Method of managing semiconductor processing apparatus
Provided is a method of managing a semiconductor processing apparatus, including irradiating, by a light source, a plurality of regions included in a diffuser on a mask stage with extreme ultraviolet (EUV) light, reflecting or transmitting, by the diffuser, the EUV light, transmitting, by an optical system, the EUV light from the diffuser, receiving, by an image sensor, the EUV light from the optical system, obtaining, by the image sensor, a plurality of original images corresponding to the plurality of regions, generating, based on an optical prediction model, a plurality of predictive images estimating a diffraction pattern in the image sensor, adjusting an optical prediction model by comparing the plurality of predictive images with the plurality of original images, and generating, based on the optical prediction model, a plurality of wavefront images corresponding to optical characteristics of each of the plurality of mirrors.
1 . A method of managing a semiconductor processing apparatus, the method comprising:
irradiating, by a light source, a plurality of regions included in a diffuser on a mask stage with extreme ultraviolet (EUV) light;
reflecting or transmitting, by the diffuser, the EUV light;
transmitting, by an optical system, the EUV light from the diffuser;
receiving, by an image sensor, the EUV light from the optical system,
obtaining, by the image sensor, a plurality of original images corresponding to the plurality of regions;
generating, based on an optical prediction model to which optical characteristics of the light source, the diffuser, and the plurality of mirrors are applied, a plurality of predictive images estimating a diffraction pattern in the image sensor;
adjusting an optical prediction model by comparing the plurality of predictive images with the plurality of original images, until consistency between the plurality of predictive images and the plurality of original images is obtained; and
generating, based on the optical prediction model, a plurality of wavefront images corresponding to optical characteristics of each of the plurality of mirrors, based on the consistency between the plurality of predictive images and the plurality of original images being obtained.
2 . The method of claim 1 , wherein the image sensor is on a wafer stage.
3 . The method of claim 2 , wherein the image sensor is on a position of the wafer different from a position of a focal plane of light passing through the optical system and incident on the wafer.
4 . The method of claim 1 , wherein each of the plurality of wavefront images correspond to at least one of amplitude modulation of the EUV light and phase aberration of the EUV light on one of a light output surface between the light source and the diffuser, and a surface of each of the plurality of mirrors.
5 . The method of claim 1 , further comprising executing a backward propagation operation based on a difference between the plurality of predictive images and the plurality of original images to adjust the optical prediction model.
6 . The method of claim 5 , wherein the optical prediction model includes a mirror prediction model reflecting at least one of amplitude modulation and phase aberration between incident light and outgoing light of each of the plurality of mirrors, and
wherein the mirror prediction model for at least one mirror, among the plurality of mirrors, is modified in the backward propagation operation.
7 . The method of claim 1 , wherein the consistency between the plurality of predictive images and the plurality of original images is obtained based on a difference between the plurality of predictive images and the plurality of original images being smaller than a predetermined reference difference.
8 . The method of claim 1 , wherein the plurality of predictive images correspond to the plurality of regions.
9 . The method of claim 1 , wherein the optical prediction model includes a diffusion prediction model corresponding to optical characteristics of each of the plurality of regions, and
wherein the generating the plurality of predictive images comprises arbitrarily generating the diffusion prediction model for each of the plurality of regions.
10 . The method of claim 1 , wherein the acquiring the plurality of original images comprises irradiating the plurality of regions with the EUV light while moving the mask stage.
11 . The method of claim 1 , wherein optical characteristics of the plurality of mirrors are obtained by executing orthogonal decomposition or matrix decomposition on some wavefront images corresponding to the plurality of mirrors, among the plurality of wavefront images, based on the consistency between the plurality of predictive images and the plurality of original images being obtained.
12 . The method of claim 1 , wherein the plurality of mirrors comprise a first mirror and a second mirror, and
wherein the plurality of wavefront images comprise a first mirror image corresponding to phase aberration between incident light and outgoing light of the first mirror, and a second mirror image representing phase aberration between incident light and outgoing light of the second mirror.
13 . The method of claim 12 , wherein aberration and contamination of the first mirror is determined based on the first mirror image, and
wherein aberration and contamination of the second mirror is determined based on the second mirror image.
14 . The method of claim 1 , wherein the plurality of wavefront images comprise a light source image corresponding to at least one of an amplitude and a phase of a light output surface, perpendicular to an optical axis of the EUV light, between the light source and the diffuser.
15 . The method of claim 14 , further comprising diagnosing states of an illumination mirror and a collector, included in an illumination optical system between the light source and the diffuser, based on the light source image.