Granted Patent B2

US 12,480,212 · App. 17/709,304 · Granted Nov 25, 2025

Inventors: Albert Barrett Hicks (Sunnyvale, CA); Serghei Malkov (Hayward, CA)

Assignee: Applied Materials, Inc.

C23C16/52C23C16/45544G06N20/20

Abstract

Claims (43)

1 . A method, comprising:

receiving, by a processing device, first data characterizing a processed surface of a film disposed on a surface of a substrate processed within a recess of a sensor assembly positioned in a first region of a processing chamber, that was processed according to a non-line of sight deposition procedure;

determining, by the processing device based on the first data, a rate of advancement of a first processed surface boundary of the film across the surface of the substrate caused by the non-line of sight deposition procedure, wherein the rate of advancement is correlated with a dosage strength of a reactive species delivered to the first region of the processing chamber during the non-line of sight deposition procedure and a predicted depth of deposition within a feature of a patterned substrate at the region;

determining, by the processing device using the rate of advancement, the dosage strength of the reactive species delivered to the first region of the processing chamber; and

altering, by the processing device, an operation of the processing chamber for future execution of the non-line of sight deposition procedure based on the dosage strength.

2 . The method of claim 1 , further comprising:

determining, using the first data, a thickness map comprising one or more values indicating a thickness of the film at one or more location across the surface of the substrate, wherein the dosage strength is determined further using the thickness map.

3 . The method of claim 2 , further comprising:

determining, using the thickness map, a first portion of the film comprising a first spatial variance below a threshold value;

determining, using the thickness map, a second portion of the film comprising a second spatial variance above the threshold value; and

determining, using the thickness map, a thickness gradient of the film in the second portion, wherein the dosage strength is determined further using the thickness gradient.

4 . The method of claim 1 , further comprising:

determining, by the processing device based on the dosage strength, a first update to one or more process parameters of the non-line of sight deposition procedure, wherein altering the operation is based on the first update.

5 . The method of claim 1 , wherein the first data comprises image data characterizing light reflected from one or more locations across the surface of the film.

6 . The method of claim 5 , wherein the light reflected from the one or more locations comprises at least one of polarization filtered light or color filtered light.

7 . The method of claim 1 , wherein the first data comprises one or more values indicative of a mass of the film at one or more locations across the surface of the substrate.

8 . The method of claim 1 , wherein the first data comprises one or more values characterizing selections of the film corresponding to discrete locations across the surface of the substrate.

9 . The method of claim 1 , wherein the first data is received while the non-line of sight deposition procedure is occurring.

10 . The method of claim 1 , further comprising:

using the first data as input to a machine learning model; and

receiving one or more outputs from the machine learning model, the one or more outputs indicating the dosage strength of the processed surface boundary of the film.

11 . The method of claim 1 , wherein altering the operation of the processing chamber results in a change to one or more of:

a gas flow rate in the processing chamber;

a gas mix provided to the processing chamber;

pressure;

temperature;

process time; or

radio frequency power.

12 . A method, comprising:

receiving, by a processing device, first data characterizing one or more process conditions of a processing chamber, wherein the one or more process conditions are associated with performing a non-line of sight deposition procedure;

determining a process result prediction by processing the first data using one or more machine learning models (MLMs), the process result prediction characterizing a processed surface of a film disposed on a surface of a substrate processed within a recess of a sensor assembly disposed within the processing chamber, the substrate processed according to the non-line of sight deposition procedure under the one or more process conditions; and

altering, by the processing device, an operation of the processing chamber for future execution of the non-line of sight deposition procedure based on the process result prediction.

13 . The method of claim 12 , wherein the process result prediction comprises a thickness profile indicating one or more thickness values of the film at one or more locations across the surface of the substrate.

14 . The method of claim 12 , wherein the process result prediction comprises a rate of advancement of a processed surface boundary of the film across the surface of the substrate.

15 . The method of claim 12 , further comprising:

determining, by the processing device based on the process result prediction, a first update to the non-line of sight deposition procedure, wherein altering the operation is based on the first update.

16 . The method of claim 12 , wherein altering the operation of the processing chamber results in a change to one or more of:

a gas flow rate in the processing chamber;

a gas mix provided to the processing chamber;

pressure;

temperature;

process time; or

radio frequency power.

Assignments (1)

Continuity (1)

References Cited (32)