CN-122003313-A - Dressing condition determining device and dressing condition determining method

CN122003313ACN 122003313 ACN122003313 ACN 122003313ACN-122003313-A

Abstract

The present invention relates to a technique for determining a dressing condition of a polishing pad used for dressing a workpiece such as a wafer, a substrate, or a panel. The trimming condition determining device (60) is provided with a storage device (60 a) for storing an estimation model (62) constructed by machine learning, and a processor (60 b) for executing an operation according to an algorithm of the estimation model (62). The estimation model (62) is configured to output a dressing condition for realizing a target track when a target track image representing a target track of a plurality of abrasive grains constituting the dressing surface (51 a) on the polishing pad (2) is input.

Inventors

Handa naotoshi
HIYAMA HIROKUNI
WADA YUTAKA
GUO JINGMING

Assignees

株式会社荏原制作所

Dates

Publication Date: 20260508
Application Date: 20240926
Priority Date: 20230929

Claims (12)

1. A dressing condition determining apparatus for determining a dressing condition for dressing a rotating polishing pad by pressing a dressing surface of the rotating dresser against the polishing pad, the apparatus comprising: a storage device for storing an estimation model constructed by machine learning, and A processor that performs an operation in accordance with an algorithm of the estimation model, The estimation model is configured to output a dressing condition for realizing a target track when a target track image is input, the target track image representing a target track of a plurality of abrasive grains constituting the dressing surface on the polishing pad.
2. The trimming condition determining apparatus according to claim 1, wherein, The dressing condition includes at least one of a rotational speed of the dresser, a rotational speed of the polishing pad, a swing speed of the dresser on the polishing pad, and a dressing time.
3. The trimming condition determining apparatus according to claim 1, wherein, The target trajectory image is a simulation image of the trajectories of the plurality of abrasive grains generated by performing a simulation of an operation of dressing the polishing pad by pressing the dressing surface of the rotating dresser against the rotating polishing pad.
4. The trimming condition determining apparatus according to claim 1, wherein, The estimation model is a learning completion model constructed by machine learning using a plurality of trimming conditions and training data including a plurality of trajectory images obtained by trimming simulation performed under the plurality of trimming conditions.
5. The trimming condition determining apparatus according to claim 1, wherein, The estimation model has: An encoder that extracts a feature amount of the target trajectory image; A decoder for restoring the target track image from the feature quantity, and A regression analyzer that performs regression analysis on the extracted feature quantity and outputs the trimming condition.
6. The trimming condition determining apparatus according to claim 1, wherein, The target track is a track in which the tracks of the abrasive grains are uniformly distributed on the polishing pad, and the intersections of the tracks of the abrasive grains are uniformly distributed on the polishing pad.
7. A dressing condition determining method for determining a dressing condition for dressing a rotating polishing pad by pressing a dressing surface of the rotating dresser against the polishing pad, characterized in that, Inputting a target track image representing a target track of a plurality of abrasive grains constituting the dressing surface on the polishing pad into an estimated model constructed by machine learning, And outputting finishing conditions for realizing the target track from the estimated model.
8. The trimming condition determining method according to claim 7, wherein, The dressing condition includes at least one of a rotational speed of the dresser, a rotational speed of the polishing pad, a swing speed of the dresser on the polishing pad, and a dressing time.
9. The trimming condition determining method according to claim 7, wherein, The target trajectory image is a simulation image of the trajectories of the plurality of abrasive grains generated by performing a simulation of an operation of dressing the polishing pad by pressing the dressing surface of the rotating dresser against the rotating polishing pad.
10. The trimming condition determining method according to claim 7, wherein, The estimation model is a learning completion model constructed by machine learning using a plurality of trimming conditions and training data including a plurality of trajectory images obtained by trimming simulation performed under the plurality of trimming conditions.
11. The trimming condition determining method according to claim 7, wherein, The estimation model has: An encoder that extracts a feature amount of the target trajectory image; A decoder for restoring the target track image from the feature quantity, and A regression analyzer that performs regression analysis on the extracted feature quantity and outputs the trimming condition.
12. The trimming condition determining method according to claim 7, wherein, The target track is a track in which the tracks of the abrasive grains are uniformly distributed on the polishing pad, and the intersections of the tracks of the abrasive grains are uniformly distributed on the polishing pad.

