Process sensitivity and optimization with full and simplified resist models

Smith, Mark D.; Mack, Chris A.

doi:10.1117/12.485541

Cited by 6 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It would be important to compare these experimental results to sophisticated stochastic modeling within the reaction and diffusion parameter ranges guided by this work. First principles methods of evaluating and testing resist models are developed by the semiconductor industry ,,,,− but lack the detailed materials constants and concentration profiles provided by neutron reflectivity experiments.…”

Section: Discussionmentioning

confidence: 99%

Neutron Reflectivity Characterization of the Photoacid Reaction-Diffusion Latent and Developed Images of Molecular Resists for Extreme Ultraviolet Lithography

Prabhu¹,

Kang²,

Sha

et al. 2012

Langmuir

View full text Add to dashboard Cite

Lithographic feature size requirements have approached a few radius of gyration of photoresist polymers used in thin-film patterning. Furthermore, the feature dimensions are commensurate with the photoacid diffusion length that defines the underlying latent image. Smaller imaging building blocks may enable reduced feature sizes; however, resolution limits are also dependent upon the spatial extent of the photoacid-catalyzed reaction diffusion front and subsequent dissolution mechanism. The reaction-diffusion front was characterized by neutron reflectivity for ccc stereoisomer-purified, deuterium-labeled tert-butoxycarbonyloxy calix[4]resorcinarene molecular resists. The spatial extent of the reaction front exceeds the size of the molecular resist with an effective diffusion constant of (0.13 ± 0.06) nm(2)/s for reaction times longer than 60 s, with the maximum at shorter times. Comparison to a mean-field reaction-diffusion model shows that a photoacid trapping process provides bounds to the spatial and extent of reaction via a reaction-limited mechanism whereas the ratio of the reaction rate to trapping rate constants recovers the effective diffusion peak. Under the ideal step-exposure conditions, surface roughness was observed after either positive- or negative-tone development. However, negative-tone development follows a surface-restructuring mechanism rather than etch-like dissolution in positive-tone development.

show abstract

Section: Discussionmentioning

confidence: 99%

Neutron Reflectivity Characterization of the Photoacid Reaction-Diffusion Latent and Developed Images of Molecular Resists for Extreme Ultraviolet Lithography

Prabhu¹,

Kang²,

Sha

et al. 2012

Langmuir

View full text Add to dashboard Cite

show abstract

“…In simulations for resists at = 193 nm for instance usually a diffusion length L = 30 nm proves to be appropriate 10 though for some resists and processes L = 60 nm have been reported. 11 Using Eq. ͑23a͒ for spaces:…”

Section: Or Y Fadingmentioning

confidence: 99%

Formulas for lithographic parameters when printing isolated and dense lines

Straaijer

2005

J. Micro/Nanolith. MEMS MOEMS

View full text Add to dashboard Cite

Some practical physics-based formulas have been derived to make quick estimates of imaging performances in lithographic systems. These formulas have a simple structure so that they can be used in mathematical worksheets or, if desired, on a pocket calculator. Formulas have been derived to predict the printing of dense lines, bright field isolated lines, and dark field isolated spaces in positive resist. They are for exposures from binary masks and give exposure latitude, focus depth, and change in critical dimension. These parameters are also derived as a function of fading of the aerial images in lateral and focal directions. Validity has been checked with simulations. It proved that, in cases of the aerial image fading and of focus depth evaluation, only a few adjustment parameters were needed to obtain a reasonably good match with simulations. A table with all basic imaging parameters and their dependence on disturbances such as fading is included.

show abstract

“…A second fundamental question for optimizing the control system is how the changes made actually affect CD uniformity across wafer in a real process. Various authors have developed models for studying the behavior of resist materials during PEB [5][6][7]. Some interesting work has also been published in modeling the transfer of heat within the hardware of the plate [8,9] in addition to experimental work regarding in-situ analysis [10].…”

Section: Optimization Of the Thermal Control System Of The Hotplatementioning

confidence: 99%

ArF processing of 90-nm design rule lithography achieved through enhanced thermal processing

Kagerer¹,

Miller²,

Chang

et al. 2006

SPIE Proceedings

View full text Add to dashboard Cite

As the lithography community has moved to ArF processing on 300 mm wafers for 90 nm design rules the process characterization of the components of variance continues to highlight the thermal requirements for the post exposure bake (PEB) processing step. In particular as the thermal systems have become increasingly uniform, the transient behavior of the thermal processing system has received the focus of attention.This paper demonstrates how a newly designed and patented thermal processing system was optimized for delivering improved thermal uniformity during a typical 90 second PEB processing cycle, rather than being optimized for steady state performance. This was accomplished with the aid of a wireless temperature measurement wafer system for obtaining real time temperature data and by using a response surface model (RSM) experimental design for optimizing parameters of the temperature controller of the thermal processing system.The new units were field retrofitted seamlessly in <2 days at customer sites without disruption to process recipes or flows. After evaluating certain resist parameters such as PEB temperature sensitivity and post exposure delay (PED) -stability of the baseline process, the new units were benchmarked against the previous PEB plates by processing a split lot experiment. Additional hardware characterization included environmental factors such as air velocity in the vicinity of the PEB plates and transient time between PEB and chill plate.At the completion of the optimization process, the within wafer CD uniformity displayed a significant improvement when compared to the previous hardware. The demonstrated within wafer CD uniformity improved by 27% compared to the initial hardware and baseline process. ITRS requirements for the 90 nm node were exceeded.

show abstract

Process sensitivity and optimization with full and simplified resist models

Cited by 6 publications

References 17 publications

Neutron Reflectivity Characterization of the Photoacid Reaction-Diffusion Latent and Developed Images of Molecular Resists for Extreme Ultraviolet Lithography

Neutron Reflectivity Characterization of the Photoacid Reaction-Diffusion Latent and Developed Images of Molecular Resists for Extreme Ultraviolet Lithography

Formulas for lithographic parameters when printing isolated and dense lines

ArF processing of 90-nm design rule lithography achieved through enhanced thermal processing

Contact Info

Product

Resources

About