Predicting and correcting for image placement errors during the fabrication of EUVL masks

Compensation of overlay errors due to mask bending and non-flatness for EUV masks

“…In the absence of a non-flatness metrology tool with an e-chuck (similar to one on the exposure tool), techniques [8][9][10][11][12] have been proposed wherein physically-based models for the e-chucking are used to estimate the shape of the FS of the mask after e-chucking. In these techniques, non-flatness measurements of the reticle in a "free-standing" configuration (i.e., from a reticle not held by an e-chuck) are used.…”

“…The EWOC procedure presented in this paper is based on using reticle non-flatness measurements in a free standing configuration, similar to that in [10], and is described as follows. First, non-flatness measurements are made on the FS and BS of the mask substrate and also after subsequent thin film deposition (ML, BS film, absorber) steps.…”

“…Since simplified models cannot accurately model all the effects of e-chucking, in [10], a procedure is described by which Finite Element Modeling (FEM) is used to predict the OPD and IPD after e-chucking a non-flat EUV mask using a non-flat e-chuck. The benefit of the FEM approach is that it takes into account the geometries of the chuck and mask, material parameters, friction, and electrostatic forces.…”

“…The benefit of the FEM approach is that it takes into account the geometries of the chuck and mask, material parameters, friction, and electrostatic forces. In [10], the resulting pattern placement error due to OPD is calculated using a fixed 6° incident angle. A table of placement errors was created which is the sum of IPD and contribution from OPD.…”