edfas.org ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 27 NO. 1 18 QUANTUM DIAMOND MICROSCOPY FOR SEMICONDUCTOR FAILURE ANALYSIS Marwa Garsi, Andreas Welscher, Manuel Schrimpf, Bartu Bisgin, Michael Hanke, Horst Gieser, Daniela Zahn, and Fleming Bruckmaier QuantumDiamonds, Munich, Germany marwa.garsi@quantumdiamonds.de EDFAAO (2025) 1:18-25 1537-0755/$19.00 ©ASM International® INTRODUCTION As heterogeneous integration (HI) and advanced packaging become increasingly prevalent for achieving the next wave of performance improvements, conventional electrical failure analysis (EFA) techniques are facing growing challenges in meeting industry demands.[1,2] Emerging trends like wafer-to-wafer and chip-to-wafer bonding, through-silicon vias, and backside power delivery are significantly increasing interconnect complexity. Because interconnects are critical to the performance gains in state-of-the-art devices, ensuring their electrical integrity is crucial for ramping up production and maintaining high yields. However, many traditional EFA techniques struggle to cope with weak signals, multiple metallization layers, and stacked dies. Moreover, the increasing adoption of wide band-gap materials such as GaN and SiC results in further complications in current EFA.[3] There is a dire need for new methods that can localize faults that are deep below the surface and surrounded by complex metallization, with three-dimensional information, high resolution, and short measurement times. One of the emerging EFA techniques addressing these problems is quantum sensing with nitrogen-vacancy (NV) centers in diamond.[4,5] The technique enables magnetic field measurements with high spatial resolution and sensitivity, allowing the user to image the electrical activity within the circuit and identify various failure modes, including shorts and opens. This process is known as magnetic current imaging (MCI).[6,7] Because magnetic fields travel through Si, GaN, and SiC unimpeded, this modality is a promising candidate for identifying buried and weak failures within the deep layers, as well as new power devices. Furthermore, compared to other magnetic current imaging techniques such as SQUIDs, this technique achieves higher resolution, does not require scanning, and operates at room conditions, making it appealing from a practical EFA standpoint. A system that performs quantum sensing using diamonds has colloquially come to be known as a quantum diamond micro- scope (QDM). This article presents the QDM as an innovative FA tool, details its operation from an FA engineer’s perspective, and discusses its place in the overall FA workflow. The focus then turns to the lateral resolution metric and analysis of two different integrated circuits: a simple Cu wire test sample, and a commercially available quadrupole NAND-gate circuit (Texas Instruments, CD4011-B). Both samples were selected to familiarize the FA engineer with magnetic field and current density data and are ideal to demonstrate the technique, due to their easy-tounderstand layouts. Here, a lateral magnetic resolution of 3.0 ± 0.5 µm is showcased, which is already competitive for EFA.[2] The article concludes by comparing QDM to existing EFA methods in terms of relevant metrics, such as resolution and measurement time. QDM FOR FAILURE ANALYSIS Currents flowing through conductive paths in integrated circuits (ICs) generate localized magnetic fields, as described by Ampère’s law.[8] Mapping these fields provides crucial insights into the current density distribution, allowing for the identification and localization of circuit anomalies. However, previous MCI methods, such as SQUIDs and GMR sensors, have faced limitations due to impractical system requirements and a steep trade-off between resolution, sensitivity, and acquisition times caused by scanning-mode systems. Quantum sensing using a QDM addresses these challenges by enabling MCI with high spatial resolution,
RkJQdWJsaXNoZXIy MTYyMzk3NQ==