An obvious example is the e-mobility boom. It is essential to keep an eye on the complete chain of interactions from power electronics via magnetic to thermal, vibration, even acoustical domains during the complete development process from initial draft till final construction. Instead of exchanging complete virtual field models between domains, departments or companies involved an effective way is to exchange reduced order models. That allows protecting intellectual properties and establishes interaction options at the same time. In parallel the reduced order models allow the simulation of systems’ behavior. In the paper presented methods are explained to reduce the field simulations’ results into reduced order models. Examples are presented for structural, thermal and electromagnetic field reduction. This will show the role of reduced order models during the complete development process. |
14:40 |
20mn |
Towards High Fidelity Silicon Microphones: Evaluating the Potential of Industrial Microsystems Applying Tailored System-level Models Gabriele Schrag, Technical University of Munich, Munich, Germany Thomas Kuenzig, Infineon Technologies AG, Munich, Germany abstract The benefits of a modular, tailored system-level modeling approach, which combines lumped with distributed models, are demonstrated for an industrial capacitive silicon microphone. The performance of such microphones is determined by distributed effects like viscous damping and inhomogeneous capacitance variation across the membrane as well as by system-level phenomena like package-induced acoustic effects and the impact of the electronic circuitry for biasing and read-out. The proposed modeling approach provides maximum insight into the device and system operation while keeping the computational expense low. All relevant figures of merit are covered by the presented model. Hence it enables to evaluate the potential of optimizing silicon microphones towards high fidelity applications. |
Session 15 |
Simulation and test applied to NEW technologies |
13:30 Leipzig |
Tuesday April 04 2017 |
Chaired by M. Lenczner, R. Metasch |
13:30 |
30mn |
Keynote presentation – Effect of Excitation Conditions on the Durability of High Standoff Electronic Components and Assemblies under Multiaxial Vibration Excitation Raman Sridharan, Abhijit Dasgupta, CALCE, University of Maryland, College Park, MD 20742, USA abstract Electronic assemblies often experience multiaxial vibration environments in use and tall, heavy components are more vulnerable under such loading than are short, light components. The added vulnerability comes from higher response due to nonlinear dynamic amplification of the response under simultaneous multiaxial excitation, termed multi degree of freedom (MDoF) excitation. This paper investigates the geometric nonlinearities and the resulting cross-axis interactions that tall and heavy electronic components experience when subjected to vibration excitation simultaneously along two orthogonal axes. Multiaxial vibration experiments were conducted on tip-loaded cantilever beams to explore the nonlinear vibration response of tall, heavy electronic components. Harmonic base-excitation was simultaneously applied in two orthogonal axes (with a different frequency in each axis), and the phase “difference” between these two harmonic signals was parametrically varied to see the effect on the response amplitude . Based on prior studies, the frequency of the transverse excitation was selected to be the fundamental natural frequency of the cantilever beam and that of the axial direction was selected to be twice as large to maximize the cross-axis interaction. Phase is seen to have a very significant effect on the nonlinear amplification of the response. Nonlinear finite element simulations were conducted to verify and explain the experimental observations. |
14:00 |
20mn |
Life Prediction Of Lead Alloy Based On Multi-Failure Criteria Krystian Jankowski, Artur Wymysłowski, Wrocław University of Science and Technology, Faculty of Microsystem Electronics and Photonics, Wrocław, Poland |
14:20 |
20mn |
Alternative Cu pillar bumps design to reduce thermomechanical stress induced during flip chip assembly Melina Lofrano, Vladimir Cherman, Mario Gonzalez, Eric Beyne, imec abstract In this work a Cu pillar design that combines a stiff metal pedestal with a soft polymer as buffer layer has been integrated in a dedicated test vehicle to investigate the thermo mechanical stress induced during flip chip assembly. In-situ electrical measurements of dedicated stress sensors during a Bump Assisted BEOL Stability Indentation (BABSI) test were performed to assess the strength of the bump designs. Furthermore, the package induced stress was monitored in different regions of the test chips by measuring and comparing the ION current of the stress sensors before and after packaging. By combining in-situ electrical measurements and finite element modeling it was possible to quantify the stress level induced in the Si die after packaging. The results show that the use of a stiff pedestal is very efficient to mitigate packaging induced stress. |
