Blank Inspection Research Articles

Carbon deposition mechanism of molten salt cleaning and optimization of multicomponent molten salt formula for remanufacturing.

In remanufacturing engineering, cleaning is the key factor for subsequent blank inspection and parts repair. Molten salt has the characteristics of low viscosity, good fluidity, and strong chemical stability. Salt bath cleaning can be comprehensively applied to remove various organic pollutants. Molten salt has the function of self-cleaning. The dirt in the cleaning pool can be decomposed and reused. Moreover, the waste molten salt is massive, and the waste residue is easy to treat. The formation and adsorption mechanism of carbon deposition was explored, and the formation mechanism of carbon deposition was verified by experiments. Then, the existing formula was improved by mixing experiment and compared with the current cleaning method; the cleaning effect was excellent. A new molten salt formula is proposed on the basis of the research on solar thermal storage medium. The composition percentage and the optimal process parameters of the new molten salt formula were determined on the basis of the experimental design of mixture. The cleaning effect of the new formula is good.

Impact of tool design on defect detection sensitivity in extreme ultraviolet actinic blank inspection

We discuss the impact of various tool design perspectives on defect detection sensitivity for dark-field-based extreme ultraviolet (EUV) actinic blank inspection. We consider the impact of pixel size, EUV source type, and photon collection efficiency on critical defect signal-to-noise ratio (SNR) performance. The results show that as the pixel size approaches the target defect image size, defect SNR increases, and that pixel size also determines the dominant noise source in the inspection system. Moreover, the choice of the EUV source affects the optimal numerical aperture (NA) and illumination settings. For plasma-discharged sources, more photons provided by larger partial coherent illumination can improve the defect SNR, while coherent illumination is needed to get a higher defect SNR for synchrotron-based source. In the end, we show that simply increasing the photon collection efficiency by using high-NA optics or increasing the source power cannot always improve the defect SNR. In a speckle-noise dominated situation, larger outer NA includes more noise than defect signal, thus resulting in a lower SNR. The impact of source power also saturates at a certain level as the system becomes speckle-noise limited compared to photon-noise limited.

Printability of buried extreme ultraviolet lithography photomask defects

Native acting phase-programmed defects, otherwise known as buried program defects, with attributes very similar to native defects, were successfully fabricated using a high-accuracy overlay technique. The defect detectability and visibility were analyzed with conventional amplitude and phase-contrast blank inspection at 193-nm wavelength, pattern inspection at 193-nm wavelength, and scanning electron microscopy. The mask was also printed on wafer, and printability is discussed. Finally, the inspection sensitivity and wafer printability are compared, leading to the observation that the current blank- and pattern-inspection sensitivity is not enough to detect all of the printable defects.

Background level in extreme-UV dark-field image for mask blank inspection

The background level (BGL) in an extreme-UV (EUV) dark-field image for EUV mask blank inspection was analyzed. The background refers to the light scattered by the multilayer roughness of the mask blank, and the variation in the BGL is attributed to random and fixed factors. We found that the BGL variation due to the fixed factor could be successfully reproduced by simulation using surface information obtained by atomic force microscope measurement with the assumption of a multilayer roughness structure. This means that the fixed factor contributing to the BGL variation was found to be speckle noise due to the multilayer roughness. Moreover, this reproduction procedure was found to be valuable for estimating the multilayer roughness structure up to several layers below the surface.

Defect mitigation considerations for EUV photomasks

The introduction of extreme ultraviolet (EUV) lithography into manufacturing requires changes in all aspects of the infrastructure, including the photomask. EUV reflective masks consist of a sophisticated multilayer (ML) mirror, capping layer, absorber layer, and anti-reflective coating thereby dramatically increasing the complexity of the photomask. In addition to absorber type defects similar to those the industry was forced to contend with for deep ultraviolet lithography, the complexity of the mask leads to new classes of ML defects. Furthermore, these approaches are complicated not only by the mask itself but also by unique aspects associated with the exposure of the photomask by the EUV scanner. This paper focuses on the challenges for handling defects associated with inspection, review, and repair for EUV photomasks. Blank inspection and pattern shifting, two completely new steps within the mask manufacturing process that arise from these considerations, and their relationship to mask review and repair are discussed. The impact of shadowing effects on absorber defect repair height is taken into account. The effect of mask biasing and the chief ray angle rotation due to the scanner slit arc shape will be discussed along with the implications of obtaining die-to-die references for inspection and repair. The success criteria for compensational repair of ML defects will be reviewed.

