Berliner Nanotest und Design

Abstract Complete expression loss of S-methyl-5′-thioadenosine phosphorylase (MTAP) is caused by homozygous 9p21 deletion and results in a critical vulnerability of cancer cells towards drugs targeting multiple different pathways. MTAP deficiency is common in urothelial cancer, but data on its intratumoral heterogeneity - a potential obstacle for targeted therapies - are lacking. To study the heterogeneity of MTAP expression loss and 9p21 deletions in advanced primary urothelial cancers of the urinary bladder, a tissue microarray (TMA) was constructed from 5 different sites from different tissue blocks of 105 pT2-4 urothelial carcinomas and analyzed by immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). In addition, all 1-15 tumor containing blocks (average 6.1) from 41 consecutive pT2-4 carcinomas were analyzed by IHC. Complete absence of MTAP staining (MTAP deficiency) was seen in 132 of 385 interpretable TMA samples. All 80 samples with a complete MTAP expression loss and FISH data had a homozygous 9p21 deletion while there were no cases with homozygous deletions within 148 samples with retained MTAP expression (100% FISH/IHC concordance). On a patient level, there was a homogeneous MTAP deficiency in 33%, a heterogeneous MTAP deficiency in 1% and a retained MTAP expression in 66% of these 98 tumors with at least 3 interpretable TMA samples. Among the 41 consecutive pT2-4 carcinomas from which whole sections were analyzed, MTAP was homogeneously deficient in 34%, heterogeneously deficient in 4.9% and homogeneously retained in 61%. It is concluded that MTAP deficiency is mostly homogeneous in advanced urothelial carcinoma. MTAP IHC is a near perfect surrogate for the detection of homozygous 9p21 (MTAP) deletions. Drugs targeting MTAP-deficiency could be highly useful in a relevant subset of invasive urothelial bladder cancers. Citation Format: Natalia Gorbokon, Viktor Reiswich, Florian Lutz, Sebastian Dwertmann Rico, Henning Plage, Maximilian Lennartz, Margit Fisch, Sarah Minner, Elena Bady, Lisa Hornsteiner, Ronald Simon, Guido Sauter, Henrik Zecha, Stefan Steurer, Thorsten Schlomm, Lukas Bubendorf, David Horst, Thorsten Ecke, Stefan Koch, Martina Kluth, Florian Viehweger, Morton Freytag. MTAP deficiency is highly homogeneous in advanced, muscle-invasive urothelial carcinoma of the urinary bladder [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2654.

A metal 3D-printed high performance multi-jet impingement cooling solution for enabling thermal management of power electronics is presented for integration in electric vehicles. In this paper, we describe the concept, metal additive manufacturing and assembling to demonstrate the proof-of-concept of direct liquid jet impingement cooling of mock-up power modules dissipating heat loss up to 400 W (~230 W/cm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>).

One of the biggest hurdles in developing an innovative, intelligent failure analysis system that caters to the needs of the manufacturing industry, is the lack of manufacturing data to train a deep learning driven system. The Finite Element Method (FEM) has the potential to bridge this data gap by developing simulations that encapsulate the underlying physical phenomena in the form of complex differential equations and solve them numerically. This work presents a detailed study comparing the results obtained for pulse thermography experiments and the FEM. Samples with artificially introduced defects in the form of flat bottom holes are fabricated and used for this study. The concept of defect signatures based on 2D axisymmetric FEM is introduced to significantly reduce the computation time for transient thermal simulations. The method for parameter estimation and boundary conditions for the FEM is explained. The evaluation of the temperature evolution of sound and defect regions, surface temperature distribution and thermal contrast shows, that the FEM results are an excellent estimate of the experimental setup and are a good fit for generating varied data for data augmentation to aid the development of a deep learning based image segmentation model.

This study shows a technique to avoid the necessary spray-coating of devices under test with low- emissivity surfaces in failure analyses by pulsed IR thermography. Such samples can be sintered or soldered electronic components. The mentioned technique is based on a vacuum foil lamination process, which does not contaminate the samples and saves the time of sample preparations. For this purpose, a suitable and optimal foil must be found, which was the goal of this study. This enables inline inspections in industrial production lines by non-destructive IR thermography methods and an additional alternative to conventional techniques such as scanning acoustic microscopy and x-ray analyses.

