Autonomous substrate processing system

Patent metadata

Field	Value
Publication number	US-11709477-B2
Application number	US-202117143072-A
Country	US
Kind code	B2
Filing date	Jan 6, 2021
Priority date	Jan 6, 2021
Publication date	Jul 25, 2023
Grant date	Jul 25, 2023

Abstract

A substrate processing system comprises one or more transfer chambers; a plurality of process chambers connected to the one or more transfer chambers; and a computing device connected to each of the plurality of process chambers. The computing device is to receive first measurements generated by sensors of a first process chamber during or after a process is performed within the first process chamber; determine that the first process chamber is due for maintenance based on processing the first measurements using a first trained machine learning model; after maintenance has been performed on the first process chamber, receive second measurements generated by the sensors during or after a seasoning process is performed within the first process chamber; and determine that the first process chamber is ready to be brought back into service based on processing the second measurements using a second trained machine learning model.

First claim

What is claimed is:

1. A substrate processing system, comprising:
one or more transfer chambers;
a plurality of process chambers connected to the one or more transfer chambers, the plurality of process chambers comprising a first process chamber comprising a first plurality of sensors and a second process chamber comprising a second plurality of sensors; and
a computing device connected to each of the plurality of process chambers, wherein the computing device is to:
receive one or more first measurements from at least one of the first plurality of sensors of the first process chamber during or after a first instance of a seasoning process performed within the first process chamber after performing maintenance on the first process chamber, wherein the one or more first measurements comprise a first set of measurements from the first plurality of sensors generated during the first instance of the seasoning process;
process the one or more first measurements using a trained machine learning model, wherein the trained machine learning model is to generate a first output based on processing of the one or more first measurements, wherein the first output comprises an indication that the first process chamber is ready to be brought back into service;
cause a first action to be performed with respect to the first process chamber based on the first output of the trained machine learning model;
determine a first result of the first action; and
update a training of the trained machine learning model based on the one or more first measurements, the first output, and the first result of the first action.

2. The substrate processing system of claim 1 , wherein the plurality of process chambers are configured to perform the seasoning process.

3. The substrate processing system of claim 1 , wherein the computing device is further to:
receive one or more second measurements from at least one of the second plurality of sensors of the second process chamber during or after a second instance of the seasoning process performed within the second process chamber;
process the one or more second measurements using the trained machine learning model to generate a second output;
cause a second action to be performed with respect to the second process chamber based on the second output of the trained machine learning model;
determine a second result of the second action; and
update the training of the trained machine learning model based on the one or more second measurements, the second output, and the second result of the second action.

4. The substrate processing system of claim 1 , further comprising:
a factory interface connected to the one or more transfer chambers via one or more load lock;
wherein the computing device is an on-tool computing device that is attached to at least one of a transfer chamber of the one or more transfer chambers, a process chamber of the plurality of process chambers, or the factory interface.

5. The substrate processing system of claim 1 , wherein the first action comprises a test process to be run on a test substrate in the first process chamber, and wherein the first result of the first action comprises one or more measurements of the test substrate generated during or after the test process.

6. The substrate processing system of claim 1 , wherein the first set of measurements comprise optical measurements, power measurements, and pressure measurements.

7. The substrate processing system of claim 1 , wherein the trained machine learning model comprises a neural network.

8. A substrate processing system, comprising:
one or more transfer chambers;
a plurality of process chambers connected to the one or more transfer chambers, the plurality of process chambers comprising a first etch chamber comprising a first plurality of sensors and a second process chamber comprising a second plurality of sensors; and
a computing device connected to each of the plurality of process chambers, wherein the computing device is to:
receive one or more first measurements from at least one of the first plurality of sensors of the first etch chamber during or after a first instance of an etch process performed within the first etch chamber, wherein the one or more first measurements comprise a reflectometry measurement of a film on a substrate generated during the first instance of the etch process;
process the one or more first measurements using a trained machine learning model, wherein the trained machine learning model is to generate a first output based on processing of the one or more first measurements, wherein the first output comprises at least one of an estimated film thickness or an estimated trench depth of the film;
cause a first action to be performed with respect to the first etch chamber based on the first output of the trained machine learning model, wherein the first action comprises stopping the etch process;
determine a first result of the first action, wherein the first result of the first action comprises at least one of a) a difference between a measured thickness of the film and the estimated film thickness of the film or b) a difference between a measured trench depth of the film and the estimated trench depth of the film; and
update a training of the trained machine learning model based on the one or more first measurements, the first output, and the first result of the first action.

9. A process tool, comprising:
a process chamber, wherein the process chamber is an etch chamber;
a plurality of sensors connected to the process chamber; and
a computing device connected to the process chamber and to each of the plurality of sensors, wherein the computing device is to:
receive one or more measurements from at least one of the plurality of sensors during or after a process performed within the process chamber, wherein the process is an etch process, and wherein the one or more measurements comprise a reflectometry measurement of a film on a substrate generated during the process;
process the one or more measurements using a trained machine learning model, wherein the trained machine learning model is to generate an output based on processing of the one or more measurements, wherein the output comprises at least one of an estimated film thickness or an estimated trench depth of the film;
cause an action to be performed with respect to the process chamber based on the output of the trained machine learning model, wherein the action comprises stopping the etch process;
determine a result of the action, wherein the result of the action comprises at least one of a) a difference between a measured thickness of the film and the estimated film thickness of the film orb) a difference between a measured trench depth of the film and the estimated trench depth of the film; and
update a training of the trained machine learning model based on the one or more measurements, the output, and the result of the action.

10. The process tool of claim 9 , wherein the trained machine learning model comprises a neural network.

11. A process tool, comprising:
a process chamber;
a plurality of sensors connected to the process chamber; and
a computing device connected to the process chamber and to each of the plurality of sensors, wherein the computing device is to:
receive one or more measurements from at least one of the plurality of sensors during or after a process performed within the process chamber, wherein the process comprises a seasoning process performed on the process chamber after performing maintenance on the process chamber, and wherein the one or more measurements comprise a set of measurements from the plurality of sensors generated during the process;
proce

Assignees

• Applied Materials Inc

Inventors

• Panda Priyadarshi
• Lian Lei
• Han Pengyu
• Egan Todd J
• Aji Prashant
• Mor Eli
• Tom Alex J
• Tedeschi Leonard Michael

Classifications

• H10P72/0604

  Process monitoring, e.g. flow or thickness monitoring · CPC title

• H10P72/0612

  Production flow monitoring, e.g. for increasing throughput · CPC title

• H10P72/0461

  characterised by the presence of two or more transfer chambers · CPC title

• H10P72/06

  Apparatus for monitoring, sorting, marking, testing or measuring · CPC title

• H10P72/0454

  surrounding a central transfer chamber · CPC title