Overview

The semiconductor industry is on the cusp of a transformative decade, driven by advancements in digitalization, automation, 6G connectivity, artificial intelligence (AI), quantum computing, and the urgent need to achieve Net Zero emissions. ZEISS is a key player in this transformation with our microscopy solutions, which, when combined with AI and ZEISS software, enhances defect detection and measurement in semiconductor manufacturing. This blog post explores how these technologies can be leveraged to drive growth and innovation in the semiconductor sector, particularly through fully automated defect analysis and reporting solutions.

The Need for Near-line Microscopy for Automated Wafer-scale Analysis and Review

In the fast-paced semiconductor manufacturing environment, the need to inspect more wafers efficiently and accurately is critical. Traditional methods that rely on wafer loading with manual inspections are becoming increasingly inadequate. Here’s why fully automated defect analysis and review solutions are essential:

  • Fully automated systems, such as those utilizing the ZEISS AxioImager Vario, can significantly increase throughput by eliminating the time-consuming process of manual wafer inspection. Automation allows for continuous operation, enabling manufacturers to inspect a larger number of wafers in a shorter period. This increased throughput is vital for meeting the growing demand for semiconductors.

  • Automated defect analysis solutions provide consistent and accurate results, reducing the variability associated with human inspections. The integration of AI algorithms within the ZEISS ZEN software architecture combined with the high resolution optical microscope from the AxioImager Vario ensures that defects are detected reliably, regardless of the operator's experience level. This consistency is crucial for maintaining high-quality standards in semiconductor manufacturing.

  • Automated systems can generate real-time reports on wafer quality and defect types, allowing manufacturers to respond quickly to issues as they arise. This immediate feedback loop enables timely adjustments to the manufacturing process, minimizing waste and improving overall efficiency. The ZEISS ZEN software platform can analyse data from the AxioImager Vario and provide actionable insights, facilitating continuous improvement.

  • By implementing fully automated detection and reporting solutions, companies can reduce their reliance on manual labour for inspections. This not only lowers labour costs but also allows skilled workers to focus on more strategic tasks, such as process optimization and innovation. Automation streamlines operations, making them more cost-effective and efficient.

The Role of ZEISS AxioImager Vario in Near-line Microscopy Analysis and Review

The ZEISS AxioImager Vario microscope is a powerful tool for semiconductor manufacturers, providing high-resolution imaging with an array of contrast techniques eg. DIC, BF and FL. Here’s how it integrates with AI and ZEISS software to enhance automated defect analysis:
ZEISS Axio Imager Vario

ZEISS Axio Imager Vario

High-Resolution Imaging

The AxioImager Vario offers exceptional imaging capabilities, allowing for the detailed examination of both patterned and un-patterned wafers. This high-resolution imaging is critical for identifying defects that could impact the performance of semiconductor devices.

Circuit board

Circuit Board​. Reflected light Brightfield, EC Epiplan-NEOFLUAR 5x/0,13

 

AI-Powered Analysis

When paired with ZEISS software, the AxioImager Vario can leverage AI algorithms for automated defect analysis. By training AI models on extensive datasets of wafer images, the software learns to recognize various defect types, such as scratches, particles, and pattern misalignments. This automation not only speeds up the inspection process but also enhances detection accuracy, reducing the likelihood of human error.

Seamless Integration into Automated Workflows

The AxioImager Vario coupled with wafer loading can be seamlessly integrated into automated workflows, allowing for continuous inspection without manual intervention. This integration ensures that manufacturers can maintain high production rates while ensuring quality control.

Monocrystalline Silicon Solar Cell

Monocrystalline Silicon Solar Cell. Reflected light, C-DIC, EC Epiplan-APOCHROMAT 50x/0.95

Data-Driven Insights

The combination of the AxioImager Vario and ZEISS software facilitates the collection and analysis of vast amounts of data. Manufacturers can generate comprehensive reports that highlight defect patterns and trends, enabling them to identify root causes and implement corrective actions. This data-driven approach fosters continuous improvement in wafer production.

Conclusion

The next decade promises to be a transformative period for the semiconductor industry, driven by digitalization, automation, 6G connectivity, AI, quantum computing, and the pursuit of Net Zero. The ZEISS AxioImager Vario microscope, in conjunction with AI and ZEISS software, represents a powerful tool for enhancing defect detection and measurement in semiconductor manufacturing.