Effectiveness of extreme ultraviolet actinic blank inspection prototype for 16-nm half-pitch for mitigation of multilayered defects

A major challenge for extreme ultraviolet (EUV) lithography is avoiding defects in the fabrication of multilayered (ML) mask blanks. Substrate defects and adders during ML coating are responsible for ML defects, which cause changes on phase and amplitude of EUV light. The ML defects must be identified by inspection prior to absorber patterning in order to reduce the effects of ML defects via covering them with patterns to permit the use of fewer ML defect blanks. Fiducial marks (FMs) on ML blanks can be used for mask alignment and to accurately and precisely determine the locations of ML defects. In this study, we fabricated an FM mask by resist exposure using an e-beam writer and etching. Then, we inspected FMs and ML defects with an EUV actinic full-field mask blank inspection tool developed by EIDEC-LaserTec (LT ABI; EIDEC, Tsukuba, Japan and LaserTec, Yokohama, Japan). Next, we evaluated the ML defect location accuracy on the mask based on FMs of several line depths. Here, we explain the estimation of defect location accuracy using FMs and the LT ABI and discuss the defect numbers which can be covered by absorber patterns. Fewer than 19 defects per blank should be required for EUV blanks to cover the ML defects with patterns.

Influence of shield roughness on Mo/Si defect density for extreme ultraviolet lithography mask blanks

The authors investigated the influence of surface roughness of stainless-steel shields in an ion beam sputtering chamber on the particle defect density of deposited 50 pairs of Mo/Si bilayer films ([Mo/Si]50) used in extreme ultraviolet mask blanks. Shields with varying arithmetic average surface roughness (Ra range approximately 3 to 20 μm) were mounted close to the sputtering targets and the substrate, and along the vacuum chamber interior wall. Silicon-rich particles (Si and Si/Mo) with diameters in the range of several tens of nanometers or more were quantified within a 142 mm× 142 mm area of the prepared blank film using a mask blank inspection tool. Si-rich particle defect density was found to be proportional to the inverse square of the shield surface roughness, suggesting that Si-rich particles arise from the shield surface. The shields with roughness exceeding 8 μm effectively suppressed the accumulation of Si-rich particle defects on the mask blank film.

Phase defect mitigation strategy: unveiling fiducial mark requirements on extreme ultraviolet lithography masks

For extreme ultraviolet lithography (EUVL), the fabrication of defect free multi-layered (ML) mask blanks is a challenge. ML defects are generated by substrate defects and adders during ML coating and are called phase defects (PDs). If we can accept ML blanks with a certain number of PDs, the blank yield will be drastically increased. We can use fewer PD blanks and reduce PD influence by covering them with an absorber layer. To do this, PDs should be located during ML blank defect inspection before absorber patterning. To locate PDs on blanks accurately and precisely, the fiducial marks (FMs) on ML blanks can be used for mask alignment. The defect location accuracy requirement is below 10 nm. We present the results of a feasibility study on the requirements of FMs on EUVL masking by simulations and experiments to establish a PD mitigation method with the EUV actinic blank inspection tool. Based on the results, the optimum ranges for FM lines etched into the ML are 3 to 5 μm in width and at least 100 nm in depth.

Impact of the phase defect structure on an actinic dark-field blank inspection signal and wafer printability

In extreme ultraviolet lithography, reducing the number of phase defects (PDs) on mask blanks is a critical issue because the PDs cause degradation of the printed pattern quality. To minimize the number of printable PDs, covering PDs beneath an absorber pattern is an effective way of addressing the problem. Therefore it becomes necessary to detect the PDs and pinpoint their locations by a blank inspector. In this work, a three-dimensional analysis of PDs has revealed that some PDs can grow and propagate in an angular direction away from the normal to the substrate surface. The impact of the inclination angle on the printing performance was evaluated by calculating printed pattern images using a simulator. A PD with an inclination angle of 1 deg corresponds to a 1-nanometer positional shift compared with a position defined as surface topography of a defect as observed by an atomic force microscope or by some nonactinic defect inspector. However, an actinic blank inspection (ABI) tool with high-magnification optics can pinpoint the actual location of the PDs. Covering a PD under an absorber pattern, by a shifting of the pattern based on the information of the PD’s location obtained by the ABI tool, works quite well.