For the calculation of correct structure function, getting the correct time constant spectrum is an important prerequisite. The time-constant spectrum is the result of the deconvolution of the derivative of the unit step response and weight function. However, deconvolution is an ill-posed problem which needs utmost care. The literature on the problem of structure function proposes two prominent solutions for the deconvolution step, namely Inverse Filtering and Bayesian deconvolution. This research work implements both deconvolution methods in the algorithm for structure function algorithm and compares the obtained structure function with the analytical solution. On the one hand, inverse filtering requires a filter function which does not change the low-frequency components but strongly attenuates high-frequency components in the frequency domain and on the other hand Bayesian deconvolution is an iterative process and might be time-consuming in case of high iteration. The comparison of structure functions for test structures procured from both deconvolution method shows that the structure function derived from Bayesian deconvolution have a better match with that of analytical structure functions.

Knowledge about thermal properties of relevant materials is an important part in development and research in electronics as well as in material sciences. This motivates the development of equipment capable to determine thermal characteristics of a sample at one’s disposal. One of those properties of interest is the thermal diffusivity a. It is commonly measured using the flash method introduced by Parker et al. As those properties are size and structure dependent, it is important to characterise them as processed and in their given thickness as used for the application. Current commercial laser flash methods cannot go below certain thicknesses. Hence, we developed a customised test stand, that allows full control of all key parameters and thus makes it possible to measure the thermal diffusivity of samples in the range of hundred microns thickness. This paper provides detailed information of the development, its features, accuracy, perspectives and limits.

Three different methods of failure analysis on die-attaches are presented and compared in the present work. For this purpose, specially designed samples with thermal test chips on top have been used for a modulated thermal excitation to investigate the underlying die-attach layer by lock-in thermography [1]. The thermoreflectance and the lock-in infrared thermography are thermal and non-destructive methods which still must prove their abilities and reliability in die-attach analysis to find the way into working production lines. The results of both methods have been compared to and verified by the common and widely used scanning acoustic microscopy, which has wide-ranging failure detection abilities but compared to thermographic methods it lacks on production line integration capabilities for 100% in-line inspections.

The mechanical characterization of copper through silicon vias (TSV) is not feasible by means of classical material characterization methods like tensile testing due to the small size of the vias. Instead, instrumented nanoindentation is an appropriate method. It has been utilized for the work reported. In addition, finite element (FE) simulations have been employed to assess the delamination risk in the BEoL stack around the copper TSV. Force/displacement curves were obtained by instrumented nanoindentation using sharp indenter tips with various tip angles (Berkovich, Cube Corner). The modelling of the resulting force/displacement curves is based on an adjusted response surface method [e.g. 3-5] that takes into account the various tip shapes. The respective response surfaces are generated by FE simulation assuming homogenous material with isotropic hardening (Ramberg-Osgood). Considering the residuum surfaces obtained for all the different tip shapes, a single point can be identified as the global minimum. This point specifies the parameters of the Ramberg-Osgood relation, hardening exponent and yield strength. Hence, the presented method enables a fast prediction of the plastic material behaviour in a very good first order approach. The accuracy of the method is evaluated by a virtual reverse check and compared to other methods found in the literature. The results of experimental investigations go into FE simulations of the copper TSV. They assess the interaction with the surrounding BEoL stack taking into account damage mechanical approaches. Since possible delamination between copper TSV and barrier contributes to increasing TSV pumping and protrusion, the authors considered silicon bi-material interface cracking as well. To this end, the cohesive surface contact method implemented in ABAQUS™ [15] is used in combination with a traction-separation law comprising a damage evolution law. Finally, suggestions for via design were provided reducing the risk of damages of BEoL-layers forced by copper vias.

The effective thermal conductivity of composites made up of silver micro-particles embedded in a resin matrix is modelled and measured. This is done for spherical and flake-like particles to analyse the effects of the particles geometry and concentration on the composite thermal performance. It is experimentally found that spherical particles yield a higher thermal conductivity than the one given by flakes, such that it takes the value of 16 Wm−1 K−1 for a 50% volume fraction of particles. Furthermore, this behaviour is well described by a simple and analytical model, which takes into account the particle-particle interactions through a crowding factor. The obtained results could be useful to optimize the design and manufacture of composites with metallic particles.