Novel absorber stack for minimizing shadow effect in extreme ultraviolet mask

Abstract

Finding an optimized absorber stack is becoming a more critical issue in the fabrication of extreme ultraviolet ͑EUV͒ mask since it is directly related to the performance of lithography such as pattern fidelity and productivity. Optical simulation, deposition, and measurement have been conducted to establish an optimized absorber stack including antireflection coating ͑ARC͒, absorber layer, and capping ͑or buffer͒ layer, which satisfies major requirements for EUV mask applications. TaN and the other absorber candidates do not show acceptable reflectivity value ͑lower than 5%͒ in deep ultraviolet ͑DUV͒ wavelength region ͑199 or 257 nm͒ for pattern inspection. DUV reflectivity can be lowered by applying C and Al 2 O 3 layers as top ARCs for 199 and 257 nm wavelengths, respectively, while keeping the EUV reflectivity at 13.5 nm less than 1%. ARC-covered TaN absorber stacks result in a reduction of printed CD variation owing to the mitigation of the shadow effect. However, long-term stability and fabricability of these stacks should be examined further.

Stochastic Patterning Simulation Using Attenuated Phase-Shift Mask for Extreme Ultraviolet Lithography

“…This effect results in a patterning limit as well as critical dimension (CD) nonuniformity. [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] The photon shot noise effect is another significant concern in EUVL. This effect arises from statistical fluctuations in the distribution of photons in the photoresist and results in the degradation of various imaging performance parameters such as the CD uniformity (CDU) and line edge roughness (LER).…”

Actinic critical dimension measurement of contaminated extreme ultraviolet mask using coherent scattering microscopy

“…The CSM results exhibited a good correlation with the CD-SEM measurements in the case of the vertical patterns (the difference in the average values was 0.3 nm); however, in the case of the horizontal patterns, the clear mask CD values measured by CSM were lower, owing to the shadowing effect. [26][27][28][29][30][31] The horizontal and vertical patterns were both measured at the best focus. Here, the horizontal-vertical (H-V) CD bias of the mask was 7.1 nm.…”

Optimized Multigrid Strategy for Accurate Flare Modeling with Three-Dimensional Mask Effect in Extreme-Ultraviolet Lithography

“…In several studies, it has been shown that flare changes with the EUV mask-making material used, which is logical because both the reflectivity values change with the material. [24][25][26][27] Initially, the pedestal model given by eq. ( 2) included the mask reflection coefficient factor, but it has been simplified by assuming a pattern density whose form is easily transformed from the layout design.…”