14:40 |
20mn |
Massively Parallel Computations of Creep Deformation in Flip-Chip Interconnections Cedrick Bouchard, Julien Sylvestre, University of Sherbrooke abstract In order to enable the computation of creep deformation in a large number of interconnections in simulations of thermal cycling or assembly processes of flip chip devices, a submodeling approach was developed to distribute the stress and creep evaluation to independent solvers for individual (or small number of) interconnections and thus allow a high level of parallelization of the solving effort. The approach uses a first, coarser model of the complete assembly to compute the general system response to thermal and mechanical loads. Displacements calculated in this first model are fed into a large number of much more detailed models of the individual interconnections, in order to obtain local creep strains. The coarser model is iteratively updated using local creep values from the interconnection models to determine the complete time evolution of the system in the presence of interconnection creep deformation. The validity of the complete parallelized procedure is verified on simplified cases. |
Session 16 |
Prognostics & Health Monitoring |
15:40 Bellevue |
Tuesday April 04 2017 |
Chaired by P. Gromala, I. Schmadlak |
15:40 |
30mn |
Keynote presentation – Prognostics & Health Management for LED-based Applications W.D. van Driel 1, B. Jacobs 2, D. Schenkelaars 2, M. Klompenhouwer 2, R. Poelma 3, B. El Mansouri 3, L.M. Middelburg 3 1 Philips Lighting, High Tech Campus, Eindhoven; Delft University of Technology, Delft, the Netherlands 2 Philips Lighting, High Tech Campus, Eindhoven, The Netherlands 3 Delft University of Technology, Delft, the Netherlands abstract Traditional lighting is focused on the prevention of hardware failures. With the trend towards controlled and connected systems, other components will start playing an equal role in the reliability of it. Here reliability need to be replaced by availability and other modeling approaches are to be taken into account. System prognostics and health management is the next step to service the connected complex systems in the most effective way possible. In this keynote we will highlight the next frontiers that will need to be taken in order to move the traditional lighting catastrophic failure thinking into a thinking more to-wards new ways how system (degraded) functions can fail or be compromised. |
16:10 |
20mn |
In-situ monitoring of field conditions and interconnect integrity for an electronic on-board module Riet Labie 1, Bart Vandevelde 1, Wesley Van Meensel 1, Mike Vogeleer 2, Daniel Werkhoven 3, Bart Allaert 4, Geert Willems 1 1 imec, Leuven, Belgium 2 Dekimo, Gentbrugge, Belgium 3 Interflux electronics, Gent, Belgium 4 Connect group, Ieper, Belgium abstract An electronic test module is designed in order to monitor field failures of board assembly interconnections while simultaneously measuring the environmental loading conditions. The test module is built-in on the ceiling of a daily driving bus. During operation, when the engine is on, daisy chain interconnections as well as temperature and accelerometer sensors are measured continuously. The purpose is to extract critical loading conditions which could be a first step towards the introduction of prognostic health monitoring units. After 1.5 years of testing, no failures were identified for two types of assembly materials: solder and conductive adhesive. Temperature fluctuations turned out to be very small with a limited maximum temperature around 35oC due to ambient regulation of the bus interior. Furthermore, relatively mild acceleration values are monitored at low frequency. New experiments in more harsh environments are solicited. |
16:30 |
20mn |
Predictive reliability with signal based meta-models Stephanie Kunath, Veit Bayer, Roland Niemeier, Dynardo GmbH, Weimar, Germany abstract The development of algorithms and models to be used for prediction of the reliability and health monitoring of components and sensors is of great importance in aerospace, automotive and power generation industry. For this purpose metamodels have been developed that are based on physical simulations and that are able to quantify the impact of uncertainties on system behavior. These surrogate metamodels for time dependent signals can approximate the failure behavior and detect symptoms of aging. Furthermore, the prediction which input parameter combination can be run by the measurement setup without risk of failure or break in testing is an important application. Our approach has been validated for a high lift system in the aerospace industry. |