Design-Aware Defect-Avoidance Floorplanning of EUV Masks

Fabricating defect-free mask blanks remains a major obstacle for the adoption of EUV lithography. We propose a simulated annealing based gridded floorplanner for single-project multiple-die reticles that minimize the design impact of buried defects. Our results show a substantial improvement in mask yield with this approach. For a 60-defect mask, our approach can improve the mask yield from 0% to 26%. If additional design information is available, it can be exploited for more accurate yield computation and further improvement in mask yield to 99.6%. These improvements are achieved with a limited area overhead of less than 0.2% on the exposure field. Our simulation results also indicate that around 10%-30% mask yield improvement is possible as a result of floorplanning compared to shifting the entire mask pattern. Our floorplanner can tolerate a defect position error (due to mask blank inspection tools) of 0.25 μm with just a 2% reduction in yield. The impact of defect dimensions and multilayer EUV patterning on the viability of floorplanning is also analyzed in this paper.

Laser-produced plasma light source for extreme-ultraviolet lithography applications

High-brightness extreme-ultraviolet light sources are required for mask inspections and metrology, including mask blank inspection, actinic pattern inspection, and aerial image measurement system to improve yield and lower cost of ownership. Laser-produced plasma (LPP) light sources have the highest potential to achieve the brightness requirements for all the range of mask inspection tools currently foreseen. High brightness of LPP sources (100 to 1000 W/mm2 sr) is the result of a smaller source size ( ∼ 0.1 mm) than that of competing technologies. Since brightness is inversely proportional to the area of the source, smaller source size corresponds with greater brightness and hence greater inspection throughput. At the Laboratory for Energy Conversion of ETH Zurich, a fully operational continuous-running multi-kHz LPP light source has been developed over the last five years and is now undergoing system optimization. Adlyte, a spin-off of ETH Zurich, is working with industry leaders to commercialize this LPP source. Individual subsystem configuration and the physical boundary conditions and limitations that affect power, brightness, stability, and lifetime management are discussed. This integrated system produces a measured brightness of 259 W/mm2 sr. Outlook for the future growth and integration of the source in high-volume manufacturing tools is then discussed.

Extreme-ultraviolet light source development to enable pre-production mask inspection

As extreme-ultraviolet (EUV) lithography moves into pre- production, the requirement for commercially available mask metrology tools becomes more urgent. A key to developing a successful tool is a reliable, high-brightness EUV light source. The Energetiq EQ-10 is a commercially available EUV light source, with an installed base of over 15 sources in the field. The source relies on an electrodeless Z-pinch™ to produce greater than 10 Watts∕2π of 13.5 nm 2% bandwidth light. In order to meet brightness and stability requirements of mask metrology tools, we have investigated modifications to the original design of the EQ-10. The result of these modifications has roughly doubled the source output power, and has achieved brightness greater than 8 Watts∕mm 2 ∕sr, without sacrificing the spatial and pulse-to-pulse stability of the original design. This level of performance is sufficient for initial mask blank and imaging inspection tools. © 2012 Society of Photo-Optical Instrumentation Engineers

Actinic characterization of extreme ultraviolet bump-type phase defects

Multilayer blank defects in extreme ultraviolet (EUV) lithography photomasks continue to be among the highest concerns impeding commercialization. The SEMATECH Berkeley Actinic Inspection Tool (AIT), an all-EUV, high-magnification microscope, is routinely used to investigate mask defects, including native and programmed defects on mask blanks. So-called phase defects, created by buried substrate bumps and pits, formed within or below the EUV-reflective multilayer coating, cause phase-shifts in the reflected light field. These small disturbances are difficult to detect, yet they can create critical defects in patterns arranged above them. The authors report the through-focus measurement of programmed defects with a range of sizes reaching below the optical resolution of the AIT, and below the detection capabilities of advanced deep ultraviolet mask blank inspection tools. Consistent with previous measurements of native phase defects, we find that the observed phase and intensity changes are much smaller th...

Evaluation of extreme ultraviolet mask defect using blank inspection, patterned mask inspection, and wafer inspection

The key challenge before extreme ultraviolet lithography is to make defect-free masks, for which it is important to identify the root cause of defects, and it is also necessary to establish suitable critical mask defect size for the production of ULSI devices. We have been developing extreme ultraviolet (EUV) mask infrastructures such as a full-field actinic blank inspection tool and 199 nm wavelength patterned mask inspection tool in order to support blank/mask supplier in reducing blank/mask defects which impact wafer printing. In this paper, by evaluating the printability of programmed phase defects and absorber defects exposed by full-field scanner EUV1, we demonstrate that defect detection sensitivities of actinic blank inspection and patterned mask inspection are higher than that of wafer inspection in HP32nm. The evaluations were done by comparing the detection sensitivities of full-field actinic blank inspection tool, 199 nm wavelength patterned mask inspection tool, and electron beam (EB) wafer inspection tool. And then, based on the native defect analysis of blank/mask, we ascertained that actinic blank inspection and patterned mask inspection are effective in detecting killer defects both at the main pattern and at the light-shield border area.

