Method and system for optical sample inspection using coherent diffraction imaging and a-priori knowledge of the sample

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent describes a method and system for inspecting samples, like semiconductor masks, using coherent diffraction imaging (CDI) combined with prior knowledge of the desired sample pattern. It works by illuminating the sample and detecting how the light diffracts, reconstructing a diffraction image of the actual pattern on the sample. This actual diffraction image is then compared to a predicted image calculated from the intended pattern. Any deviations indicate defects, which can be quickly identified for further inspection.

Use CasesContent extracted from patent full text and abstract with AI.

• Rapid inspection and quality control of photolithography masks (including EUV masks) in semiconductor manufacturing.
• Detecting and characterizing defects in patterned substrates for micro/nanoelectronic device production.
• Evaluation of mask cleaning processes by comparing defect densities before and after cleaning.
• Inspection of other patterned optical components or devices where high-resolution, non-destructive defect identification is needed.
• Quality assurance of advanced photomasks used in processors, memory chips, and other microelectronics.

BenefitsContent extracted from patent full text and abstract with AI.

• Enables fast, non-destructive defect detection with high sensitivity and resolution.
• Works for both amplitude and phase defects, which traditional optical methods may miss or be less sensitive to.
• Lens-less method reduces the complexity and cost compared to high-NA optics-based systems.
• Can use a wide range of light sources (visible, UV, EUV, X-rays), making the technique highly adaptable.
• Improved throughput and potentially lower cost-of-ownership for inspection in industrial settings.
• Depth of focus is not critical, simplifying sample handling and setup.
• Can precisely locate and characterize defects for targeted corrective actions, improving yield and product reliability.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

The present invention discloses a method and a system for reflective and scanning CDI for the identification of defects (4) in an actual pattern (6) of a sample (2) as compared to the desired pattern of the sample (2), comprising: a) providing the sample (2), either blank or having the actual pattern (6), said pattern (6) comprising absorbing and/or phase-shifting materials, b) providing a light source (8) for generating a light beam (10) for scanning the sample (2) in transmission mode or reflection mode; c) illuminating the sample (2) with the light beam (10), preferably under an angle of 0 to 45° relative to the surface normal of the sample, thereby diffracting the light beam (10) according to the actual pattern (6) present on the sample (2); d) detecting the diffracted light beam (11) in terms of its position related intensities with a position sensitive detector (12); e) analyzing the detected intensities, and thereby obtaining a reconstructed diffraction image responsive to the actual pattern (6) of the sample (2); f) calculating a predicted diffraction image of the desired pattern, preferably using Fourier or Fresnel calculations; g) comparing the reconstructed diffraction image with the predicted diffraction image for the detection of an intensity variation deviating from the predicted diffraction image; and h) identifying the position of the deviating intensity variation for further inspection this position of the sample (2).