16:50 |
20mn |
Prognostics and Health Monitoring of Electronic System: A Review Alexandru Prisacaru 1, Przemyslaw Jakub Gromala 1, Mateus Bagetti Jeronimo 1, Bongtae Han 2, Guo Qi Zhang 3 1 Robert Bosch GmbH, Reutlingen, Germany 2 University of Maryland, College Park, USA 3 Delft University of Technology, Delft, Netherlands |
Session 17 |
Characterization, experiments and modelling |
15:40 Leipzig |
Tuesday April 04 2017 |
Chaired by I. Maus, E. Bektas |
15:40 |
30mn |
Keynote presentation – Thermo-mechanical properties of 3D printed ABS parts fabricated by fused deposition modelling and vapor smoothing Sung-Uk Zhang 1, Jonghyeuk Han 2, Hyun-Wook Kang 2, Byoung-Chul Shin 1 1 Dong-Eui University, Busan, South Korea 2 Ulsan National Institute of Science of Technology, Ulsan, South Korea abstract FDM is one of the popular 3D printing technologies due to an inexpensive extrusion machine and multi-material printing. However, FDM could only use thermoplastics such as ABS and PLA and has a problem related to the post-processing. In this study, we measured the mechanical property of ABS parts fabricated by FDM and the vapor smoothing technique which is one of the post-processing methods. Using dynamic mechanical analysis (DMA) and dilatometer, we observed temperature-dependent storage modulus and CTE for specimens varying with amount of acetone in the vapor smoothing process. In result, we could not observe the effect of the amount of acetone in the given range but differentiate one without the vapor smoothing process and the others with the vapor smoothing process in terms of the storage modulus and CTE. Moreover, we could perform finite element analysis using the measured mechanical properties and make a design guideline for an ABS product. We could conclude that the vapor smoothing process weakens thermal stability of ABS 3D printed parts. |
16:10 |
20mn |
Design Optimization for a Power Package by Simulation Haibo Fan 1, WW Chow 1, Pompeo V Umali 1, Fei Wong 1, Kai Zhang 2, Haibin Chen 2, Jingshen Wu 2 1 Nexperia, Hong Kong 2 Hong Kong University of Science and Technology, Hong Kong abstract In this study, different clip designs are proposed and effect of clip tip, die pad size and die size on junction temperature under surge test are studies. Simulation data show that clip design is the only key for device to pass surge test without any soft solder melting issue. A clip design guideline is thereafter provided with that making clip tip small as possible for better view inspection and solder flux cleaning, but also keeping a good thermal performance to pass surge test without reliability concerns. |
16:30 |
20mn |
An AlAs/germanene heterostructure with outstanding tunability of electronic properties Chunjian Tan 1, Qun Yang 1, Huaiyu Ye 1, Xianping Cheng 1, G.Q.Zhang 2 1 aKey Laboratory of Optoelectronic Technology 2 Delft University of Technology, Delft, Netherlands abstract By means of comprehensive first-principles calculations, we investigate the stability and electronic properties of AlAs/germanene heterostructures. Especially, electric field and strain are used to tailor its electronic band gap. The binding energy and interlayer distance indicate that germanene and AlAs monolayers in AAI pattern are bound together via van der Waals interaction with a maximum indirect-gap of 0.494 eV, which is expected to has potential application in the field of field-effect transistors. Under the negative E-field and compressive strain, the bandgaps of the AAI-stacking show a near-linear and linear decrease behavior respectively, whereas the response of the bandgaps to the positive E-field and tensile strain displays a dramatic and monotonous decrease relationship. All these nontrivial and tunable properties endow AlAs/germanene nanocomposite great potentials for FETs, strain sensors, and photonic devices. |
Session 18 |
Thermo-mechanical Analysis |
15:40 Elbterrasse |
Tuesday April 04 2017 |
Chaired by Jiajie Fan, M. Kuczynska |
15:40 |
30mn |
