CMP Endpoint Detection Using Copper Reference Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemical mechanical polishing (CMP) processes face challenges in determining the polishing endpoint due to variations in material removal rates caused by initial substrate thickness, slurry composition, polishing pad conditions, and load variations, leading to inaccurate optical monitoring results, especially in back-end-of-line processes where silicon substrates do not provide a reliable normalization signal.
Innovation Solution
The method involves using a copper background substrate to improve the measured spectra by normalizing raw spectra with a metal reference spectrum, which reduces noise and enhances the accuracy of endpoint detection by comparing normalized spectra to a library of reference spectra, allowing for precise control of polishing endpoint and rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If silicon substrate is used for normalization in optical monitoring, then the measurement can be performed, but the normalization signal is unreliable and produces significant noise in the spectrum
Solution Approach 1:
The patent introduces a copper underlayer as an intermediary between the silicon substrate and the dielectric layers being polished. This copper underlayer serves as a stable reference surface for optical monitoring, replacing the unreliable silicon substrate as the normalization reference. The copper layer provides consistent optical properties that enable accurate endpoint detection without the noise problems associated with using silicon directly.
2Ease of operation
If polishing time is used to determine endpoint, then the process is simple to control, but variations in material removal rate cause inaccurate endpoint detection
Solution Approach 1:
The patent implements real-time optical monitoring with feedback control by continuously measuring the reflectance spectrum during polishing and comparing it to reference spectra. The system detects changes in the optical signal that indicate the polishing endpoint is approaching, providing feedback that allows for precise endpoint detection independent of material removal rate variations. This feedback mechanism enables the system to adapt to different polishing conditions and achieve accurate endpoint detection.
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 improves the reliability of endpoint detection and reduces within-wafer and wafer-to-wafer thickness non-uniformity by effectively normalizing the spectra, providing a more accurate assessment of the polishing progress and achieving uniformity in substrate thickness.
Implementation Method 1
A sequence of raw spectra of light is reflected from the substrate during polishing with an in-situ optical monitoring system
Implementation Method 2
Each raw spectrum in the sequence of raw spectra is normalized to generate a sequence of normalized spectra. Normalizing includes a division operation in which the raw spectrum is in the numerator and the metal reference spectrum is in the denominator
Data Source
AI summary
A method of controlling polishing includes polishing a substrate of a non-metallic layer undergoing polishing and a metal layer underlying the non-metallic layer; storing a metal reference spectrum, the metal reference spectrum being a spectrum of light reflected from a same metal material as the metal layer; measuring a sequence of raw spectra of light reflected from the substrate during polishing with an in-situ optical monitoring system; normalizing each raw spectrum in the sequence of spectra to generate a sequence of normalized spectra, of which normalizing includes a division operation where the measured spectrum is in the numerator and the metal reference spectrum is in the denominator; and determining at least one of a polishing endpoint or an adjustment for a polishing rate based on at least one normalized predetermined spectrum from the sequence of normalized spectra.