Granted Patent B2

US 12,314,032 · App. 17/735,303 · Granted May 27, 2025

Inventor: Tzu-Ching Tsai (Taipei, TW)

Assignee: NANYA TECHNOLOGY CORPORATION

G05B19/4099G05B2219/45031H01L21/67H01L22/12

Abstract

Claims (31)

1. A method for controlling a deposition tool, comprising:

executing a first deposition recipe on a current wafer;

generating a first set of data of the current wafer by a first measurement module, wherein the first set of data is one of a data of critical dimension of the current wafer, a data of film thickness of the current wafer, and an electrical data collected by the first measurement module;

analyzing the first set of data by an artificial intelligence module coupled to the first measurement module, wherein the artificial intelligence module uses at least one of a table-based control and a formula-based control to analyze the first set of data;

generating, by the artificial intelligence module, a second deposition recipe and applying the second deposition recipe to the deposition tool when the first set of data is not within a predetermined range, wherein the artificial intelligence module updates the first deposition recipe according to a feedback data to provide the second deposition recipe for a next wafer, wherein the feedback data is one of a feedback CD data, a feedback film thickness data, and a feedback electrical data; and

executing the second deposition recipe on the next wafer, wherein a wafer-to-wafer time frame is updated and adjusted in response to the second deposition recipe.

2. The method for controlling the deposition tool of claim 1 , wherein the artificial intelligence module is integrated in the deposition tool, such that a graphic user interface (GUI) and a database are shared therebetween.

3. The method for controlling the deposition tool of claim 2 , wherein the artificial intelligence module comprises algorithms comprising one or more of the following, alone or in combination: machine learning, hidden Markov models; recurrent neural networks; convolutional neural networks; Bayesian symbolic methods; general adversarial networks; or support vector machines.

4. The method for controlling the deposition tool of claim 3 , further comprising;

transferring the current wafer to a second measurement module before the deposition tool; and

feeding forward at least one parameter of the deposition tool to the artificial intelligence module from the second measurement module before executing the first deposition recipe on the current wafer.

5. A method for controlling a deposition tool, comprising:

executing a first deposition recipe on a current wafer;

generating a first set of data of the current wafer by a first measurement module, wherein the first set of data is one of a data of critical dimension of the current wafer, a data of film thickness of the current wafer, and an electrical data collected by the first measurement module;

analyzing the first set of data by an artificial intelligence module coupled to the first measurement module, wherein the artificial intelligence module uses at least one of a table-based control and a formula-based control to analyze the first set of data;

executing the first deposition recipe on a next wafer when the first set of data is within a predetermined range;

transferring the current wafer to a second measurement module before the deposition tool; and

feeding forward at least one parameter of the deposition tool to the artificial intelligence module from the second measurement module before executing the first deposition recipe on the current wafer;

wherein the artificial intelligence module is integrated in the deposition tool, such that a graphic user interface (GUI) and a database are shared therebetween;

wherein the artificial intelligence module comprises algorithms comprising one or more of the following, alone or in combination: machine learning, hidden Markov models; recurrent neural networks; convolutional neural networks; Bayesian symbolic methods; general adversarial networks; or support vector machines.

6. A method for controlling a deposition tool, comprising:

executing a first deposition recipe on a current wafer;

generating a first set of data of the current wafer by a first measurement module, wherein the first set of data is one of a data of critical dimension of the current wafer, a data of film thickness of the current wafer, and an electrical data collected by the first measurement module;

analyzing the first set of data by an artificial intelligence module coupled to the first measurement module, wherein the artificial intelligence module uses at least one of a table-based control and a formula-based control to analyze the first set of data;

generating, by the artificial intelligence module taking into consideration at least one of the first deposition recipe, an etching recipe, or an implanting recipe, a second deposition recipe and applying the second deposition recipe to the deposition tool when the first set of data is not within a predetermined range, wherein the artificial intelligence module updates the first deposition recipe according to a feedback data to provide the second deposition recipe for a next wafer, wherein the feedback data is one of a feedback CD data, a feedback film thickness data, and a feedback electrical data; and

executing the second deposition recipe on the next wafer, wherein a wafer-to-wafer time frame is updated and adjusted in response to the second deposition recipe.

7. The method for controlling the deposition tool of claim 6 , wherein the artificial intelligence module is integrated in the deposition tool, such that a graphic user interface (GUI) and a database are shared therebetween.

8. The method for controlling the deposition tool of claim 7 , wherein the artificial intelligence module comprises algorithms comprising one or more of the following, alone or in combination: machine learning, hidden Markov models; recurrent neural networks; convolutional neural networks; Bayesian symbolic methods; general adversarial networks; or support vector machines.

9. The method for controlling the deposition tool of claim 8 , further comprising;

transferring the current wafer to a second measurement module before the deposition tool; and

feeding forward at least one parameter of the deposition tool to the artificial intelligence module from the second measurement module before executing the first deposition recipe on the current wafer.

Assignments (1)

Continuity (1)

References Cited (46)