Ion beam irradiation apparatus and program therefor

Patent metadata

Field	Value
Publication number	US-2022301817-A1
Application number	US-202217835875-A
Country	US
Kind code	A1
Filing date	Jun 8, 2022
Priority date	Mar 25, 2019
Publication date	Sep 22, 2022
Grant date	—

Abstract

An ion beam irradiation apparatus includes modules for generating an ion beam according to a recipe, and a control device. The control device receives the recipe including a processing condition for new processing, reads, from a monitored value storage, a monitored value that indicates a state of a module during a last processing immediately before the new processing, inputs the processing condition and the monitored value to a trained machine learning algorithm and receives, as an output from the trained machine learning algorithm, an initial value for the module, and outputs the initial value to the module to set up the module for generating the ion beam.

First claim

What is claimed is:

1 . An ion beam irradiation apparatus comprising:
a plurality of modules for generating an ion beam according to a recipe; and a control device configured to:
receive the recipe including a processing condition for new processing,
read, from a monitored value storage, a monitored value that indicates a state of a module of the plurality of modules during a last processing immediately before the new processing;
input the processing condition and the monitored value to a trained machine learning algorithm and receive, as an output from the trained machine learning algorithm, an initial value for the module, and
output the initial value to the module to set up the module for generating the ion beam.

2 . The ion beam irradiation apparatus as recited in claim 1 , wherein the module is operated based on an adjusted value obtained by adjusting the initial value.

3 . The ion beam irradiation apparatus as recited in claim 2 , wherein the control device is further configured to store, in a storage, the processing condition, the initial value, and the adjusted initial value in association with each other, as learning data.

4 . The ion beam irradiation apparatus as recited in claim 1 , further comprising a machine learning part configured to update the trained machine learning algorithm based on a plurality of data sets, each data set including, as an explanatory variable, a processing condition during new processing and a monitored value that indicates a state of at least one of the modules during a last processing immediately before the new processing.

5 . The ion beam irradiation apparatus as recited in claim 1 , wherein the control device is configured to receive the trained machine learning algorithm from an off-site device that is located at a location different from a location of the ion beam irradiation apparatus.

6 . The ion beam irradiation apparatus as recited in claim 1 , wherein the control device is further configured to:
based on the processing condition and a setup sequence, select an initial value of a basic operation parameter; input the selected initial value to one of the plurality of modules; and adjust the input initial value to setup the one of the plurality of modules.

7 . The ion beam irradiation apparatus as recited in claim 6 , wherein the control device is further configured to, when the setup of the one of the plurality of modules fails to be completed, acquire an abnormal signal indicative of the failure, and based on the abnormal signal, output the initial value from the trained machine learning algorithm to the one of the plurality of modules.

8 . The ion beam irradiation apparatus as recited in claim 6 , wherein the control device is further configured to, based on the initial value of the basic operation parameter, predict whether or not the setup of the one of the plurality of modules will be completed, and
based on predicting that the setup will not be completed, output the initial value from the trained machine learning algorithm to the one of the plurality of modules.

9 . The ion beam irradiation apparatus as recited in claim 1 , wherein the last processing comprises a plurality of processes corresponding respectively to a plurality of wafers, the monitored value being changed as the plurality of processes are performed on the plurality of wafers, and
wherein the monitored value that is stored is a monitored value after completing a processing of the plurality of processes in a last quartile of the plurality of processes that are performed on the wafers in the last processing.

10 . The ion beam irradiation apparatus as recited in claim 1 , wherein the last processing comprises a plurality of processes corresponding respectively to a plurality of wafers, the monitored value being changed as the plurality of processes are performed on the plurality of wafers, and
wherein the monitored value that is stored is a monitored value after completing a last process of the plurality of processes on a last wafer of the last processing.

11 . A non-transitory computer readable storage medium storing program code which, when executed by at least one central processing unit (CPU) of an ion beam irradiation apparatus that includes a plurality of modules for generating an ion beam according to a recipe, causes the CPU to:
receive the recipe including a processing condition for new processing, read a monitored value that indicates a state of a module of the plurality of modules during a last processing immediately before the new processing; input the processing condition and the monitored value to a trained machine learning algorithm and receive, as an output from the trained machine learning algorithm, an initial value for the module, and output the initial value to the module to set up the module for generating the ion beam.

12 . The non-transitory computer readable storage medium as recited in claim 11 , wherein the at least one CPU is further configured to:
operate the module based on the initial value, obtain an adjusted value based on adjusting the initial value, and store the adjusted value.

13 . The non-transitory computer readable storage medium as recited in claim 12 , wherein the at least one CPU is further configured to store, in a storage, the processing condition, the initial value, and the adjusted initial value in association with each other, as learning data.

14 . The non-transitory computer readable storage medium as recited in claim 11 , wherein the at least one CPU is further configured to receive the trained machine learning algorithm from an off-site device that is located at a location different from a location of the ion beam irradiation apparatus.

15 . The non-transitory computer readable storage medium as recited in claim 11 , wherein the last processing comprises a plurality of processes corresponding respectively to a plurality of wafers, the monitored value being changed as the plurality of processes are performed on the plurality of wafers, and
wherein the monitored value that is stored is a monitored value after completing a processing of the plurality of processes in a last quartile of the plurality of processes that are performed on the wafers in the last processing.

16 . The non-transitory computer readable storage medium as recited in claim 11 , wherein the last processing comprises a plurality of processes corresponding respectively to a plurality of wafers, the monitored value being changed as the plurality of processes are performed on the plurality of wafers, and
wherein the monitored value that is stored is a monitored value after completing a last process of the plurality of processes on a last wafer of the last processing.

17 . A system comprising:
a plurality of ion implanters, each ion implanter comprising a plurality of modules and each ion implanter configured to:
setup at least one module of the plurality of modules according to an initial value obtained from a trained machine learning algorithm based on a processing condition for a new processing and a monitored value that indicates a state of the at least one module during a last processing immediately before the new processing, and
store learning data including the processing condition, the initial value, and an adjusted value obtained by adjusting the initial value during the setup, in association with each other; and
an off-site device communicatively coupled to the plurality of the ion implanters, the off-site device configured to receive the learning data from at least one of the plura

Assignees

• Nissin Ion Equipment Co Ltd

Inventors

• Takemura Shinya

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• H01J37/08

  Ion sources; Ion guns · CPC title

• H01J37/304

  Controlling tubes by information coming from the objects {or from the beam}, e.g. correction signals · CPC title

• H01J37/3171

  for ion implantation · CPC title

• H01J37/3056 Primary

  for microworking, e. g. etching of gratings or trimming of electrical components · CPC title