Actinic Mask Blank Inspection and Signal Analysis for Detecting Phase Defects Down to 1.5 nm in Height

The capability of an actinic (at-wavelength) inspection system for extreme ultraviolet lithography (EUVL) mask blank has been analyzed by experiment and simulation. The actinic inspection optics, that we developed to obtain a two-dimensional dark field image, consists of illumination optics, Schwarzschild optics with concave and convex mirrors as dark-field imaging optics, and a back-illuminated charge-coupled-device (BI-CCD). A test mask blank with programmed bump defects of smaller sizes and lower heights compared to those used in a previous work was fabricated and the bump defects were detected by the tool. The inspection experiments demonstrated that fabricated multilayer defects down to 1.5 nm in top height and 60 nm in width can be successfully detected. The simulation further indicated that the inspection optics performed well in detecting phase defects of 1.5 nm in height and 40 nm in width.

Actinic extreme ultraviolet lithography mask blank defect inspection by photoemission electron microscopy

A new method for the actinic inspection of defects inside and on top of extreme ultraviolet (EUV) lithography multilayer-coated mask blanks is presented. The experimental technique is based on photoemission electron microscopy supported by the generation of a standing wave field inside and above the multilayer mask blank when illuminated near the resonance Bragg wavelength at around 13.5nm. Experimental results on programed defect samples based on electron beam lithographic structures or silica balls overcoated with an EUV multilayer show that buried defects with a lateral size down to 50nm are detectable. Furthermore, phase structures as shallow as 6nm in height on a programed phase grating sample have been detected by this technique. The contrast of the phase defect structures has shown to be strongly dependent on and controlled by the phase of the standing wave field at the mask blank surface, and thus can be optimized by tuning the inspection wavelength.

Factors Affecting Magnetic Flux Leakage Inspection of Tailor-Welded Blanks

The development of a laboratory-based tailor-welded blank (TWB) inspection system using the principles of magnetic flux leakage (MFL) is presented. The effects of variations in inspection system operating parameters are quantified to allow for optimized system performance. The parameters examined include the applied magnetic field strength, inspection scanning velocity, spatial resolution of acquired signals, specimen end effects, and pole piece liftoff. The results indicate that this inspection method is relatively insensitive to operating parameters and is therefore robust.

Particle Contamination Control Technology in Electron Beam Mask Writing System for Next-Generation Mask Fabrication

An in-situ measurement system for particles in an electron beam (EB) writer is developed to improve mask yield management. The system has satisfied the required installation specifications for a mask blank inspection system for the EB writer, and the results of an experiment using the system prove that particles added from the mask handling system in our EB writer satisfy the total particle count specification (<1.25 counts/cycle for 6 inch mask). The investigation of particle increase after repeated mask movement has been carried out in each segmented mask handling route. It has been clarified that the segmentation test using this system is helpful for investigation of the origin of particle production on a mask. Effective application of information such as particle position and size obtained by this system will be very useful for improving mask yield management in the mask fabrication process in terms of pattern inspection and repair system.

High sensitivity actinic detection of native defects on extreme ultraviolet lithography mask blanks

We present recent experimental results of actinic extreme ultraviolet lithography mask blank defect inspection. The detection sensitivity of the current actinic inspection system is predicted to be able to reach approximately 30 nm in cross correlation experiments done with commercial visible-light, scanning electron microscopy (SEM), and atomic force microscopy (AFM) inspection tools. Random, native defects identified using the visible-light tool were scanned by the actinic tool and the scattering characteristics were compared. SEM and AFM characterization of selected defects were also performed for physical measurement. We found a defect whose size was as small as 60 nm and AFM analysis showed that this defect was a 3 nm high substrate defect that was smoothed-out by the multilayer coating.

At-wavelength detection of extreme ultraviolet lithography mask blank defects

We report the design and operation of an at-wavelength system for extreme ultraviolet lithography mask blank defect detection. Initial results demonstrate sensitivity to submicron size phase defects. The performance of the system is compared with the practical requirements for a mask blank inspection system in terms of the sensitivity and scanning time.