Machine learning and hybrid metrology using scatterometry and LE-XRF to detect voids in copper lines

Kong, Dexin; Motoyama, K.; Pena, Abraham Arceo de la; Huang, Huai; Mendoza, Brock; Breton, Mary; Muthinti, Gangadhara Raja; Shobha, H.; Jiang, Liying; Li, Juntao; Demarest, J.; Gaudiello, John G.; Karve, Gauri; Cepler, Aron; Sendelbach, Matthew; Emans, Susan; Isbester, Paul; Shah, Kavita; Wolfing, Shay; Ger, Avron

doi:10.1117/12.2515257

Cited by 8 publications

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With respect to the modelling-assisted understanding of the measurements of nanomaterials, some aspects of the characterization techniques become crucial when combining them in a hybrid approach: both the penetration and information depths as well as the probing volumes of different techniques may differ, and, in addition, may not match well with some of the sample dimensions or the spatial sample heterogeneity. With the advent of machine learning algorithms that can handle large measurement data sets as quickly as the actual measurement times last, part of the missing knowledge needed for a complete modelling-assisted understanding of the measurement processes in hybrid metrology may be compensated for [69]. However, substantial knowledge on the physical relationship between the techniques' measurands and the particular physical or chemical quantities of interest can further reduce uncertainties and allow for mutual validations of quantification schemes.…”

Section: Hybrid Metrology-determination Of Dimensional and Analytical...mentioning

confidence: 99%

Traceable Characterization of Nanomaterials by X-ray Spectrometry Using Calibrated Instrumentation

Beckhoff

2022

Nanomaterials

View full text Add to dashboard Cite

Traceable characterization methods allow for the accurate correlation of the functionality or toxicity of nanomaterials with their underlaying chemical, structural or physical material properties. These correlations are required for the directed development of nanomaterials to reach target functionalities such as conversion efficiencies or selective sensitivities. The reliable characterization of nanomaterials requires techniques that often need to be adapted to the nano-scaled dimensions of the samples with respect to both the spatial dimensions of the probe and the instrumental or experimental discrimination capability. The traceability of analytical methods revealing information on chemical material properties relies on reference materials or qualified calibration samples, the spatial elemental distributions of which must be very similar to the nanomaterial of interest. At the nanoscale, however, only few well-known reference materials exist. An alternate route to establish the required traceability lays in the physical calibration of the analytical instrument’s response behavior and efficiency in conjunction with a good knowledge of the various interaction probabilities. For the elemental analysis, speciation, and coordination of nanomaterials, such a physical traceability can be achieved with X-ray spectrometry. This requires the radiometric calibration of energy- and wavelength-dispersive X-ray spectrometers, as well as the reliable determination of atomic X-ray fundamental parameters using such instrumentation. In different operational configurations, the information depths, discrimination capability, and sensitivity of X-ray spectrometry can be considerably modified while preserving its traceability, allowing for the characterization of surface contamination as well as interfacial thin layer and nanoparticle chemical compositions. Furthermore, time-resolved and hybrid approaches provide access to analytical information under operando conditions or reveal dimensional information, such as elemental or species depth profiles of nanomaterials. The aim of this review is to demonstrate the absolute quantification capabilities of SI-traceable X-ray spectrometry based upon calibrated instrumentation and knowledge about X-ray interaction probabilities.

show abstract

Section: Hybrid Metrology-determination Of Dimensional and Analytical...mentioning

confidence: 99%

Traceable Characterization of Nanomaterials by X-ray Spectrometry Using Calibrated Instrumentation

Beckhoff

2022

Nanomaterials

View full text Add to dashboard Cite

show abstract

“…For the first use case, we tested the sensitivity of model-less ellipsometry approach to detect intra wafer signature resulting from difference in stress of nitride layers [3] (both on blanket and on patterned wafer). These nitride layers used for the study are dielectric films acting as isolation between metallic Al contact pads of patterned wafer.…”

Section: A Intrawafer Nitride Passivation Layers Properties Variationmentioning

confidence: 99%

“…Machine Learning is a subset of artificial intelligence which enables computer systems to learn from data without being explicitly programmed. This new alternative for data treatment has already been explored for semi-conductor process control in several previous studies such as for example: predictive metrology [1][2] or detection of voids in copper lines [3].…”

Section: Introductionmentioning

confidence: 99%

Spectroscopic Ellipsometry Imaging for Process Deviation Detection via Machine Learning Approach

Alcaire¹,

Cunff²,

Gredy³

et al. 2020

2020 31st Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC)

View full text Add to dashboard Cite

Spectroscopic ellipsometry is a very sensitive metrology technique to accurately measure the thickness and the refractive index of the different layers present on specific dedicated metrology targets. In parallel, optical defectivity techniques are widely implemented in production lines to inspect a large number of dies and catch physical and patterning defects during the process flow. It becomes then of interest to explore a new approach overlapping metrology and defectivity by using the sensitivity of metrology tools on a full wafer scale. In our case, spectroscopic ellipsometry's optical response was collected directly on the dies to capture specific deviations such as film properties and thickness variation. This is an innovative strategy that requires a model-less approach, combining an automatic ellipsometry mapping generation and a smart classification via a machine learning algorithm. In this paper, we will present such approach on two industrial use cases and explain how an image classification algorithm can be implemented to automatically detect the process drift on the latter.

show abstract

“…AFM has a high resolution and full 3D contouring capability 3 , but its ability to measure shallow structures' internal contours and maximum measurable aspect ratio is limited by the tip shape and insufficient stiffness, and its measurement throughput may not meet in-line measurement requirements. X-ray techniques such as small-angle X-ray scattering (SAXS) 4 or X-ray fluorescence (XRF) 5 can measure tiny devices, but require a high-power X-ray source and a specialized work environment, limiting their practicality. WLI offers nondestructive, full-field measurement with decent resolution, but accurately measuring the inner parameters of high-aspectratio structures remains a challenge 6 .…”

Section: Introductionmentioning

confidence: 99%

Artificial-neural-network-assisted DUV scatterometry for OCD on HAR sub-micron structures

Chein

Yang

et al. 2023

Metrology, Inspection, and Process Control XXXVII

View full text Add to dashboard Cite

A neural network-assisted spectral scatterometry method is presented to measure multi-dimensional critical dimensions (CDs) on high aspect ratio (HAR) structures with micron or submicron scales. With the rise of 3D integrated circuit packaging, there is a need for accurate characterization of HAR sub-micron structures. This method uses DUV scatterometry and a broadband light source from DUV to visible light to gather multi-channel reflection data. The inverse modeling method and artificial neural network model enable accurate measurement of multiple CDs of test structures. The results showed accurate measurement of deep trench critical dimensions with a nominal line width of 0.6 μm and aspect ratio up to 5:1, with accuracy within a few nanometers comparable to SEM results using the same sample.

show abstract

Machine learning and hybrid metrology using scatterometry and LE-XRF to detect voids in copper lines

Cited by 8 publications

References 3 publications

Traceable Characterization of Nanomaterials by X-ray Spectrometry Using Calibrated Instrumentation

Traceable Characterization of Nanomaterials by X-ray Spectrometry Using Calibrated Instrumentation

Spectroscopic Ellipsometry Imaging for Process Deviation Detection via Machine Learning Approach

Artificial-neural-network-assisted DUV scatterometry for OCD on HAR sub-micron structures

Contact Info

Product

Resources

About