Fractilia, a next-generation semiconductor industry leader leading the way in stochastic-based analysis and control, provides analysis capabilities for EUV pattern control and yield improvement.

Fractilia announced on the 26th that it is providing the Fractilia Overlay Package.

The Fractilia Overlay Package is an optional product that adds core overlay measurement and analysis capabilities to Fractilia’s MetroLER™ and FAME™ products.

MetroLER™ and FAME™ combine Fractilia’s patented FILM™ (Fractilia Inverse Linescan Model) technology with true computational analysis to provide the only proven fab solution capable of measuring all key stochastic effects responsible for patterning errors at advanced nodes with exceptional accuracy and precision.

Currently, several semiconductor chip manufacturers are working with Fractilia to utilize the new Fractilia overlay package for their Scanning Electron Microscope (SEM)-based overlay data analysis.

halfIn conductor manufacturing, overlay refers to the process of placing each patterned layer precisely over the previous layer to ensure good electrical connection and proper operation of the device.

Traditionally, chipmakers have used optical-based measurement equipment to analyze and control this pattern overlay, a process essential to producing high-performance semiconductor devices with high yields.

This analysis is performed on a specific target at the scribe line rather than on the device itself.

As chip line widths continue to shrink and stochastic variation increases with the introduction of EUV patterning, the gap between what is analyzed at the scribe line and what actually happens on the device is widening.

As a result, there is growing interest in analyzing overlays using SEM equipment, which has higher resolution and precision than optical equipment.

On the other hand, SEM introduces noise into the SEM image, which can easily be mistaken for stochastic variability on the wafer.

“SEM-based overlay analysis is increasingly being used to improve control of advanced patterning processes, but like other stochastic analysis methods, it is subject to system-level random SEM errors,” said Chris Mack, CTO of Fractilia. “Fractilia is uniquely positioned to analyze and remove SEM noise with our proven FILM technology, and as a result, many customers have come to us to leverage our technology to improve the accuracy of their SEM overlay analyses. Moreover, we are confident that combining SEM stochastic analysis with optical overlay analysis through the Fractilia overlay package will not only improve the accuracy of SEM overlay analysis, but also enable better lot dispositioning and compensation, ultimately improving patterning control and reducing NZO (non-zero offset), or stochastic variation,” he said.

Practilia’s FAME portfolio of solutions uses a unique, physics-based SEM modeling and data analysis approach to analyze and remove system-level random errors from SEM images, ensuring accurate measurements of what is actually happening on the wafer, not just what is seen in the image.

FAME simultaneously measures all major stochastic effects such as Line Edge Roughness (LER)/Line Width Roughness (LWR), Local CD Uniformity (LCDU), Local Edge Placement Error (LEPE), and stochastic defects, as well as providing Critical Line Width (CD) analysis. It boasts the industry’s highest signal-to-noise ratio (SNR) edge detection (up to 5x higher SNR than competing solutions) and extracts over 30x more data from each SEM image.

With the new Fractilia Overlay Package, Fractilia adds highly accurate SEM-based overlay analysis (including stochastic feature analysis) to its existing analysis capabilities.

Practilia's products are adopted by dozens of semiconductor industry companies, including leading major semiconductor manufacturers, equipment companies, material suppliers, and research institutes.