Keynote presentation – Analytical and simulation-based risk assessment of imprint depth and brittle fracture in bond pad stacks Georg M. Reuther 1, Ivan Penjovic 1, Angel Ochoa Brezmes 2, Reinhard Pufall 1 1 Infineon Technologies AG, Am Campeon 1-12, 85579 Neubiberg, Germany 2 Technische Universität Dresden, Institut für Energietechnik, Helmholtzstraße 14, 01069 Dresden, Germany abstract During electrical wafer testing and wire bonding onto pad metallization, oxide layers in Backend-of-Line (BEOL) pad stacks are exposed to the risk of mechanical damage. Subsequent metal migration into oxide cracks leads to electrical device failure. We undertook simulation-based risk assessment using analytical and Finite Element Modelling (FEM) with regard to critical imprint depths in top metallization layers of elementary metal-oxide test vehicles. Our modelling outcomes cope well with results obtained by instrumented indentation and, thus, constitute a promising physics-of-failure approach towards minimizing the risk of lifetime-limiting oxide fracture. |
16:10 |
20mn |
A study on the correlation between experiment and simulation of the board level drop test for SSD(Solid State Drive) Tae Min Kang 1, Yong Chang Lee 2, Byung Kwon Bae 1, Won Seob Song 1, Jae Sung Lee 1 1 SK hynix NAND Solution Development Division, Seongnam-si, Republic of Korea 2 SK hynix PKG R abstract Recently, handheld electronic products are prone to being dropped during their useful service life because of their size and weight. Board level solder joint reliability performance of IC packages during drop impact becomes a great concern to semiconductor and electronic product manufacturers. The packages are susceptible to solder joint failures, induced by a combination of PCB bending and mechanical shock during impact. Therefore, board level drop testing is an effective method to characterize the solder joint reliability performance of miniature handheld products.In this study, we examine and compare simulation the board level drop test of SSD. Applying the JEDEC (JESD22-B111) standard present a finite element modeling of the BGA package assembly was performed to study the stress and strain behavior of the solder joints during drop test. The simulation revealed that maximum stress was located at the outermost solder ball in the PCB or Package side, which consisted well with the location of crack initiation observed in the failure analysis after drop reliability tests. |
16:30 |
20mn |
The onset of plastic flow in copper materials used for rigid and flexible PCB Steffen Wiese, Frank Kraemer, Joseph Al Ahmar, Saarland University, Saarbruecken, Germany abstract The study focused on the onset of plastic flow in various copper raw materials. The tested copper sheets varied by the way they were manufactured (e.g.electroplated, cold rolled), by post treatments (e.g.chromate coating) or by pre ageing. Tests were conducted at different temperatures (25°C, 60°C, 100°C, 150°C).The results show that the onset of plastic flow depends strongly on the manufacturing condition of the copper materials. Microstructural investigations were carried outin order to correlate the particular mechanical behaviour with established knowledge about copper as a fcc material. The paper will point out how certain properties of the copper sheets relate to the onset of plastic flow |
16:50 |
20mn |
Delamination-induced stich crack of copper wires M. van Soestbergen 1, A. Mavinkurve 1, S. Shantaram 2, J.J.M. Zaal 1 1 NXP Semiconductors, Nijmegen, the Netherlands 2 NXP Semiconductors, Austin (TX), United States of America abstract Stitch crack is a recurring failure mechanism for many years. In this work, we will show that delamination is a prerequisite for stitch cracking to occur because it has a pronounced effect on the stress in the stitch. Simulations reveal increased stress for a stitch where the mold compound is delaminated compared to a fully adherent compound. The results of the simulations match the results of TC reliability tests, and scanning acoustic tomography to identify delaminated regions within the package. |
Session 20 |
Solder joint fatigue characterisation and simulation |
10:30 Bellevue |
Wednesday April 05 2017 |
Chaired by R. Dudek, Xuejun Fan |
10:30 |
30mn |
Keynote presentation – Validation of different SAC material models calibrated on isothermal tests using in-situ TMF measurement of thermally induced shear load M. Kuczynska 1, N. Schafet 1, U. Becker 1, R. Metasch 2, M. Roellig 2, A. Kabakchiev 3, S. Weihe 4 1 Robert Bosch GmbH, Automotive Electronics Division, Schwieberdingen, Germany 2 Fraunhofer Institute for Ceramic Technologies and Systems, Branch Materials Diagnostics, Dresden, Germany 3 Robert Bosch GmbH, Corporate Research Division, Renningen, Germany 4 Materials Testing Institute (MPA), University of Stuttgart, Germany abstract In the past, a large number of material models for Sn-based solder alloys have been proposed, which are usually calibrated based on the material testing under isothermal the conditions. However, their ability to map the lifetime differences depending on the temperature rate under