Spectral Mapping of Photo Emission for IC Fault Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current probing systems face challenges in identifying, localizing, and classifying malfunctioning devices within integrated circuits due to faint light emission and high background noise, especially with shrinking dimensions and operational voltage reductions, making it difficult to determine the exact faulty device and its type.
Innovation Solution
The use of a transmissive grating in the optical path of the probing system, which allows for the imaging of emission spectra without disrupting the view, enabling the identification, localization, and classification of faulty devices by projecting zero and first-order diffractions onto a sensor, with the grating's orientation and configuration optimizing the separation of spectral responses for different fault types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If emission microscopy is used to collect photon emission from integrated circuits, then non-invasive testing capability is improved, but measurement precision deteriorates due to faint light emission and high background noise
Solution Approach 1:
The patent transitions from spatial imaging to spectral domain analysis by capturing emission spectra across multiple wavelengths. This dimensional change from 2D spatial images to 3D spectral data (spatial + wavelength) enables differentiation of signal types that appear identical in conventional emission microscopy, thereby improving measurement precision without compromising non-invasive testing capability
Solution Approach 2:
The patent applies spectral analysis across different wavelengths (analogous to color analysis) to distinguish between various emission sources. By examining the spectral fingerprint at each pixel location, the system can differentiate between transistor switching emissions, leakage currents, and background thermal radiation, significantly enhancing detection precision while maintaining the non-invasive nature of emission microscopy
2Productivity
If integrated circuit dimensions are reduced and operational voltage is lowered, then device density and power consumption are improved, but detection difficulty increases due to fainter light emission
Solution Approach 1:
The patent changes the detection parameter from intensity-only measurement to spectral distribution analysis. By capturing emission intensity across multiple wavelength bands, the system can detect faint emissions from scaled-down devices that would be indistinguishable from background noise in conventional single-wavelength emission microscopy, thereby overcoming the detection difficulty imposed by reduced device dimensions and voltage
Solution Approach 2:
The patent introduces spectral information as an intermediary dimension between the faint emission signal and the detection system. This additional spectral dimension acts as a mediator that enhances the contrast between actual device emissions and background noise, enabling reliable detection of signals from densely packed, low-voltage circuits that would otherwise be undetectable
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the ability to non-invasively identify and classify faulty devices, even in dense areas, by providing clear spectral profiles that distinguish between different fault types, such as overloaded transistors and thermal emissions, improving failure analysis and fault localization.
Implementation Method 1
The transmissive grating is configured to receive collected light from the objective lens and provide diffracted light to the sensor, with the grating's orientation and configuration optimizing the separation of spectral responses for different fault types
Data Source
AI summary
An apparatus and method for optical probing of a DUT is disclosed. The system enables identifying, localizing and classifying faulty devices within the DUT. A selected area of the DUT is imaged while the DUT is receiving test signals, which may be static or dynamic, i.e., causing certain of the active devices to modulate. Light from the DUT is collected and is passed through a transparent diffracting grating prior to imaging it by a sensor and converting it into an electrical signal. The resulting image includes the zero order and first order diffraction of the grating. The grating is configured such that the zero order is in registration with emission sites imaged when the grating is outside the optical path.