Self-healing semiconductor wafer processing
Implementations of the present disclosure generally relate to methods for processing substrates, and more particularly, to methods for predicting, quantifying and correcting process drift. In one implementation, the method includes performing a design of experiments (DOE) in a process chamber to obtain sensor readings and film properties at multiple locations on a substrate for every adjustable process control change associated with the process chamber, building a regression model for each location on the substrate using the sensor readings and film properties obtained from the DOE, tracking changes in sensor readings during production, identifying drifting in sensor readings that can lead to a change in film properties using the regression model, and adjusting one or more process controls to correct the drifting in sensor readings to minimize the change in film properties.
1. A method, comprising:
performing a design of experiments in a process chamber, the design of experiments comprising:
positioning a test substrate into the process chamber;
performing a process on the test substrate, wherein the process is controlled by a plurality of process controls;
adjusting a first of the plurality of process controls during the process performed on the test substrate;
recording a first set of process chamber sensor readings following the adjusting of the first of the plurality of process controls;
adjusting a second of the plurality of process controls during the process performed on the test substrate; and
recording the first set of process chamber sensor readings following the adjusting of the second of the plurality of process controls;
building a regression model for each location on the test substrate using the first set of sensor readings obtained from the design of experiments;
positioning a production substrate in the process chamber; and
depositing a film on the production substrate, wherein the depositing is controlled by the plurality of process controls, the depositing comprising:
tracking changes of a second set of process chamber sensor readings during the depositing of the film on the production substrate in the process chamber;
identifying drifting in the second set of process chamber sensor readings that drift, using the regression model, during the depositing of the film on the production substrate from a plurality of set points by 0.5 percent to 3.0 percent;
adjusting one or more of the plurality of process controls during the depositing of the film on the production substrate to correct the identified drift in the second set of process chamber sensor readings; and
continuing the depositing of the film on the production substrate after the adjusting of the one or more of the plurality of process controls.
2. The method of claim 1 , wherein the regression model includes multiple sensor readings relating to one of the film properties.
3. The method of claim 2 , wherein the one of the film properties is a thickness of the film.
4. The method of claim 1 , wherein the one or more process controls are identified from the design of experiments.
5. The method of claim 1 , further comprising validating the regression model.
6. The method of claim 5 , wherein the validating the regression model comprises:
performing metrology on the production substrate to obtain film properties of the film; and
comparing the film properties obtained from the metrology to predicted film properties based on the regression model.
7. A method, comprising:
performing a design of experiments in a process chamber, the design of experiments comprising:
positioning a test substrate into the process chamber;
performing a process on the test substrate, wherein the process is controlled by a plurality of process controls;
adjusting a first of the plurality of process controls during the process performed on the test substrate;
recording a first set of process chamber sensor readings following the adjusting of the first of the plurality of process controls;
adjusting a second of the plurality of process controls during the process performed on the test substrate; and
recording the first set of process chamber sensor readings following the adjusting of the second of the plurality of process controls;
building a regression model for each radial zone on the test substrate using the first set of sensor readings from the design of experiments;
positioning a production substrate in the process chamber; and
depositing a film on the production substrate, wherein the depositing is controlled by the plurality of process controls, the depositing comprising:
tracking changes of a second set of process chamber sensor readings during the depositing of the film on the production substrate in the process chamber; and
identifying drifting in the second set of process chamber sensor readings that drift, using the regression model, during the deposition of the film on the production substrate; wherein
if the drifting in the second set of process chamber sensor readings is less than 0.5 percent, then continuing the depositing of the film on the production substrate without adjusting a process control of the plurality of process controls associated with the drifting;
if the drifting in the second set of process chamber sensor readings is within a range of 0.5 percent to 3.0 percent, then adjusting one or more process controls of the plurality of process controls during the depositing of the film on the production substrate, and then continuing the depositing of the film on the production substrate after the adjusting; and
if the drifting in the second set of process chamber sensor readings is greater than 3.0 percent, then stopping the depositing of the film on the production substrate.
8. The method of claim 7 , wherein the regression model includes multiple sensor readings relating to one of the film properties.
9. The method of claim 8 , wherein the one of the film properties is a thickness of the film.
10. The method of claim 7 , further comprising validating the regression model.
11. The method of claim 10 , wherein the validating the regression model comprises:
performing metrology on the production substrate to obtain film properties; and
comparing the film properties obtained from the metrology to predicted film properties based on the regression model.