the field and test-lab conditions, as well as on the mean operating temperature, is still not completely investigated and validated.The novel thermo-mechanical fatigue (TMF) measurement set-up is employed for in-situ measurement of the material degradation driven by temperature cycles. The experimental system involves different materials, which impose thermally induced displacements onto the solder interconnections. The acceleration of the test duration can be controlled by the placing the sample into the loading positions with the different level of the thermally induced displacement. The measurement enables monitoring of the force-reduction and the concurrent change of displacement. In the current study, the samples comprising a real-scale geometry of the four Ball Grid Array (BGA) connections were stressed with the temperature cycles relevant for the typical lab-tests and field conditions. The level of the thermally induced shear displacement in the solder joints was significantly higher than in an Engine Control Unit ECU. Since the experimental set-up includes various geometrical and material features, an extensive FE-based sensitivity study has been performed. The simulation of the free-expanding system as well as of the system with different pre-characterized dummy samples (without solder joints) revealed the capabilities and specific mechanical behavior of the experimental set-up. Finally, for Sn96.5Ag3.0Cu0.5 solder alloy the ability of the different material formulations to reproduce the trends of the measured hysteresis was analyzed: for double power-law creep model (DPL), unified inelastic strain formulation by Anand, and unified visco-plastic model proposed by Chaboche. Their accuracies in predicting of the acceleration factor between the different temperature profiles are summarized and discussed. |
11:00 |
20mn |
Accelerated Life Time Measurement with In-situ Force and Displacement Monitoring during Thermal Cycling on Solder Joints R.Metasch 1, M. Roellig 1, M. Kuczynska 2, N. Schafet 2, U. Becker 2, K. Meier 3, I. Panchenko 3 1 Fraunhofer Institute for Ceramic Technologies and Systems, Branch Materials Diagnostics, Dresden, Germany 2 Robert Bosch GmbH, Automotive Electronics Division, Stuttgart, Germany 3 Technische Universität Dresden, Electronics Packaging Laboratory, Dresden, Germany abstract This paper is focused on the thermally-induced thermo-mechanical load in interconnections of electronic products, which are necessary to connect an electronic component on substrate material (such as Printed Circuit Board). In order to improve of the material selection, the Fraunhofer IKTS has developed a measurement setup enabling a cost-efficient material characterization in terms of mechanical and fatigue values.The developed setup is able to measure mechanical quantities on a realistic specimen which is stressed under thermo-mechanical load of predefined magnitude. The setup consists of a construction (load frame) with integrated force and contactless displacement sensors to monitor the changes in strength of a specimen in-situ. The first measurement results present a comparison of the influence of two temperature cycling profiles relevant for automotive electronics industry. The realistic specimen comprising four BGA-like solder joints is subjected to temperature cycles until a substantial structure-mechanical failure is observed.Further, the paper presents the calibration steps necessary to understand the unique force-deformation behaviour of the system as function of temperature. |
11:20 |
20mn |
Lead free solder joints characterisation using Single Lap Shear tests Samuel Pin 1, Hélène Frémont 2, Alexandrine Guédon-Gracia 2 1 Institut de Recherche Technologique Saint-Exupéry, Toulouse, France 2 Univ. Bordeaux, IMS Lab, Bordeaux, France abstract The study aims at characterizing the mechanical behaviour of a lead free solder material at a representative scale. The specific alloy considered in this work is SnAg0.3Cu0.5. The test bench comprises a specific single lap shear specimen geometry with grooves associated to a contactless method of monitoring. This allows to test solder joints with a thickness representative of microelectronics applications with good thermal gradients during the reflow process and thus metallurgic properties similar to the ones get with BGA dimensions. The campaign consists in monotonic shear tests and creep tests at 3 different temperatures. An emphasis on the fabrication process and the monitoring data treatment is given in this paper as it can leads to non-consistent results if not managed with precautions due to the reduced scale of the joint. Once the microstructure has been validated, experimental data are used to identify a