Author's Latest Posts


ML-Assisted IC Test Binning With Real-Time Prediction At The Edge


IC Test is a critical part of semiconductor manufacturing and proper die binning and material disposition has an important impact on the overall yield and on the process monitoring and failure mode diagnostics. Edge analytics are becoming an increasingly important aspect of die disposition. By intercepting parts in real-time at the wafer test step, we can save downstream processing needs. In th... » read more

In-Product BTI Aging Sensor For Reliability Screening And Early Detection Of Material At Risk


We have developed a new reliability monitoring suite, within a proprietary IP block that we call a CV Core, with aging sensors embedded in the product layout and testable through the product I / O interface. We illustrate the application of the sensor suite with an example of the PMOS NBTI monitor, testable at the wafer level during product electrical wafer sort (EWS), as well after packaging a... » read more

New Method For BEOL Overlay And Process Margin Characterization


This paper presents a new method, design for inspection (DFI) to characterize overlay. Using design-assisted voltage contrast measurement, the method enables in-line test and monitoring of process induced OVL and CD variation of backend-of line (BEOL) features with litho-etch-lithoetch (LELE) patterning. While only some of the features of multi-color patterning scheme are chosen to be aligned d... » read more

Novel E-Beam Techniques For Inspection And Monitoring


In this paper, we report an advanced e-beam defect inspection tool (eProbe®250) and the Design-for-Inspection™ (DFI) system that has been built and deployed by PDF Solutions down to 4nm FinFET technology nodes. This tool has a very high throughput which allows for in-line inspection of nanometer-level defects in the most advanced technology nodes. We also present eProbe applications for... » read more

Advanced High Throughput e-Beam Inspection With DirectScan


Optical inspection cannot resolve critical defects at advanced nodes and cannot detect subsurface defects. Especially at 7nm and below, many yield and reliability killer defects are the result of interactions between lithography, etch, and fill. These defects often will have part per billion (PPB) level fail rates. Conventional eBeam tools lack the throughput to measure PPB level fail rates. A ... » read more

Combining Machine Learning With Advanced Outlier Detection To Improve Quality And Lower Cost


In semiconductor manufacturing, a low defect rate of manufactured integrated circuits is crucial. To minimize outgoing device defectivity, thousands of electrical tests are run, measuring tens of thousands of parameters, with die that are outside of specified parameters considered as fails. However, conventional test techniques often fall short of guaranteeing acceptable quality levels. Given t... » read more

Design And Measurement Requirements For Short Flow Test Arrays To Characterize Emerging Memories


Emerging non-volatile memories are becoming increasingly attractive for embedded and storage-class applications. Among the development challenges of Back-End integrated memory cells are long learning cycle and high wafer cost. We propose a short-flow based characterization of Memory Arrays using a Cross Point Array approach. A detail analysis of design requirements and testability confirms feas... » read more

IEEE S3S 2019 — Characterization Challenges And Solutions For FDSOI Technologies


FDSOI technology has been proposed as an alternative device scaling path which offers benefits of tunable, superior electrostatics transistor while maintaining simplicity of planar integration. New device type and integration elements brought up challenges in device and process characterization and monitoring across the whole lifecycle of the technology. This paper presents successful applicati... » read more

Yield And Reliability Challenges At 7nm And Below


Layout Design Rules have been scaled very aggressively to enable the 7nm technology node without EUV. As a result, achieving acceptable performance and yield in High Volume Manufacturing (HVM) has become an extremely challenging task. Systematic yield and parametric variabilities have become quite significant. Moreover, due to overlay tolerance requirements and diminishing process windows, reli... » read more

Holistic Yield Improvement Methodology


As new products and processes are being introduced into IC manufacturing at an accelerated rate, yield learning and ramping are becoming more challenging due to the increased interaction between the design and process. Compared to random defect caused yield losses, systematic yield loss mechanisms are becoming more important, thus initial yield ramping process becomes more challenging. A “hol... » read more

← Older posts