Method and system for generation of lithography patterns

A lithography pattern generation system comprising:

• a particle source providing an incoming particle beam with a wavelength (λ);
• and
• a mask comprising a plurality of through holes arranged in a path of the incoming particle beam,
• wherein the lithography pattern generation system is configured to generate a pattern on a pattern target, wherein particles in the incoming particle beam arrive at the mask in an incoming particle wave,
• wherein the mask is configured to induce phase shifts across a wavefront of the particle wave as it propagates through the mask,
• wherein the lithography pattern generation system is configured to generate the pattern a target distance (L2) from the mask on the opposite side of the incoming particle beam,
• wherein the through holes have a minimum hole width (w) and the mask has a mask extension (a), which is the maximum distance between two through holes in the mask exposed by the incoming particle beam, and
• wherein L2≤wa/λ.

A method for generating a pattern by lithography comprising:

• exposing a mask comprising through holes to an incoming particle beam from a particle source, wherein particles in the incoming particle beam arrive at the mask in a particle wave, and wherein the mask is configured to induce a phase shift of the particle wave as it propagates through the mask;
• arranging the mask in front of a pattern target;
• and
• generating the pattern on the pattern target,
• wherein the pattern target is arranged a target distance (L2) from the mask, the through holes have a minimum hole width (w) and has a mask extension (a) between the through holes exposed by the incoming particle beam, and wherein L2≤wa/λ.

A computer implemented method for generating a lithography mask model, comprising:

• generating one or more training masks comprising through holes, wherein properties of the training masks are output variables to a machine learning algorithm;
• generating training patterns based on the generated training masks, wherein properties of the training patterns are input parameters of the machine learning algorithm;
• and
• generating the lithography mask model based on a desired pattern by feeding the desired pattern into the machine learning algorithm.