DETERMINING POTENTIAL DEFECTS IN MASK PATTERNS
The technology involves differentiable mask manufacturing model that helps predict defects on the wafer. According to one aspect, a method includes receiving a mask design. Based on a gradient optimization, one or more parameters of one or more convolution kernels of a mask manufacturing model is determined. Using the one or more parameters of the one or more convolution kernels of the mask manufacturing model, a simulated mask pattern is generated based on the mask design. Whether the simulated mask pattern includes a defect is determined. Whether the mask design needs to be adjusted to avoid defects may also be determined.
1 . A method comprising:
receiving, by one or more processors, a mask design;
determining, by the one or more processors based at least on a gradient optimization, one or more parameters of one or more convolution kernels of a mask manufacturing model;
generating, by the one or more processors using the one or more parameters of the one or more convolution kernels of the mask manufacturing model, a simulated mask pattern based at least on the mask design; and
determining, by the one or more processors, whether the simulated mask pattern includes a defect.
2 . The method of claim 1 , further comprising, adjusting, by the one or more processors, the mask design based at least on a determination that the simulated mask pattern includes the defect.
3 . The method of claim 2 , further comprising:
generating, by the one or more processors using the mask manufacturing model, another simulated mask pattern based at least on the adjusted mask design; and
determining, by the one or more processors, whether the other simulated mask pattern includes the defect.
4 . The method of claim 1 , further comprising, prior to generating the simulated mask pattern, initializing, by the one or more processors, one or more convolution kernels of the mask manufacturing model.
5 . The method of claim 4 , wherein initializing the one or more convolution kernels includes initializing a first kernel of the one or more convolution kernels as a gaussian kernel with a sigma corresponding to a point spread function.
6 . The method of claim 5 , wherein initializing the one or more convolution kernels further includes initializing a second kernel of the one or more convolution kernels as another gaussian kernel with a random, positive sigma.
7 . The method of claim 1 , further comprising:
determining, by the one or more processors, one or more convolutions of the mask design with one or more kernels; and
applying, by the one or more processors, a threshold to the one or more convolutions to generate one or more thresholded convolutions.
8 . The method of claim 7 , wherein generating the simulated mask pattern is further based at least on the thresholded convolutions.
9 . The method of claim 7 , further comprising, prior to determining the one or more convolutions, converting, by the one or more processors, the mask design to a raster format.
10 . The method of claim 1 , further comprising training, by the one or more processors based on the simulated mask pattern, a machine learning model to adjust the mask design.
11 . A system comprising:
memory configured to store at least one of a mask design and a mask manufacturing model; and
one or more processors operatively coupled to the memory, the one or more processors being configured to:
determine, based at least on a gradient optimization, one or more parameters of one or more convolution kernels of the mask manufacturing model;
generate, using the one or more parameters of the one or more convolution kernels of the mask manufacturing model, a simulated mask pattern based at least on the mask design; and
determine whether the simulated mask pattern includes a defect.
12 . The system of claim 11 , wherein the one or more processors are further configured to adjust the mask design based at least on a determination that the simulated mask pattern includes the defect.
13 . The system of claim 12 , wherein the one or more processors are further configured to:
generate, using the mask manufacturing model, another simulated mask pattern based at least on the adjusted mask design; and
determine whether the other simulated mask pattern includes the defect.
14 . The system of claim 11 , wherein the one or more processors are further configured to, prior to generation of the simulated mask pattern, initialize one or more convolution kernels of the mask manufacturing model.
15 . The system of claim 14 , wherein the one or more processors are further configured to initialize the one or more convolution kernels by being configured to initialize a first kernel of the one or more convolution kernels as a gaussian kernel with a sigma corresponding to a point spread function.
16 . The system of claim 15 , wherein the one or more processors are further configured to initialize the one or more convolution kernels by being configured to initialize a second kernel of the one or more convolution kernels initialized as another gaussian kernel with a random, positive sigma.
17 . The system of claim 11 , wherein the one or more processors are further configured to:
determine one or more convolutions of the mask design with one or more kernels; and
apply a threshold to the one or more convolutions to generate one or more thresholded convolutions.
18 . The system of claim 17 , wherein the one or more processors are further configured to generate the simulated mask pattern further based at least on the thresholded convolutions.
19 . The system of claim 17 , wherein the one or more processors are further configured to, prior to determination of the one or more convolutions, convert the mask design to a raster format.
20 . The system of claim 11 , wherein the one or more processors are further configured to train, based at least on the simulated mask pattern, a machine learning model to adjust the mask design.