By adopting fully automated microscopy for near-line analysis and reporting solutions, UK companies can significantly increase throughput, improve consistency and accuracy, and reduce labour costs. The time to act is now—embracing innovation will ensure that the UK remains a key player in the global semiconductor landscape, ready to meet the demands of the future while driving quality and efficiency in production.

Application in the Semiconductor World

  • Smartphone-power management-ic-interconnects-3d-xrm-cross-section.

    Smartphone-power management-ic-interconnects-3d-xrm-cross-section.

    Power Electronics

    In power electronics, the reliability and efficiency of semiconductor devices are paramount. ZEISS 3D X-ray imaging allows for the non-destructive testing (NDT) and inspection of solder joints, wire bonds, and internal connections in power modules. This capability helps identify defects such as voids or cracks that could lead to failure under high-stress conditions.

    FIB techniques, including Focused Ion Beam Scanning Electron Microscopy (FIBSEM), can then be used to perform detailed analysis on specific areas of interest, providing high-resolution insights into the microstructural integrity of these components. The combination of non-destructive imaging and failure analysis ensures that potential issues are addressed early in the design and manufacturing processes, ultimately enhancing the performance and longevity of power electronic devices.

  • A12-A14 Chip Package

    A12-A14 Chip Package

    Artificial Intelligence (AI)

    As AI applications demand increasingly complex semiconductor architectures, the need for precise characterization of electronic packages becomes critical. ZEISS 3D X-ray imaging can visualize the intricate interconnections and packaging of AI chips, allowing engineers to assess the integrity of multi-layer designs without damaging the samples.

    This is particularly important in AI hardware, where performance can be severely impacted by even minor defects. FIB techniques, including Focused Ion Beam Scanning Electron Microscopy (FIBSEM), can complement this by enabling localized analysis of specific features, such as interconnects and vias, providing additional context for any observed anomalies in the failure analysis process.

  • DRAM package sample
    DRAM package sample

    DRAM package sample – Metal lines and small submicron features are clearly resolved with greater certainty with the 40X-P objective

    DRAM package sample – Metal lines and small submicron features are clearly resolved with greater certainty with the 40X-P objective

    Memory

    In memory devices, such as DRAM and NAND flash, the density of components continues to increase, making defect detection more challenging. ZEISS 3D X-ray imaging facilitates the examination of internal structures, enabling the identification of defects that could affect performance, such as incomplete etching or material delamination.

    The non-destructive testing (NDT) nature of this imaging allows for the assessment of multiple samples in a single run, enhancing throughput. FIB can then be employed to analyze specific memory cells in detail, allowing for targeted investigations into failure mechanisms and improving the overall reliability of memory devices.

  • Hybrid bond workflow
    Hybrid bond workflow

    The workflow detail of the hybrid bond case study. a) move the SEM sample stage to the target hybrid bond. b) SEM Pt deposition. c) the SEM micrograph shows the opening defect between the Cu bonds
    on the target interconnectors. d) X-ray image validated that the target hybrid bond was FIB cross-sectioned.

    The workflow detail of the hybrid bond case study. a) move the SEM sample stage to the target hybrid bond. b) SEM Pt deposition. c) the SEM micrograph shows the opening defect between the Cu bonds
    on the target interconnectors. d) X-ray image validated that the target hybrid bond was FIB cross-sectioned.

    Packaging

    The packaging of semiconductor devices is crucial for performance and reliability. ZEISS 3D X-ray imaging provides insights into the integrity of the packaging, including the quality of encapsulation and the presence of voids or delamination. This is essential for ensuring that devices can withstand environmental stresses.

    FIB techniques can further enhance this analysis by allowing for the preparation of cross-sections of packaged devices, enabling detailed examination of the interfaces between different materials. This combination of non-destructive imaging and defect inspection plays a vital role in ensuring the longevity and functionality of packaged semiconductor devices.

  • Smartphone camera module 3d x-ray microscopy

    Non-destructive 3D X-ray imaging of camera module acquired at 18 µm/voxel resolution shows CMOS imaging sensor and other mechanical components

    Photonics

    For photonics applications, such as integrated photonic circuits and optical devices, the ability to visualize internal structures and interfaces is critical. ZEISS 3D X-ray imaging can help in assessing the quality of photonic waveguides, lenses, and other components, ensuring that they meet the stringent requirements for light transmission and manipulation.

    FIBSEM can be utilized to create cross-sections of photonic devices, providing valuable information about layer thicknesses and material interfaces, which are crucial for optimizing device performance. The integration of non-destructive microscopy techniques and advanced imaging technologies enhances the overall effectiveness of photonics development.