specific viscoplastic material behaviour. Indeed, the primary creep strain has been found to be significant regarding the common secondary creep rate. Primary creep strain accumulation in cyclic loads could lead to an error of estimation in future predictive calculations if only secondary creep strain was considered as for the former solder materials. It is highlighted by successive creep steps at various stress levels performed in this study. Cyclic hardening has also been investigated but turns to be not relevant to model this particular plastic behaviour. The final set of material coefficients is provided for a full creep behaviour combining primary and secondary creep states coupled with Young’s moduli at -40, 25 and 125°C monotonic loads. |
11:40 |
20mn |
Non-Local Continuum Damage Mechanics Approach in the Finite Element Simulation of Lead-Free Solder Joints Youssef Maniar 1, Benjamin Metais 1, Marta Kuczynska 2, Alexander Kabakchiev 1, Peter Binkele 3, Siegfried Schmauder 3 1 Robert Bosch GmbH, Corporate Research Division, Renningen, Germany 2 Robert Bosch GmbH, Automotive Electronics Division, Schwieberdingen, Germany 3 University Stuttgart, Institute for Materials Testing, Materials Science and Strength of Materials (IMWF), Stuttgart, Germany abstract Realistic material modelling is at the heart of accurate reliability prognosis of electronics hardware by means of Finite Element (FE) calculations. It is usually achieved on the basis of material testing using standardized samples, where well defined, homogeneous stress states and loading conditions can be realized. Both the deformation behaviour in the initial state, as well as the materials degradation during repetitive loading can then be mapped by calibrated damage mechanics FE-models. Such models employ the calculation of internal damage state variables at integration point level, which are functions of stress, strain, time and temperature. However, in the simulation of real components accommodating inhomogeneous stress states due to their complex geometries, damage localization effects can take place. As reported in previous works, local enhancement of damage variables at integration points and significant impact of the FE-mesh quality on the results are commonly obtained in the simulations. Such numerical features can hamper the applicability of damage mechanics models derived from standard material testing samples onto real solder joints geometries.In order to overcome mesh dependency and localization of damage evolution, we adopt a spatial weighted averaging of the damage state variable in a visco-plastic Chaboche material model. This approach overcomes the numerical localization of damage on discrete numerical points and is often referred to as nonlocal damage concept. Here, we highlight the algorithm of nonlocal damage calculation at integration point level, which we implemented for the use in a commercial FE software package. We achieve a spatial damage distribution on a finite microscopic scale with the aim to resemble the physical material degradation, known to happen on the scale of microscopic cracks, grain and grain-boundary modifications. Finally, we discuss the advantages and implications of the nonlocal damage approach on the basis of the damage evolution obtained by simulations of a Low Cycle Fatigue (LCF) specimen. |
12:00 |
20mn |
Fatigue Measurement Setup under Combined Thermal and Vibration Loading on Electronic SMT Assembly Karsten Meier 1, René Metasch 2, Mike Roellig 2, Karlheinz Bock 1 1 Technische Universität Dresden, Electronics Packaging Laboratory 2 Fraunhofer Institute for Ceramic Technologies and Systems – Material Diagnostics abstract In the past vibration experiments have been conducted in the field of automotive, railway transportation, aerospace and other applications where electronics face dynamic mechanical loads. The tests were used to either qualify components or systems, analyse interconnect reliability or to research the fatigue behaviour of the involved interconnect materials. Dominant investigation has been in solder alloys. Field conditions of the named applications usually introduce coupled mechanical and thermal loads to electronics systems. Great efforts have been made to enable these conditions in the experimental environment and extract reasonable reliability results. However, experimental methodologies considering coupled vibration and temperature loads for qualification purpose, accelerated reliability testing or fatigue analysis lack for well-defined load conditions, ability of in-situ load measurement or unwanted behaviour of the experimental setup (e. g. deflection change or rigging due to temperature changes).In recent work we introduce an experimental setup