Description

Dressing condition determining device and dressing condition determining method Technical Field The present invention relates to a technique for determining a dressing condition for dressing a polishing pad used for polishing a workpiece such as a wafer, a substrate, or a panel. Background Chemical mechanical polishing (hereinafter referred to as CMP) is a process of supplying a polishing liquid onto a polishing pad and bringing a workpiece (e.g., a wafer, a substrate, a panel, etc.) into sliding contact with the polishing pad to polish the workpiece. The polishing liquid is usually a slurry containing abrasive grains. A polishing apparatus for performing CMP is provided with a polishing table for supporting a polishing pad having a polishing surface, and a polishing head for pressing a workpiece against the polishing pad. The polishing apparatus polishes the workpiece as follows. The polishing platen and the polishing pad are integrally rotated, and a polishing liquid (typically, slurry) is supplied onto the polishing surface of the polishing pad. The polishing head rotates the workpiece and presses the surface of the workpiece against the polishing surface of the polishing pad. The workpiece is in sliding contact with the polishing pad in the presence of the polishing liquid. The surface of the workpiece is polished by chemical action of the polishing liquid and mechanical action with the polishing particles and/or polishing pad contained in the polishing liquid. The polishing rate (also referred to as removal rate) of a workpiece varies depending on various factors. The state of the polishing surface of the polishing pad is one of factors affecting the polishing rate. That is, the polishing liquid is held on the polishing surface of the polishing pad, and the workpiece is in sliding contact with the polishing surface. The polishing pad has a plurality of minute protrusions on a polishing surface thereof. These many protrusions facilitate mechanical grinding of the workpiece. Therefore, the state of the polishing surface of the polishing pad affects the polishing rate of the workpiece. Therefore, in order to improve and stabilize the polishing rate of the workpiece, dressing of the polishing surface of the polishing pad (hereinafter referred to as pad dressing) is performed before polishing the workpiece. Pad dressing is performed using a dresser having a dressing surface formed of abrasive grains such as diamond particles. More specifically, the polishing surface is roughened (i.e., a plurality of minute protrusions are formed on the polishing surface) by pressing the dressing surface of the dresser against the polishing surface of the polishing pad while rotating the polishing pad and the dresser, respectively. Prior art literature Patent literature Patent document 1 Japanese patent laid-open publication No. 2010-76049 Problems to be solved by the invention Pad conditioning is performed under predetermined conditioning conditions. And should provide uniform roughness (asperity) across the entire polishing surface of the polishing pad. Accordingly, the dressing conditions are determined so that such a polishing surface can be realized. However, in the past, a user may have failed to obtain a desired trimming result by determining the trimming condition by attempting an error. Disclosure of Invention Accordingly, the present invention provides a technique capable of automatically determining dressing conditions capable of achieving a desired state of a polishing surface. (Means for solving the problems) A dressing condition determining device for determining a dressing condition for dressing a polishing pad by pressing a dressing surface of a rotating dresser against the polishing pad includes a storage device for storing an estimated model constructed by machine learning, and a processor for executing an operation in accordance with an algorithm of the estimated model, wherein the estimated model is configured to output, when a target track image representing a target track of a plurality of polishing particles constituting the dressing surface on the polishing pad is input, a dressing condition for realizing the target track. One embodiment is that the dressing condition includes at least one of a rotational speed of the dresser, a rotational speed of the polishing pad, a swing speed of the dresser on the polishing pad, and a dressing time. In one embodiment, the target trajectory image is a simulation image of trajectories of the plurality of abrasive grains generated by performing a simulation of an operation of dressing the polishing pad by pressing the dressing surface of the rotating dresser against the rotating polishing pad. In one embodiment, the estimation model is a learning completion model constructed by machine learning using a plurality of trimming conditions and training data including a plurality of trajectory images obtained by trimming simulation performed under the plurality of trimmi