which overcomes these drawbacks. A test vehicle and an associated mount as well as an in-situ measurement approach were developed and tested. This novel setup offers test vehicle clamping options which represent system conditions including rigging for qualification purposes or enable fatigue research which rely on constant rigging-free load conditions. The specimen is a standard PCB with mounted SMT components. Typically vibration will be done as a sine dwell with frequencies in the range of 50 Hz…1 kHz. The setup may be used for isothermal vibration experiments at room, lower or higher temperatures (e.g. -40°C or 125°C) but it enables vibrations experiments at temperature cycling conditions as well. A low weight mount and PCB design targeting temperature cycling with reasonable gradients of about 1 K/min. Combined with integrated PCB heaters even local component heating is enabled. Dependent on the temperature condition either optical or capacitive contact-free in-situ displacement or deflection measurements can be utilised. The latter is very important for small or low weight specimens since acceleration sensors usually add to much mass and strain gages are difficult to mount. Using the contact-free measurement systems high spatial and time measurement resolution can be achieved. |
Session 21 |
3D-Printing & Power Electronics Applications |
10:30 Leipzig |
Wednesday April 05 2017 |
Chaired by Xianping Chen, Sung-Uk Zhang |
10:30 |
30mn |
Keynote presentation – Multi-Physics Models and Condition-based for 3D-Printing of Electronic Packages Chris Bailey, Stoyan Stoyanov, Tim Tilford, Georgios Tourloukis, University of Greenwich |
11:00 |
20mn |
Simulation-driven Development of a Novel SiC Embedded Power Module Design Concept Yafan Zhang 1, Klaus Neumaier 2, Olaf Zschieschang 2, Gerald Weis 3, Gerhard Schmid 3, Mietek Bakowski 1, Hans-Peter Nee 4 1 RISE Acreo AB, Kista, Sweden 2 ON Semiconductor, Aschheim / München, Germany 3 AT 4 KTH School of Electrical Engineering, Stockholm, Sweden abstract Silicon carbide embedded power modules enable a compact and cost competitive packaging solution for high-switching frequency and high-temperature operation applications. Power module packaging technologies span several engineering domains. At the early design stage, simulation-driven development is necessary to shorten the design period and reduce the design cost. This paper presents a novel design concept of a three-phase embedded power module (1200 V, 20 A, 55 mm × 36 mm × 0.808 mm) including silicon carbide metal-oxide-semiconductor field-effect transistor and antiparallel diode dies. Based on the E/CAD design data different layer built-up designs have been tested against thermal, mechanical, and electrical behavior. The obtained simulation data then have been evaluated against a commercial available power module (Motion Smart Power Module SMP33) which utilizes over mold direct bonded copper substrates with soldered semiconductor dies and bond wire contacts. Compared to the conventional module, the loop conductive interconnection parasitic inductance and resistance of the design concept (Vdc to Vdc-) reduces approximately by 88 % and 72 %, respectively. The average junction to case thermal resistance has been improved by approximately more than 10 % even though the total package size reduces by approximately 88 %. Furthermore, the contours of deformation and stresses have been investigated for the design concept in the thermomechanical simulation. |
11:20 |
20mn |
Design of High Voltage 140X100 Footprint IGBT Module Daohui Li, Matthew Packwood, Fang Qi, Steve Jones, Xiaoping Dai, Power Semiconductor RD Center, Dynex Semiconductor Ltd abstract A new type of novel high voltage IGBT module package with a 140mmX100mm footprint half-bridge has been developed. The module aims to provide the market a new type of high power density, high reliability standardised package. The module is scalable and suitable for multi-module parallel connection without degradation, is a low inductance design for both the internal structure and external connection, and has balanced current distribution at internal substrate level and module level, etal. At the design and sample build stage, new types of electromagnetic (EM), electrical circuitry, thermal, mechanical simulation techniques have been used to verify the design to speed up the development cycle. |
11:40 |
20mn |
Topology Optimization of a 3D printed Acoustic Chamber for Photoacoustic Spectrometry Rachid Haouari 1, Veronique Rochus 2, Liesbeth Lagae 1, Xavier Rottenberg 2 1 Imec; Katholiek Universiteit van Leuven, Laboratory of Solid State Physics and Magnetsim, Department of Physics and Astronomy, Leuven, Belgium 2 Imec, Leuven, Belgium |